Radiology Research Forum


Radiology Research Forum

Radiology Research Forum is an ongoing lecture series held approximately every two weeks at the Center for Biomedical Imaging, which reflect the translational research goals of the Department of Radiology at the NYU Langone Medical Center. The forum rotates on a regular basis between research reports highlighting the work of departmental researchers (“research seminars”), reviews of journal articles in topic areas of interest (“journal club”), and lectures by distinguished visiting researchers (“guest lectures”). The lectures are normally held at noon.

Thursday, May 25th at 10:00 am

Peder Larson, PhD

Associate Professor, Principal Investagor
University of California, San Francisco
MRI of Ultra-fast relaxing spins for PET/MRI, Lung imaging, and Myelin Imaging

Abstract: MRI has historically performed poorly when imaging ultra-fast relaxing tissues such as bone, lung tissue, and tendons as well as components of other connective tissues including cartilage and myelin. Specialized pulse sequences such as ultrashort echo time (UTE) and zero echo time (ZTE) MRI offer the potential to image these tissues, and have several promising new applications that will broaden the capability of MRI. These include
1. PET/MRI – Hybrid PET/MRI systems require attenuation correction for accurate PET reconstructions, which should include estimates of bone density. This talk will present work using ZTE MRI for generating pseudo/synthetic CT images that include bone density estimates in the head and pelvis. Most recently, we have applied Deep Learning for this synthetic image generation task.
2. Lung Imaging – Pulmonary MRI has been very challenging due to the short T2* of lung parenchyma and motion, but is important for assessing pulmonary nodules in PET/MRI and for dose reduction in pediatric populations. This talk will present an approach using UTE MRI, where self-navigation is achieved through a local low-rank reconstruction of dynamic 3D image navigators and motion-corrected images are reconstructed similarly to XD-GRASP.
3. Myelin Imaging – Myelin facilitates crucial long-range communication across the brain, and is typically assessed in MRI through diffusion-weighting, magnetization transfer, and myelin water imaging. It has been shown through ex vivo studies that there are fast relaxing components in myelin associated with protons in the myelin phospholipid membranes, which are not captured in these conventional approaches. This talk will present in vivo characterizations of the ultrashort-T2* components in the brain that have the potential to provide a more direct measurement of myelination.

About the speaker: Peder Larson, PhD, is an Associate Professor in Residence and a Principal Investigator in the Department of Radiology and Biomedical Imaging at the University of California, San Francisco. Dr. Larson's research program is primarily centered around developing new MRI scanning and reconstruction technology for improved clinical outcomes.

Tuesday, May 16th at noon

Hong-Hsi Lee

PhD Candidate
Sackler Institute of Biomedical Sciences
NYU School of Medicine
Time-Dependent Diffusion in the Brain

Abstract: Diffusion MRI is sensitive to the length scale of tens of microns, which coincides to the scale of microstructure in the human brain tissue. By changing the diffusion time or diffusion gradient pulse width, we can probe the brain micro-geometry via time-dependent diffusion measurements. To increase the sensitivity to the microstructure, STEAM sequence is often used for extending the range of diffusion time. However, water exchange between myelin water and intra-/extra-axonal water may bias the parameter estimations. This talk will focus on the time dependence either along or perpendicular to white matter axons and corresponding micro-geometries, and the correction for the time dependence measured by STEAM.

About the speaker: After finishing the major in Medicine and Physics in Taiwan, Hong-Hsi Lee is working with Dmitry S. Novikov and Els Fieremans as a PhD candidate in the biomedical imaging program at the Sackler Institute of Biomedical Sciences. Hong-Hsi’s recent focus is on the Monte Carlo simulation of the diffusion and water exchange phenomenon between highly aligned, randomly packed, myelinated axons.

Tuesday, May 9th at noon

Gregory Lemberskiy

PhD Candidate
Sackler Institute of Biomedical Sciences
NYU School of Medicine
Time-Dependent Diffusion in the Body

Abstract: Diffusion of water molecules is directly influenced by the mountainous landscape of biological tissue. By modeling time-dependent diffusion, it is possible to reverse engineer various features of this landscape. The proposed model will depend on the underlying tissue microstructure, which poses an additional challenge of model selection. This talk will focus on the efforts of modeling diffusion time-dependence in the prostate, which embodies modeling problems, which concern partial volume and model selection, as well as imaging problems, such as geometric distortion and low SNR.

About the speaker: Greg attended NYU as an undergrad, where he studied Physics. After briefly working in astronomical imaging, he decided to try his hand at medical imaging under the guidance of Dmitry S. Novikov and Els Fieremans. Greg is currently a PhD candidate in the biomedical imaging program at the Sackler Institute of Biomedical Sciences. Most of his work focuses on the development of time-dependent diffusion applications “outside-the-brain.”

Tuesday, April 18th at noon

Giuseppe Carlucci, PhD

Assistant Professor of Radiology
NYU School of Medicine
Novel agents for PET targeted imaging and theranostics

Abstract: PET radiochemistry can be a great resource for imaging, treatment and point-of-care response/monitoring in Cancer and Cardiovascular disease. A novel small molecule targeting the cyclin-dependent kinases CDK4/6 and a series of radiolabeled nanobodies and peptides for atherosclerotic plaques imaging will be presented. The seminar will also focus on the fundamentals of Radiochemistry and how the newly established CAi2R Radiochemistry facility will operate.

About the speaker: Dr. Giuseppe Carlucci is assistant professor of Radiology at NYU School of Medicine. Dr. Carlucci completed his postdoctoral training in radiochemistry at the Memorial Sloan Kettering Cancer Center, and his PhD degree at Rijksuniversiteit - University of Groningen (The Netherlands). Dr. Carlucci research program revolves around the development and validation of novel imaging tracers. This involves small molecule–, peptide-, and nanoparticle-based probes for both optical and PET imaging. His work is focused on the design of companion imaging agents for determining drug susceptibility and target engagement in the preclinical and translational setting.

Friday, April 14th at noon

Nikola Stikov, PhD

Assistant Professor of Biomedical Engineering
Co-director, NeuroPoly, École Polytechnique, University of Montreal
Producer of MRM Highlights and founder of OHBM blog
To microstructural imaging and beyond: separating the signal from the noise

Abstract: Over the past decade, the number of microstructural imaging papers has been doubling every 2.7 years. With such growth, it is becoming increasingly difficult to perform a fair comparison between competing approaches. Some simplify the tissue modelling and overlook physiological constraints. Others overparametrize the models and amplify the noise. The outcome is a field of research with great promise, but few checks and balances.
This lecture will introduce several frameworks for interpreting, validating and communicating microstructural imaging data. Examples will be drawn from myelin imaging in the brain, focusing on the challenges associated with mapping the longitudinal relaxation time (T1), the axon caliber, and the myelin thickness (g-ratio). The last part of the lecture will put these frameworks in a broader science communication context, discussing how medical imaging researchers can set new standards for reviewing, publishing, and publicizing their findings.

About the speaker: Dr. Nikola Stikov is assistant professor of Biomedical Engineering, a researcher at the Montreal Heart Institute, and co-director of NeuroPoly, the Neuroimaging Research Laboratory at École Polytechnique, University of Montreal. His research runs the gamut of quantitative magnetic resonance imaging, from basic issues of standardization and accuracy, to biophysical modelling, microstructural imaging, and clinical applications. Prior to joining the Polytechnique faculty, Dr. Stikov completed his postdoctoral training at the Montreal Neurological Institute, and his B.S., M.S., and PhD degree at Stanford University. In 2014 Dr. Stikov was elected Junior Fellow of the International Society for Magnetic Resonance in Medicine (ISMRM), and in 2015 he joined the Editorial Board of the journal Magnetic Resonance in Medicine (MRM), spearheading the journal's Highlights initiative. Continuing with his science outreach activities, in 2016 Dr. Stikov established the official blog of the Organization for Human Brain Mapping (OHBM).

Monday, April 10 at noon

Guanshu Liu, PhD

Assistant Professor of Radiology
Johns Hopkins University
MRI-guided drug delivery without chemical labeling

Abstract: Recently, Chemical Exchange Saturation Transfer (CEST) has emerged as an attractive MRI contrast mechanism. In CEST, the MRI contrast is generated by transferring the magnetic labeled water-exchanging protons (OH, NH, or NH2) from a CEST agent to its surrounding water molecules. Many natural biological compounds naturally carry exchangeable protons, making them possibly detected by CEST MRI directly in a “label-free” manner. In our studies, we utilized this unique feature to directly detect drugs and drugs carriers, which makes MRI-guided drug delivery possible even without any chemical labeling, a strategy we called “natural labeling”. This new MRI labeling strategy in principle can be tailored to many existing drug delivery systems, and portends a new path to safe, rapid clinical translation of image-guided drug delivery.

Tuesday, April 4th at noon

Harikrishna Rallapalli, BS

Graduate Student
NYU School of Medicine
A High Throughput, MEMRI-Based Imaging Pipeline to Study Mouse Models of Sporadic Human Cancer

Abstract: A high-throughput imaging pipeline is presented to characterize the heterogeneity in longitudinal disease progression in mouse models of human brain cancer and to test the efficacy of novel anti-cancer therapeutics in accurate mouse models of sporadic human cancer.

About the speaker: Hari Rallapalli is a graduate student in the Turnbull lab. He received his BS in Biomedical Engineering from the University of California, Davis.

Tuesday, March 28th at noon

Ines Blockx, PhD

Assistant Research Scientist
Center for Biomedical Imaging
Department of Radiology
NYU Langone Medical Center
In vivo characterization of rat models of Huntington’s disease using Diffusion Imaging

Abstract: Huntington disease (HD) is a dominantly inherited and progressive neurodegenerative disorder, caused by a CAG trinucleotide repeat expansion (≥ 39 repeats) within the HD gene. The median age at which HD occurs, is around 40 years and the disease progresses over time and is invariably fatal 15–20 years after the onset of the first symptoms. The major goals of current HD research are to improve early detection and monitor pathological changes in individuals both at early and advanced stages of the disease. Animal models of inherited neurological diseases provide an opportunity to test potential biomarkers of disease onset and progression and evaluate treatments for translation to clinical care. Using several diffusion MR techniques we studied two different rat models of HD. In this talk I will present data that shows that diffusion MRI is a sensitive and quantitative method to detect HD related neurodegenerative changes, at both microstructural and subcellular levels.

About the speaker: Ines Blockx holds a master in biomedical sciences and graduated at the University of Antwerp, Belgium. After her master degree in 2007, she was rewarded with an IWT (agency for innovation by science and technology) PhD research grant and had the opportunity to start a PhD at the Bio-Imaging lab (University of Antwerp - Belgium) under supervision of Prof. Dr. Annemie Van der Linden for. An important focus of her PhD was the identification of symptom-independent biomarkers of Huntington’s disease neuropathology and progression using in vivo MRI in different rodent models. After receiving her PhD in 2011, she started working as a postdoctoral researcher at the Bio-Imaging lab. During her postdoctoral career, she was the coordinating scientist and key player in an industrial research grant (Janssen Alzheimer Immunotherapy project) where she focussed on the Exploration of multiple MRI modalities to study cerebral vascular dynamics and blood-brain-barrier integrity (BBB) in a mouse model of Alzheimer disease. Currently she is working as an assistant research scientist at the Centre for Biomedical imaging at NYU, where she has the opportunity to expand her research field from preclinical to clinical neuroimaging. Here she is involved in the research of Dr. Mariana Lazar, focussing on MR spectroscopy in Schizophrenia.

Tuesday, March 7th at noon

Richard Spencer, MD, PhD

Chief, Magnetic Resonance Imaging and Spectroscopy Section
NIH/National Institute on Aging
Magnetic Resonance Relaxometry and Macromolecular Mapping: An Inverse Problem Framework

Abstract: There is an ongoing need for non-invasive identification of macromolecular changes in tissue. An important application is to the diagnosis of early osteoarthritis (OA). Our work in this area combines basic science studies in magnetic resonance imaging and relaxometry with emerging methodologies that carry translational potential. We will discuss multi-exponential transverse relaxation analysis as a means to identify underlying macromolecular compartments in normal and degraded cartilage, as well as important extensions of this work, based on higher dimensional relaxometry and compressed sensing. We will describe the mathematical setting for this work as a linear inverse problem. Further work in human subjects requires introduction of a nonlinear model system. We will describe several approaches to these problems and indicate the potential for improved detection of early cartilage degradation. Our methods are also applicable to directly mapping myelin in the brain, and we have obtained results showing myelination pattern alterations with age and in cognitive impairment. All of these studies are centered around the clinical goal of improving the ability of magnetic resonance methods to diagnose pathology and to monitor disease progression.

Tuesday, February 14th at noon

Eric E. Sigmund, PhD

Associate Professor of Radiology
New York University School of Medicine
Simultaneous PET/MRI in Advanced Breast Cancer : Initial Experiences and Future Potential

Abstract: Separately, PET and MRI have longstanding roles in diagnosis, prognosis and monitoring of breast cancer. Since the recent advent of the simultaneous PET/MRI platform, intense research has taken place to identify unrealized applications of their fusion. Initial work around the world has included study of a range of practical advantages (feasibility, efficiency, patient retention, physiologic simultaneity, co-registration), but always with an eye toward future 'breakout' applications beyond those with separate scans. I will describe efforts within our breast cancer research group that pursue both practical and fundamental benefits with the unique capabilities in our research center. Whole body evaluation of metastatic breast cancer patients is nearly equivalently done with PET/MRI as with PET/CT but with half the radiation dose. Dynamic contrast enhanced (DCE) MRI and intra-voxel incoherent motion (IVIM) MRI offer a range of quantitative characterizations of the primary tumor microenvironment (cellularity, vascular volume, vascular permeability).
that when combined with fluorodeoxyglucose (FDG) uptake provide a comprehensive characterization of malignancy in one imaging session. Simultaneity also supports detailed intralesional correlations that may increase classification ability even further. Finally, future planned work with more specific microenvironment tracers and integrated PET and MRI pharmacokinetic modeling holds remarkable potential for oncologic management with noninvasive imaging.

Monday, February 6th at noon

Markus H¨llebrand

Fraunhofer MEVIS
Bremen, Germany
Multi Cycle analysis of cardiac function in real-time

Abstract: Analyzing moving organs such as the heart in MRI is a challenging task. In clinical routine images are acquired over several heartbeats to reconstruct all contraction phases of one representative cardiac cycle using ECG-gating and breath-hold techniques.
Real-time MRI techniques allow the acquisition of serial 2D images with a temporal resolution of up to 20 ms under free breathing. The analysis of real-time MRI sequences, however, requires adapted segmentation techniques as well as an advanced analysis providing information about temporal evolution of parameters during individual heart cycles in amulti cycle analysis workflow.

Wednesday, February 1st at noon

Giulio Ferrazzi, PhD

Research Associate
Biomedical Engineering Department
King’s College London, UK
Imaging the fetus and the neonate using MR
Tuesday, January 24th at noon

Dr. Timothy Shepherd

Assistant Professor, Director of Brain Mapping
Department of Radiology
New York University School of Medicine
Integrated PET-MRI for current clinical neuroradiology dilemmas—a CAI2R perspective
January 23rd at noon

Jeffrey Fessler, PhD

William L. Root Professor of EECS
University of Michigan
Optimal first-order convex minimization methods with applications to image reconstruction and machine learning

Abstract: Many problems in signal and image processing, machine learning, and estimation require optimization of convex cost functions. For convex cost functions with Lipschitz continuous gradients, Nesterov's fast gradient method decreases the cost function at least as fast as the square of the number of iterations, a rate order that is optimal. This talk describes a new first-order optimization method called the optimized gradient method (OGM) that converges twice as fast as Nesterov's famous method yet has a remarkably similar simple implementation. Interestingly, Drori recently showed that OGM has optimal complexity among first-order methods. I will discuss other recent extensions and show examples in machine learning and X-ray computed tomography (CT). Combining OGM with ordered subsets provides particularly fast reconstruction for CT. This work is joint with Donghwan Kim.

About the speaker: Jeff Fessler is the William L. Root Professor of EECS at the University of Michigan. He received the BSEE degree from Purdue University in 1985, the MSEE degree from Stanford University in 1986, and the M.S. degree in Statistics from Stanford University in 1989. From 1985 to 1988 he was a National Science Foundation Graduate Fellow at Stanford, where he earned a Ph.D. in electrical engineering in 1990. He has worked at the University of Michigan since then. From 1991 to 1992 he was a Department of Energy Alexander Hollaender Post-Doctoral Fellow in the Division of Nuclear Medicine. From 1993 to 1995 he was an Assistant Professor in Nuclear Medicine and the Bioengineering Program. He is now a Professor in the Departments of Electrical Engineering and Computer Science, Radiology, and Biomedical Engineering. He became a Fellow of the IEEE in 2006, for contributions to the theory and practice of image reconstruction. He received the Francois Erbsmann award for his IPMI93 presentation, and the Edward Hoffman Medical Imaging Scientist Award in 2013. He has served as an associate editor for IEEE Transactions on Medical Imaging, the IEEE Signal Processing Letters, and the IEEE Transactions on Image Processing, and is currently serving as an associate editor for the IEEE Transactions on Computational Imaging. He has chaired the IEEE T-MI Steering Committee and the ISBI Steering Committee. He was co-chair of the 1997 SPIE conference on Image Reconstruction and Restoration, technical program co-chair of the 2002 IEEE International Symposium on Biomedical Imaging (ISBI), and general chair of ISBI 2007. His research interests are in statistical aspects of imaging problems, and he has supervised doctoral research in PET, SPECT, X-ray CT, MRI, and optical imaging problems.

Tuesday, January 10th at noon

Gadi Wollstein, MD

Professor of Ophthalmology
Director, Ophthalmic Imaging Research Laboratory
Vice Chair for Clinical Research
NYU Langone Medical Center

Joel Schuman, MD

Professor and Chairman of Ophthalmology
Professor of Neuroscience and Physiology
NYU Langone Medical Center
Professor of Electrical and Computer Engineering
NYU Tandon School of Engineering

Hiroshi Ishikawa, MD

Professor of Ophthalmology
Director, Ocular Imaging Center
NYU Langone Medical Center
Part II: In-vivo High Resolution Ocular Imaging - Innovative Technologies and Clinical Challenges

Summary: In recent years ocular imaging has become the cornerstone for clinical diagnosis and disease monitoring as well as a primary research tool in ophthalmology. In this presentation we will discuss state-of-the-art, in-vivo, high resolution ocular imaging technologies. We will present the utility and challenges of the technologies to advance clinical practice and research of glaucoma - a leading cause of blindness and visual morbidity.

Friday, January 6th at noon

Alexander P. Lin, PhD

Director, Center for Clinical Spectroscopy
Department of Radiology, Brigham and Women’s Hospital
Assistant Professor, Harvard Medical School
The Virtual Biopsy: Magnetic Resonance Spectroscopy of Traumatic Brain Injury

Abstract: Advances in neuroimaging provide us with greater insight to brain injury than ever before. Magnetic resonance spectroscopy is a non-invasive method of measuring brain chemistry altered by bran injury using readily available MRI, thus providing a virtual biopsy of concussions. A review of the technology and current findings from the acute to chronic stages of mild brain injury, including the rising concern of chronic traumatic encephalopathy in sports and military-related repetitive brain trauma, will be discussed.

About the speaker: Alexander P. Lin, PhD is an assistant professor at Harvard Medical School and director of the Center for Clinical Spectroscopy at Brigham and Women’s Hospital. Dr. Lin earned his doctorate in Molecular Biochemistry and Biophysics from the California Institute of Technology. He has been involved in magnetic resonance spectroscopy since 1997 both from a research and clinical perspective. His primary research focus is on sports and military-related brain injury which has been featured by the media and funded by that National Institutes of Health and Department of Defense.

Tuesday, December 13th at noon

Gadi Wollstein, MD

Professor of Ophthalmology
Director, Ophthalmic Imaging Research Laboratory
Vice Chair for Clinical Research
NYU Langone Medical Center

Joel Schuman, MD

Professor and Chairman of Ophthalmology
Professor of Neuroscience and Physiology
NYU Langone Medical Center
Professor of Electrical and Computer Engineering
NYU Tandon School of Engineering

Hiroshi Ishikawa, MD

Professor of Ophthalmology
Director, Ocular Imaging Center
NYU Langone Medical Center
Part I: In-vivo High Resolution Ocular Imaging - Innovative Technologies and Clinical Challenges

Summary: In recent years ocular imaging has become the cornerstone for clinical diagnosis and disease monitoring as well as a primary research tool in ophthalmology. In this presentation we will discuss state-of-the-art, in-vivo, high resolution ocular imaging technologies. We will present the utility and challenges of the technologies to advance clinical practice and research of glaucoma - a leading cause of blindness and visual morbidity.

Monday, November 21st at noon

Martin Burger, PhD

Institute for Computational and Applied Mathematics
University of Münster
Variational Methods for Motion-Corrected Image Reconstruction

Abstract: In this talk we discuss the use of advanced physical modeling to build successful image reconstruction approaches for dynamic imaging including motion, noise, and undersampling. The variational approach is based on minimizing energy functionals in a spatio-temporal domain including advanced models of the image formation process, noise, and motion. For the latter hyperelastic or fluid-dynamic constraints are used in order estimate feasible motion vectors jointly with the image sequence. We present the potential use of the methods for dynamic PET and highly undersampled dynamic CT. Finally we comment on extensions to include cross-modality information such as available in PET-MR. We discuss potential issues when using pre-estimated motion vectors from the MR sequence and propose a mathematical model to overcome those.

About the speaker: Martin Burger received a MS (Diplom) in Mathematics (1998) and a Ph.D. in Mathematics (2000) from the Johannes Kepler University Linz. After a period as CAM assistant professor at UCLA he returned to Johannes Kepler University, where he did his habilitation in Mathematics (2005). Briefly afterwards, he was offered a full professor position in applied mathematics at the Westfälische Wilhelms-Universität Münster, where he moved in 2006. Since then he heads the mathematical imaging group and has contributed strongly to building up interdisciplinary research structures related to biomedical imaging. In particular, he serves as PI, executive board member, and research area coordinator for the Cluster of Excellence "Cells in Motion" funded by the German Science Foundation (DFG). He received several awards and recognitions, including the Calderon Prize 2009 by the Inverse Problems International Association, an ERC consolidator grant 2013, and an offer to become director of the Weierstrass Institute for Applied Analysis and Stochastic in Berlin. His research is centered around mathematical imaging and inverse problems, currently with a strong focus on dynamic and high-dimensional image reconstruction.

Friday, November 18th at noon

Thomas Benkert, PhD, (NYU Langone Medical Center) and Bjorn Stemkens (Utrecht University Medical Center)

Advanced methods for motion-robust free-breathing MRI

Abstract: MRI of motion-sensitive applications such as abdominal examinations usually relies on strict breath-holding. However, breath-holding can fail especially for sick, elderly, or pediatric patients, which can render image quality non-diagnostic. Furthermore, sudden motion events such as global body shifts, bulk motion, or coughing may induce further artifacts.
This talk presents methods which solve these problems and enable motion-robust free-breathing MR acquisitions. First, recent advances for comprehensive one-stop shop free-breathing imaging are presented by Thomas Benkert. Second, a technique to automatically detect and exclude bulk motion is described by Bjorn Stemkens. In summary, the presented techniques have the potential to increase patient comfort, improve clinical workflow, and eliminate the risk for failed exams caused by imperfect breath-holding or sudden patient movements.

About the speakers: Dr. Thomas Benkert obtained his PhD in Physics at the University of Wuerzburg, where he worked on novel steady-state techniques for fast MRI under the supervision of Dr. Felix Breuer. In August 2015, he joined CBI as a postdoctoral researcher in the team of Dr. Tobias Block, where he is developing methods for comprehensive free-breathing imaging.
Bjorn Stemkens is a PhD candidate at the department of Radiotherapy at the University Medical Center in Utrecht where he is working on the implementation of novel MRI applications for MRI-guided radiotherapy, with a focus on abdominal treatments. In July he started a 4-month internship at CBI in the team of Dr. Tobias Block to develop a technique to detect bulk motion for robust free-breathing abdominal imaging.

Tuesday, November 15th at noon

Bruce Berkowitz, PhD

Professor, Department of Ophthalmology
Director, Small-Animal MRI Facility
Wayne State University School of Medicine
Oxidative Stress and its Functional Consequences Measured In Vivo by MRI

Abstract: In 1992, it was not obvious that MRI, a relatively insensitive and still developing imaging method, would be useful for examining the retina, one of the thinnest organs in the body. Since then, Dr. Berkowitz has established a body of work that highlights MRI as a surprisingly useful discovery tool in vision research. These methods have been successful transitioned into cancer and brain research areas, and are used by drug companies and other investigators world-wide. Improvements in resolution and methodology have even allowed us to measure the physiology of sub-compartments within rod cells in vivo. These data are spatially grounded based on optical coherence tomography images and compared to visual performance using optokinetic tracking. His current pioneering efforts uses MRI to measure neuronal oxidative stress without a contrast agent in untreatable neurodegenerative disease, including Alzheimer’s disease, to optimize antioxidant treatment in vivo.

Tuesday, November 8th at noon

Haris Sair, MD

Assistant professor of Radiology
Johns Hopkins University School of Medicine
Presurgical brain mapping using resting state fMRI – promises and challenges

Abstract: Assessment of intrinsic brain networks using resting state functional MRI (rs-fMRI) has resulted in a paradigm shift in evaluating brain function. Changes in functional connectivity have been described in numerous disorders, and normal intrinsic brain networks characterized in thousands of subjects. Several studies have examined the use of rs-fMRI in presurgical brain mapping. Following an overview of rs-fMRI basics, the benefits of rs-fMRI over task-fMRI in presurgical brain mapping will be discussed. Challenges in characterizing rs-fMRI at the single subject level for presurgical brain mapping will be reviewed.

About the speaker: Haris Sair MD is Assistant Professor of Radiology in the Department of Radiology at Johns Hopkins University School of Medicine. He completed a 2 year fellowship in Neuroradiology at the Massachusetts General Hospital, where he developed an interest in clinical functional MRI. His primary research interest is in application of resting state fMRI at the single subject level for clinical use, concentrating on presurgical brain mapping, but also including development of rs-fMRI based prediction models in disease and outcome.

Tuesday, October 25th at noon

Christoph Juchem, PhD

Departments of Biomedical Engineering and Radiology
Columbia University in the City of New York
Magnetic Resonance Engineering – from Bench to Bedside

Abstract: My laboratory pursues technology and method developments in the fields of magnetic resonance imaging (MRI) and spectroscopy (MRS) to advance their clinical potential for the study of multiple sclerosis (MS) and other neurological disorders. MRI and MRS allow the non-invasive measurement of brain anatomy and physiology, but excellent B0 magnetic field homogeneity is required for meaningful results. In the first part of my talk, I will present a technique for magnetic field modeling and correction, i.e. shimming, that is based on the combination of fields generated by an ensemble of individual, generic coils. This multi-coil approach enables the accurate generation of simple and complex magnetic field shapes in a flexible fashion. B0 shimming with the dynamic multi-coil technique (DYNAMITE) is shown to outperform conventional methods based on spherical harmonic (SH) functions and provides unrivaled magnetic field homogeneity in mouse, rat and human brain.
MS is a chronic disorder of the central nervous system that leads to demyelination and neurodegeneration. Its underlying pathobiochemical mechanisms, however, remain poorly understood. MRS promises non-invasive access to the brain's biochemistry in vivo, but suffers from methodological limitations and experimental imperfections. The goal of our work is to establish MRS as a clinical research tool towards in vivo metabolomics of the pathogenesis of MS through a combination of ultra-high 7 Tesla field, state-of-the-art B0/B1 shimming and optimized MRS methods. The second part of my talk will focus on the specific MRS infrastructure and implementations that enables us to assess pathological changes from the earliest stage of the disease.

Tuesday, October 18th at noon

Jullie Pan, MD, PhD

Professor of Neurology
University of Pittsburgh School of Medicine
7T and 3T Imaging in Epilepsy

Abstract: This talk will focus on Dr. Pan's work in the development and application of high field imaging approaches to better understand the metabolic and functional pathophysiology of epilepsy. These methods include high degree B0 shimming, high resolution MRSI and in vivo detection of amino acids and will discuss some of her results from 7T and 3T.

Tuesday, October 4th at noon

Ivan Kirov, PhD

Assistant Professor of Radiology
Center for Biomedical Imagign
Department of
Proton MR spectroscopy of lesion evolution in Multiple Sclerosis: steady-state metabolism and its relationship to conventional imaging

Abstract: Although MRI assessment of white matter lesions is essential for the clinical management of multiple sclerosis, the processes leading to the formation of lesions and underlying their subsequent MRI appearance are incompletely understood. We used proton MR spectroscopy to study the evolution of N-acetyl-aspartate (NAA), creatine (Cr), choline (Cho) and myo-inositol (mI) in pre-lesional tissue, persistent and transient new lesions, as well as in chronic lesions, and related the results to quantitative MRI measures of T1-hypointensity and T2-volume. Within 10 patients with relapsing-remitting course, there were 180 regions-of-interest consisting of up to seven semi-annual follow-ups of normal-appearing white matter (NAWM, n=10), pre-lesional tissue giving rise to acute lesions which resolved (n=3) or persisted (n=3), and of moderately (n=9) and severely hypointense (n=6) chronic lesions. Compared to NAWM, pre-lesional tissue had higher Cr and Cho, while compared to lesions, pre-lesional tissue had higher NAA. Resolving acute lesions showed similar NAA levels pre- and post-formation, suggesting no long-term axonal damage. In chronic lesions, there was an increase in mI, suggesting accumulating astrogliosis. Lesion volume was a better predictor of axonal health than T1-hypointensity, with lesions larger than 1.5 cm3 uniformly exhibiting very low (<4.5 millimolar) NAA concentrations. A positive correlation between longitudinal changes in Cho and in lesion volume in moderately hypointense lesions implied that lesion size is mediated by chronic inflammation. These and other results are integrated in a discussion on the steady state metabolism of lesion evolution in Multiple Sclerosis, viewed in the context of conventional MRI measures.

About the speaker: Ivan Kirov received his Bachelor of Science in Biology from University of California, Irvine. After graduation, he worked for 2 years as a molecular biologist on retinal stem cells. In 2004 he entered the Ph.D. program at the Sackler Institute at NYU, graduating in 2009 from the program in Physiology and Neuroscience. He then completed a post-doctoral fellowship under Oded Gonen, training on applications of proton MR spectroscopy. Ivan has been an independent investigator since 2014 as an Assistant Professor with research interests mainly in Traumatic Brain Injury and Multiple Sclerosis.

Tuesday, September 20th at noon

Teodora Chitiboi, PhD

Postodctoral Associate
Center for Biomedical Imaging
Department of Radiology
NYU School of Medicine
Myocardium Segmentation and Motion Analysis from Time-varying Cardiac Magnetic Resonance Imaging

Abstract: Magnetic Resonance Imaging (MRI) is a reference method for noninvasive examination of the global and local cardiac function. Using the latest real-time MRI sequences, cardiac function can be monitored over multiple consecutive heart beats, enabling the study of cardiac cycle variability, for example, in patients with arrhythmia. An essential precondition for the analysis of cardiac functional is the segmentation of the heart muscle (myocardium). To address this task, a hierarchical object-based segmentation approach was devised, which combines bottom-up region grouping with a top-down optimization strategy. This principle takes steps towards bridging the semantic gap between low-level image features and high-level, complex and heterogeneous structures. This algorithm is part of a comprehensive pipeline for automatic segmentation of the myocardium from short-axis MRI. Furthermore, tissue phase mapping (TPM) offers the means to inspect local cardiac motion by acquiring the velocity of individual myocardium voxels. This talk presents a semi-automatic probabilistic segmentation approach for TPM that combines contour displacement with particle tracing for estimating the uncertainty of the segmentation result. An automatic quantification method was additionally developed to compute global myocardial torsion.

About the speaker: Teodora Chitiboi received her PhD in Computer Science from Jacobs University Bremen, after having received a Bachelor and Master in Computer Science. Teodora was a researcher at Fraunhofer MEVIS in Bremen where she contributed to the development of an object-based image analysis (OBIA) library and was part of the group for Cardiovascular Research and Development. Her research interests are medical image analysis, visualization and image segmentation.


Taehoon Shin, PhD

Assistant Professor
Department of Diagnostic Radiology and Nuclear Medicine
University of Maryland, Baltimore
Advanced Magnetic Resonance Imaging Methods for Cardiovascular Applications

Abstract: This talk presents advanced MRI methods for two cardiovascular applications: non-contrast-enhanced (NCE) angiography and cardiac late gadolinium enhanced (LGE) imaging. First I will present new NCE MRA methods using Fourier based velocity-selective (VS) magnetization preparation which can generate positive vessel contrast in single acquisition with high spatial resolution in all three dimensions. The principle of VS excitation is explained under excitation k-space formalism, followed by a few designs with improved B0 and B1 immunity, and applications for various vascular territories. Second, I will present 3D LGE MRI methods based on stack-of-spirals acquisitions. Two strategies will be shown, including single breath-hold whole-heart LGE using parallel imaging acceleration, and free-breathing near-isotropic resolution LGE using outer-volume-suppressed projection-based navigator.

July 25th at noon

Pål Erik Goa, PhD

Associate Professor
Department of Physics
Norwegian University of Science and Technology
Diffusion Weighted MRI in Breast Cancer and the link to Tissue Microstructure

Abstract: Diffusion-weighted MRI (DWI) has become a standard component in most clinical breast MRI protocols. The main reason for this is the reduced apparent diffusion coefficient (ADC) observed in cancer tissue compared to healthy fibroglandular tissue and benign lesions. This effect is loosely attributed to increased cellularity and reduced extracellular volume in cancer tissue. Other flavours of DWI, like diffusion tensor imaging (DTI), diffusion kurtosis imaging (DKI), intravoxel incoherent motion (IVIM) and most recently stimulated echo diffusion imaging (STE-DWI), have also been applied in breast cancer MRI. Each of these methods show some interesting results, including differences between malignant and benign tissues.
Since DWI draws its contrast from the microscopic structural features of the tissue, the link between DWI and specific microstructural parameters has always been a topic of interest. However, due to the non-unique mapping from DWI-signal to microstructure, this has proven very difficult in practice. As an example, only weak correlation between cellularity and ADC has been shown in breast cancer. A biophysical interpretation of the DWI signal is therefore often avoided, and the results from DWI are analysed with respect to its link with other parameters, like malignancy/grade or molecular subtype.
In this presentation we will discuss this issue in some detail, looking at the structure of healthy and malignant tissues together with results from the various methods in DWI.

About the speaker: Pål Erik Goa obtained his PhD in Physics at the University of Oslo, Norway, in 2002, after building the first microscope capable to capturing the live motion of quantized magnetic flux-lines in a superconductor. After working on remote sensing applications for the Norwegian Defence Research Establishment for a couple of years, he and his family moved to Trondheim in 2005, and he started on a new career in the field of medical imaging. In the period 2006-2013 he worked as MR-physicist at the Department of Radiology and Nuclear Medicine, St.Olavs Hospital, and in 2013 he took up the position as Associate Professor in Medical Physics at NTNU. Goa has been involved in a wide variety or research projects in MRI, both clinical and pre-clinical, ranging from retro-gated cardiac MRI in mice to the development of new sequences for BOLD-fMRI at 7 T. His current research interest is focused on the application of different methods of Diffusion-Weighted MRi in Cancer Management.

Tuesday, June 14th at noon

Alexandra Badea, PhD

Assistant Professor of Radiology
Center for In Vivo Microscopy
Duke Medical Center, Durham, NC
Learning from Small Animal Models of Neurological Conditions – An MR-Based Phenotyping Approach

Abstract: The rich contrast and flexibility of MR offers the possibility of quantifying multiple image based biomarkers in small animal models of neurological and psychiatric conditions. This talk focuses on a mouse model of Alzheimer’s disease (AD). Mouse models provide opportunities to study characteristics of AD in well-controlled environments and can facilitate early interventions. Multivariate biomarkers are needed for detecting AD, helping to understand its etiology, and quantifying the effect of therapies. The CVN-AD mouse model replicates multiple AD hallmark pathologies, and we identified multivariate biomarkers characterizing a brain circuit disruption predictive of cognitive decline. We used manganese-enhanced MRI to locate areas of differential uptake of manganese in CVN mice relative to age matched controls, and in association with learning and memory deficits. In vivo and ex vivo MRI revealed that CVN-AD mice replicate the hippocampal atrophy (6%), characteristic of humans with AD, and also present changes in subcortical areas. The largest effect was in the fornix (23% smaller), which connects the septum, hippocampus, and hypothalamus. In characterizing the fornix with diffusion tensor imaging, fractional anisotropy was most sensitive (20% reduction), followed by radial (15%), and axial diffusivity (2%) in detecting pathological changes. These findings were strengthened by optical microscopy and ultrastructural analyses. CD68 staining showed that white matter pathology could be secondary to neuronal degeneration or due to direct microglial attack. In conclusion, these findings strengthen the hypothesis that the fornix plays a role in AD, and can be used as a disease biomarker and as a target for therapy.

About the speaker: Dr. Badea is an Assistant Professor in the Department of Radiology at Duke, and a member of the Center for In Vivo Microscopy, where she co-directs the Visual Informatics Core, while maintaining her focus on models of neurodegenerative conditions. She was born in Romania, where she studied Physics for her BSc. She graduated with a PhD in Biomedical Engineering from the University of Patras, Greece, where she has learned to love working with images, and in particular brain images. Her interest in computational imaging has led to the development of image processing pipelines for structural and diffusion imaging. She uses such pipelines with the aim to understand the lessons that small mouse models can teach us about human neurological and psychiatric conditions.

May 31st at noon

Kerstin Hammernik

Doctoral Candidate
Research Assistant
Graz University of Technology
Insights into deep learning for MRI reconstruction

Abstract: Compressed sensing techniques allow MRI reconstruction from under sampled k-space data. However, most reconstruction methods suffer from high computational costs and are limited to low acceleration factors for non-dynamic 2D imaging protocols. Furthermore, existing image reconstruction methods are based on simple regularizers such as sparsity in the wavelet domain or Total Variation (TV). However, these simple and handcrafted regularizers make assumptions on the underlying image statistics and the reconstructed images appear unnatural. In this work, we propose a novel and efficient approach to overcome these limitations by learning a sequence of optimal regularizers that removes typical undersampling artifacts while keeping important details in the imaged objects and preserving the natural appearance of anatomical structures. We test our approach on patient data and show that we achieve superior results in terms of both runtime and image quality compared to commonly used reconstruction methods.

About the speaker: Kerstin Hammernik received a BSc and MSc in Biomedical Engineering from Graz University of Technology in 2011 and 2015, respectively. Currently,she is a research assistant and PhD student supervised by Dr. Thomas Pock at the Institute of Computer Graphics and Vision, Graz University of Technology. Her current research interests include optimization and learning of variational models with application to medical inverse problems such as magnetic resonance (MR) and photo acoustic image reconstruction.

Special Seminar: May 27th at noon

William Grissom, PhD

Assistant Professor of Biomedical Engineering
Vanderbilt University Institute of Imaging Science (VUIIS)
Nashville, TN
Array-Compressed Parallel Transmit MRI

Abstract: Many-coil transmit arrays are desirable in parallel transmission (pTx), since with many coils multidimensional pulses can be shortened, more uniform radiofrequency shims can be produced, and specific absorption rate can be more effectively controlled. However, the high cost and the large physical footprint and cabling requirements of the corresponding power amplifiers required to drive many-coil arrays has limited the number of transmit coils/channels that are used in practice, and most ultra-high field MR scanners in use today have only eight transmit channels. Inspired by recent work in MRI receive array compression, we proposed array-compressed pTx (acpTx) to overcome these limitations. In acpTx, a large number of coils is connected to a small number of channels via a virtual or physical array compression network that splits the input pulse waveforms to the coils and applies attenuations and phase shifts that are optimized jointly with the pulse waveforms. In this way, the excitation spin physics directly informs the construction of the compressed transmit coil array. This talk will describe how pulses can be designed for acpTx and how it can be implemented in hardware. We will also talk about its potential embodiments and impact on ultra-high field MRI, including how it might be used to improve transmit coil design.

Special Seminar: May 26th at noon

Ernest Feleppa, PhD, Research Director
Jeffrey Ketterling, PhD, Associate Research Director
Jonathan Mamou, PhD, Research Manager
Daniel Rohrbach, PhD, Member of the Research Staff

Biomedical Engineering at Riverside Research

Abstract: Riverside Research has been engaged in biomedical research since it was the Electronics Research Laboratory of Columbia University in the 1950s. Over the past several decades, the Biomedical Engineering Laboratory at Riverside Research has become internationally recognized as a leader in advanced biomedical ultrasound. The history, capabilities, and interests of the Laboratory will be summarized.

High-frequency ultrasound annular-array probes operating at frequencies higher than 15 MHz provide resolution superior to linear arrays operating at the same frequencies. We have developed custom imaging systems based on five-ring, 20-MHz and 40-MHz annular arrays, and have shown that the devices permit a significant improvement in image quality over current technology for small-animal- and ophthalmic-ultrasound imaging. An overview of the systems and examples of in vivo human and in utero mouse-embryo scans will be shown. Extensions of the work to photoacoustic imaging of mouse embryos as well as applications such as characterization of the human vitreous and analysis of brain development in mouse embryos will be discussed.

Quantitative ultrasound (QUS) methods permit characterizing tissue microstructure at a sub-resolution level. Our group is considered to be a pioneer in using these methods for tissue characterization. As an example, a high-frequency ultrasound study focusing on 3D imaging and characterization of lymph nodes freshly-excised from cancer patients will be presented. QUS images were formed and used to detect metastases using a transducer that has a 26-MHz center frequency. Classification results suggest that these QUS methods may provide a clinically-important means of identifying small metastatic foci that might not be detected using standard pathology procedures.

Scanning acoustic microscopy (SAM) is a well-established method for fine-resolution material characterization in particular for non-destructive testing. However, accurate estimation of the mechanical properties for soft-tissue applications is still challenging. We developed a novel quantitative acoustic microscope (QAM) operating at 250-MHz and 500-MHz center frequencies that allows characterizing soft-tissue material properties (i.e. mass density and bulk modulus) at resolutions down to four micrometers. The presentation will provide an overview of the device and its working principle. We will present current research results obtained for ophthalmologic tissues and human lymph nodes, and will discuss the potential applications for the measured material properties.

May 3rd at noon.

Frederic Noo, PhD

Professor of Radiology and Imaging Sciences
The University of Utah
Salt Lake City, Utah
Early assessment of image quality in X-ray computed tomography using model-based iterative reconstruction

Abstract: Radiation dose associated with CT scans has become an important concern in medical imaging. Fortunately, there are many pathways to reducing dose, one of which amounts to using a model-based iterative reconstruction method. A major strength of this approach is its flexibility: there are many ways to design such a reconstruction, allowing adaptation to both anatomy and diseases. This strength however comes with major challenges in terms of gain assessment. Early assessment of image quality, before clinical deployment, is critical to identify and refine solutions. Moreover, given the non-linear nature of model-based iterative reconstruction methods, task-based assessment must be embraced, which further complicates the problem. Currently, there are few publications reporting on early, task-based assessment of image quality achieved with iterative reconstruction methods. This talk will present results in this direction using LROC analysis with computer-simulated data read by human observers. At the same time, it will be demonstrated that the grayscale used for image display is a critical factor in such image quality comparison studies.

About the speaker: Frederic Noo is a Professor of Radiology and Imaging Sciences at The University of Utah. His education took place in Belgium, where he completed a Ph.D. degree in Engineering Sciences in 1998, with emphasis on image reconstruction problems in cone-beam tomography. The National Science Foundation in Belgium supported his research from 1993 until 2001, first as a Ph.D. student, then as a post-doctoral research. In 2001, he decided to move to The University of Utah, where he had built strong collaboration ties. Since then, he has expanded his range of expertise to encompass all aspects of X-ray computed tomography, including image reconstruction algorithms, scanner design, simulation models and Monte-Carlo transport of photons, noise and dose evaluations, and task-based assessment of image quality using both model and human observers. His publications include 65 peer-reviewed articles, 112 conference proceedings, and 9 patents. His CT expertise is recognized in the industry as well as in the academia. He launched a highly successful biennial stand-alone conference in 2010, called "The International Conference on Image Formation in X-ray Computed Tomography". This effort was offered as a community service to address a growing need for CT scientists. His work has been continuously supported by the NIH and by corporate funds since 2001. He has supervised a number of Ph.D. students and post-doctoral researchers, who have become prolific scientists with Siemens, GE, Philips, and the FDA. A number of his image reconstruction methods are or have been used by vendors; and the FDA supports his methods for image quality assessment.

April 26th at noon.

Nicole Wake

Graduate Student
Biomedical Imaging Program
Sackler Institute, Radiology Department
NYU School of Medicine
3D Printing: Applications in Urologic Oncology

Abstract: Three-dimensional (3D) printing in radiology represents the fabrication of physical objects from imaging data, with the intent of impacting patient care. 3D printing of anatomical data allows radiologists, surgeons, and other physicians to physically hold in their hands patient-specific models and use visuo-haptic inputs to better understand both complex anatomy and the condition being treated. In this talk, I will describe the steps required to derive anatomically accurate, patient-specific models in the context of urological oncology. In particular, the application of 3D printing in the pre-operative evaluation of prostate and kidney cancer will be demonstrated.

About the speaker: Nicole Wake received her Bachelors in Biology and History from the University of Pennsylvania and her Masters in Science from the Mount Sinai School of Medicine. She has extensive experience working as a research assistant in a cardiovascular CT and MR imaging lab at Brigham and Women's Hospital, Boston, MA. Nicole is currently a PhD Candidate at NYU School of Medicine, where she works under the direction of Hersh Chandarana and Daniel K. Sodickson on applications of 3D printing in urologic oncology.

April 19th at noon.

Dan Wu, PhD

Research Associate
Department of Radiology
Johns Hopkins University School of Medicine
Road onto Microstructural Imaging: Diffusion MRI at high-resolution and varying time scales

Abstract: Diffusion MRI is a powerful tool for noninvasive mapping of the microstructural organization in the brain. One part of my work focuses on developing high-resolution in vivo imaging techniques to resolve structures and connections in the live mouse brain. With a localized high-resolution imaging technique, we achieved in-utero diffusion MRI of the embryonic mouse brain. I also worked on probing brain microstructural features using oscillating gradient diffusion MRI (OGSE). In a neonatal mouse model of hypoxia-ischemia, OGSE diffusion MRI showed drastic change of contrast in the edema tissue and enhanced sensitivity in mild edema region compared to conventional pulsed gradient diffusion MRI. We have explored the diffusion-time dependence of kurtosis property of water diffusion and the time dependence of intra-voxel incoherent motion at varying oscillating frequencies. These work may lead to better understanding of the relation between diffusion MRI signals and the underlying tissue microstructural properties.

About the speaker: Dr. Wu obtained Masters and PhD degrees from Johns Hopkins University, Department of Biomedical Engineering, where she conducted the thesis study mainly on diffusion MRI. Currently, she is a Research Associate in the Department of Radiology at Hopkins, starting her independent research in the technical development and biomedical applications of advanced diffusion MRI techniques.

Special Seminar: April 8th at 2:00 p.m.

Choong Heon Lee, PhD

Postdoctoral Researcher
Rush University
Chicago, IL
Magnetic Resonance Microscopy from Tissues to Potential Clinical Applications

Abstract: MR microscopy has developed over the last 25 years as a complementary microimaging technique. Although it offers the potential to study tissues in vivo, the inherently low sensitivity of NMR has limited MR microscopy to the study of relatively large cells, i.e. frog ova (~1mm in diameter) and Aplysia neurons (~ 300-350 μm in diameter). Recently, using new surface microcoils and high field magnets to improve sensitivity, we performed the first MR microscopy of neurons in mammalian tissue, and potential identification of mammalian nuclei in the tissue. These findings have the potential to change the way we interpret clinical MR images by revealing unique signal and contrast characteristics of the microstructural elements that comprise tissues: perikarya, nuclei, neurites, vasculature etc. Developing a better understanding of subcellular elements and how their MR characteristics change under the influences of pathology will lead to advances in tissue modeling and provide diagnostic criteria for earlier and more accurate disease detection. Such improvements are critically needed in the case of neuropathologies which often present with abnormalities at the cellular level many years prior to the development of symptoms which spur patients to seek treatment. In this work, we offer an overview of the progress made in MR microscopy of neural tissues and non-neural tissue applications, and potentials of offering a better references to clinical treatment.

April 5th at noon

Assaf A. Gilad, PhD

Associate Professor
Russell H. Morgan Department of Radiology and Radiological Science
Division of MR Research and the Institute for Cell Engineering
Johns Hopkins University
Developing MRI reporter genes for optimal drug, virotherapy, and stem cell delivery

Abstract: The tremendous developments in the field of (bio) medical imaging that have revolutionized modern medicine have opened a new niche for technologies that enable the collection of information above and beyond anatomical, metabolic, and functional information. Our lab has been focusing on the development of one such technology, which is based on genetically encoded systems that can generate MRI contrast from specific cellular and molecular events. These are genes–synthetic, semi–synthetic, and adopted from other organisms that we introduced into the cell's genome. These genes, once expressed, can be used for numerous applications. Here, we will demonstrate how such genes can be used to monitor:

  • Cell-specific expression
  • Drug delivery
  • Oncolytic viral therapy in cancer
  • Cancer immunotherapy
  • Tracking transplanted stem cells in the heart.
While most of the studies were performed in live rodents, we have recently demonstrated the feasibility of these technologies in pigs, using clinical MRI scanners. Our research is a part of an on-going effort to expand the toolkit of MRI technologies for more comprehensive diagnostics.

Special Seminar: March 30th at noon

Wyger Brink

Department of Radiology
Leiden University Medical Center
The Netherlands
Dielectric Shimming – Exploiting Dielectric Interactions in High Field MR

Abstract: One of the main challenges in MR operation at high fields is to acquire images when the dimensions of the body section being imaged are comparable to the RF wavelength. The resulting RF interferences within the body can severely reduce diagnostic image quality. However, the underlying electromagnetic interactions also raise the question of whether these mechanisms may be exploited to improve performance. This approach, termed "Dielectric Shimming," is a very simple method which allows for adjusting the radiofrequency (RF) fields in high field MR. Previous work has shown that this can improve MR operation in various body applications at 3T as well as neuro applications at 7T. Currently, numerical methods are being developed to harness and exploit this approach.

Special Seminar: March 25th at noon

Luis J. Garay

Associate Professor
Universidad Complutense de Madrid
Pieces of time

Abstract: In this overview presentation I will describe from a personal—and necessarily biased—point of view a few aspects of TIME which I have dealt with during my research and teaching on gravity and quantum theory. They encompass various contexts: from Newtonian mechanics and general relativity to quantum gravity and the microscopic structure of spacetime.

About the speaker: Luis J Gray is a Associate Professor at the Universidad Complutense de Madrid. His area of research is classical and quantum gravity. In particular he has worked on black holes, quantum fields in curved backgrounds, Hawking radiation, Analog models of gravity, emergent gravity, acoustic black holes in Bose-Einstein condensates, Quantum gravity and cosmology, Relativistic Quantum Information.

Special Seminar: March 23rd at 2:00 p.m.

Marcelo Victor Wüst Zibetti, Dr. Eng.

Associate Professor
Graduate Program in Electrical and Computer Engineering
Federal University of Technology - Paraná (UTFPR), Brazil
Visiting Scholar
Computer Science
Graduate Center, City University of New York (CUNY)
Improving compressed sensing in MRI with separate magnitude and phase priors

Abstract: Compressive sampling/compressed sensing (CS) has shown that it is possible to perfectly reconstruct non-bandlimited signals sampled well below the Nyquist rate. Magnetic Resonance Imaging (MRI) is one of the applications that has benefited from this theory. Sparsifying operators that are effective for real-valued images, such as finite difference and wavelet transform, also work well for complex-valued MRI when phase variations are small. As phase variations increase, even if the phase is smooth, the sparsifying ability of these operators for complex-valued images is reduced. If the phase is known, it is possible to remove it from the complex-valued image before applying the sparsifying operator. Another alternative is to use the sparsifying operator on the magnitude of the image, and use a different operator for the phase, i.e., one related to a smoothness enforcing prior. The proposed method separates the priors for the magnitude and for the phase, in order to improve the applicability of CS to MRI. An improved version of previous approaches, by ourselves and other authors, is proposed to reduce computational cost and enhance the quality of the reconstructed complex-valued MR images with smooth phase. The proposed method utilizes L1 penalty for the transformed magnitude, and a modified L2 penalty for phase, together with a non-linear conjugated gradient optimization. Also, this paper provides an extensive set of experiments to understand the behavior of previous methods and the new approach.

March 22nd at noon

Gang Chen

Graduate Student
Biomedical Imaging Program
Sackler Institute, Radiology Department
NYU School of Medicine
Approaching the Ultimate Intrinsic SNR with dense arrays of electric dipole antennas

Abstract: Radiofrequency(RF) Coil designs motivated by the ideal current patterns corresponding to the Ultimate Intrinsic SNR (UISNR) have been used to boost central SNR at 3T and 7T for MR imaging. For a cylindrical phantom and a current distribution defined on a concentric cylindrical surface, the ideal current pattern for optimal central SNR includes both divergence-free and curl-free components. At low field, divergence-free current patterns saturate the UISNR and arrays with an increasing number of loops can approach the UISNR. While loops are exclusively divergence-free, recent work has shown that electric dipole antennas include both divergence-free and curl-free current components. To shorten a dipole compared to its self-resonant l/2 length it is necessary to incorporate inductors, which are lossy. In this talk I will present that arrays with an increasing number of electric dipole antennas can approach UISNR for all currents in the center of a head-sized phantom at 7T despite these losses.

Special Seminar: March 16th at 2:00 p.m.

Pep Pàmies, PhD

Chief Editor
Nature Biomedical Engineering
Help shape Nature Biomedical Engineering

Abstract: Launching in January 2017, Nature Biomedical Engineering will publish original research, reviews and commentary of high significance to the biomedical engineering community, including bench scientists interested in devising materials, methods, technologies or therapies to understand or combat disease; engineers designing or optimizing medical devices and procedures; and clinicians leveraging research outputs in biomedical engineering to assess patient health or deliver therapy across a variety of clinical settings and healthcare contexts. In this discussion, the Chief Editor will welcome suggestions about what the journal could do for your field and for the broader biomedical engineering community.

About the speaker: Pep is leading the editorial team of Nature Biomedical Engineering. He has been an editor for Nature Materials for more than 5 years, where he championed the biomaterials content, handling manuscripts and commissioning articles in a wide variety of subjects, including tissue engineering, medical imaging, regenerative medicine, cancer therapy and diagnostics. Previously, Pep conducted research in computational soft matter and biophysics at Columbia University's Chemistry Department in New York City, at the Max Planck Institute of Colloids and Interfaces in Potsdam, and at the Atomic and Molecular Physics Institute in Amsterdam. Pep obtained a PhD in Chemical Engineering in December 2003 from Rovira i Virgili University in Catalonia, Spain.

March 15th at noon

Alberto Pepe

Research Associate
Harvard University
Data-driven, Interactive Scientific Articles in a Collaborative Environment with Authorea

Abstract: Most tools that scientists use for the preparation of scholarly manuscripts, such as Microsoft Word and LaTeX, function offline and do not account for the born-digital nature of research objects. Also, most authoring tools in use today are not designed for collaboration and as scientific collaborations grow in size, research transparency and the attribution of scholarly credit are at stake. In this talk, I will show how Authorea allows scientists to collaboratively write rich data-driven manuscripts on the web–articles that would natively offer readers a dynamic, interactive experience with an article’s full text, images, data, and code–paving the road to increased data sharing, data reuse, research reproducibility, and Open Science.

About the speaker: Alberto Pepe is the co-founder of Authorea. He recently finished a Postdoctorate in Astrophysics at Harvard University. During his postdoctorate, Alberto was also a fellow of the Berkman Center for Internet and Society and the Institute for Quantitative Social Science. Alberto is the author of 30 publications in the fields of Information Science, Data Science, Computational Social Science, and Astrophysics. He obtained his Ph.D. in Information Science from the University of California, Los Angeles with a dissertation on scientific collaboration networks which was awarded with the Best Dissertation Award by the American Society for Information Science and Technology (ASIS&T). Prior to starting his Ph.D., Alberto worked in the Information Technology Department of CERN, in Geneva, Switzerland, where he worked on data repository software and also promoted Open Access among particle physicists. Alberto holds a M.Sc. in Computer Science and a B.Sc. In Astrophysics, both from University College London, U.K. Alberto was born and raised in the wine-making town of Manduria, in Puglia, Southern Italy.

February 23th at noon

Gillian Haemer

Graduate Student
Biomedical Imaging Program
Sackler Institute, Radiology Department
NYU School of Medicine
Incorporation of high permittivity materials into RF transmit coil design

Abstract: High permittivity, low conductivity materials (HPMs) placed between RF coils and a subject can be used to passively vary the spatial distribution of electric and magnetic fields, independent of or in combination with RF shimming or parallel transmission. This field redistribution has the potential to improve both receive sensitivity and transmit efficiency, and therefore HPMs have the potential to greatly benefit transmit-receive coil array design. In this talk I will present a method for determining the optimal relative permittivity and placement of HPMs close to a transmit array, and the practical restrictions that come along with placing materials with very high permittivities close to resonant loops.

About the speaker: Gillian Haemer received here Bachelors in Biomedical (Electrical) Engineering from the University of Southern California. During her time in Los Angeles she discovered medical imaging research working as a research assistant on CTA/SPECT data registration at Cedars Sinai Medical Center. She then completed her Masters at the joint program in Biomedical Engineering and Medical Imaging at the University of Tennessee and the University of Memphis, with the design and development of a prototype variable-resolution x-ray breast CT scanner. She is currently a PhD student at the NYU School of Medicine, where she works under the direction of Daniel K Sodickson and Graham C Wiggins on MRI hardware engineering challenges at ultra high field strengths.

Special Seminar: February 19th at noon

Marios Georgiadis, PhD

Postdoctoral Fellow
ETH Zurich
X-ray scattering for microstructural anisotropy of tissues—the examples of bone and brain

Abstract: Small-angle X–ray scattering (SAXS) occurs when part of the X–ray beam that probes a sample is scattered at small angles, due to differences in electron density distributions within the sample. Moreover, it gives a particularly strong signal in the presence of ordered and periodic systems, that act like slits. Recently, we developed two techniques based on SAXS that can reconstruct the 3D organization of tissue microstructure. In the first technique, called 3D scanning SAXS, local 3D tissue anisotropy is derived by scanning thin sections at different rotation angles. In the second technique, called small–angle scattering tensor tomography, a non–destructive method to reconstruct local anisotropy is introduced by the use of a second sample rotation axis and an iterative reconstruction algorithm based on spherical harmonics. Small-angle scattering tensor tomography extends the concept of traditional tomography: it reconstructs not only scalar values, but multiple parameters per voxel, providing a 3D representation of local material anisotropy. These methods were demonstrated for reconstructing the orientation of the mineralized collagen fibrils in bone trabeculae. Similar experiments can also be performed in other tissues and materials which exhibit structural anisotropy, such as the human brain.

About the speaker: Marios Georgiadis received his Mechanical Engineering diploma from the National Technical University of Athens, Greece. He did his Masters studies in Biomedical Engineering at ETH Zurich, Switzerland, where his thesis “Microfluidic probe for tissue staining in advanced pathology” was awarded the ETH medal. In his PhD at the Institute for Biomechanics of ETH Zurich he developed methods for investigating local tissue anisotropy using X-ray scattering, and applied that to investigate local tissue anisotropy of human trabecular bone. He was runner-up for the Student Award of the European Society of Biomechanics in 2015. He is currently at the Institute for Biomedical Engineering of ETH Zurich and the University of Zurich, where he will be looking at the microstructural anisotropy of brain tissue using X-ray scattering and diffusion MRI.

February 9th at noon

Gregory Lemberskiy

Graduate Student
Biomedical Imaging (BIO) Program
Sackler Institute, Radiology Department
NYU School of Medicine
Time-Dependent Diffusion in the Body

Abstract: Diffusion of water molecules is directly influenced by the biological tissue architecture at the micrometer length scale. Capturing this effect using Diffusion MRI has led to the development of numerous applications including early detection of stroke and cancer. However, despite the overwhelming tissue complexity, important non-Gaussian nuances of the diffusion signal are ignored. I have been focused on using time-dependent diffusion as a probe for non-Gaussian behavior within the muscle and prostate. This presents a unique challenge, as acquisition techniques such as STEAM, PGSE, and OGSE, must be tailored to the tissue of interest. In this talk, I will discuss the clinical motivation for pursuing time-dependent diffusion as well as advances in diffusion modeling and acquisition.

About the speaker: Greg immigrated from the city of Tomsk, Russia to Brooklyn, NY in 1994, where he grew up and attended school in Sheepshead Bay. He attended NYU as an undergrad, where he majored in Physics and followed the premed track. However, after numerous experiences working with and shadowing both scientists and clinicians he concluded that a PhD in MRI physics was a suitable middle ground between the two disciplines. He has since been working closely with Dmitry S. Novikov and Els Fieremans on development of Time-Dependent diffusion applications in the body.

Special Seminar: February 5th at 1:00 p.m.

Kyunghyun Cho, PhD

Assistant Professor
Department of Computer Science
New York University
Deep Learning: what it is and what it is becoming

Abstract: Deep learning has become one of the hottest topic in machine learning research in recent years. It began with the 2012 breakthrough in computer vision, the breakthrough that essentially transformed the whole field of computer vision. This breakthrough was followed by those in automatic speech recognition, natural language processing and machine translation. Beyond these recent success stories, deep learning promises much more especially in the areas of multimodal, multitask learning and sequential decision making. In this talk, I will start with a high-level overview on deep learning and discuss these future promises and challenges.

Special Seminar: February 3rd at noon

Baiyu Chen, PhD

Research Fellow
Department of Radiology
Mayo Clinic
Development and optimization of CT reconstruction algorithms – Challenges and my solutions

Abstract: The rising public concerns on CT radiation dose have greatly motivated the development of dose-reducing reconstruction algorithms. However, the development of a reconstruction algorithm is challenged by several aspects. First, collecting CT projection data (i.e., raw materials for CT reconstruction) is challenging: The CT projection data acquired on commercial CT scanners are proprietary and vendor-specific, and therefore not accessible to researchers who do not have research agreements with the vendor. Second, optimizing the algorithm is challenging: Any optimization needs to use diagnostic accuracy as the end goal, but the assessment of diagnostic accuracy via reader studies is time-consuming. Last but not least, validating the algorithm via clinical trials is challenging: The process can be very expensive and labor-intensive. In this talk, I will discuss solutions to these three challenges using examples from my current research: A library of patient projection data in an open and standardized format; a mathematical model that predicts the detection performance of human observers based on the image quality, the viewing condition, and the lesion characteristic; and a computer program that creates positive cases for clinical trials by inserting lesion of known characteristics into images of healthy patients. The same framework not only facilities the development of CT reconstruction algorithms, but can also be adapted by clinical practices (such as the optimization of clinical protocols) to improve diagnostic performance.

January 26th at noon

Matthew Gounis, PhD

Associate Professor
University of Massachusetts Medical School
Worcester, MA
From Bench to Brain: Toward Quantitative Assays of Brain Aneurysm Vulnerability

Abstract: The last two-decades have seen an explosion of technology to improve upon the safety and efficacy of brain aneurysm treatment. Despite remarkable improvements in treatment modalities, risk of severe neurological morbidity varies between 5 and 15% of patients with treated unruptured aneurysms. In parallel, increased access to noninvasive neuroimaging has led to a historically unprecedented detection rate of unruptured brain aneurysms. Although the risk of aneurysm rupture is often quite low, the consequences of aneurysmal subarachnoid hemorrhage are devastating with approximately half of the patients not surviving the rupture. Therefore, an approach enabling appropriate selection of patients who would benefit from treatment is urgently needed. Currently, best evidence indicates that size, ethnicity (Finnish, Japanese, or other), location, prior history of subarachnoid hemorrhage, and hypertension should all be considered. Other potential factors elevating rupture risk are family history of subarachnoid hemorrhage, cigarette smoking, and aneurysm morphology. However, given the uncertainty of aneurysm pathophysiology in the progression toward rupture, the precise model to accurately predict aneurysm rupture risk remains elusive. Over the last decade, a plethora of data from human brain aneurysm specimens as well as animal models of intracranial aneurysms has highlighted the role of aneurysm wall inflammation in mural destabilization. Our leading hypothesis is that stable aneurysms can become active, and hence undergo a process of remodeling that involves the invasion of immune cells. This invasion and pursuant inflammation precedes the breakdown of the structural components of the aneurysm wall. Capitalizing on models of vulnerable plaque, we have focused our efforts on in vivo imaging to detect active myeloperoxidase (MPO) in brain aneurysms as a precursor to structural destabilization. We have identified that human brain aneurysms that contain MPO have a statistically higher estimated 5-year rupture risk. Logistic regression modeling of 5-year aneurysm rupture risk and irregular aneurysm morphology when coupled are strong predictors of histologically confirmed MPO presence. Taken together, these data on human brain aneurysm specimens indicate the potential role of MPO as a biomarker for aneurysm instability. In parallel, MR probes have been tested in both animal models of inflamed aneurysms as well as using a unique micro-MRI approach to imaging human brain aneurysm specimens to quantify MPO presence. In summary, MPO detection by MRI may provide clinicians critical information on aneurysm wall biology to make informed decisions regarding treatment.

Special Seminar: January 14th at noon

Mootaz Eldib

Doctoral Candidate
Senior Associate Researcher
Mount Sinai Mecical Center
Optimization of PET Imaging on Simultaneous PET/MR Scanners

Abstract: With the recent introduction of simultaneous PET/MR imaging, various opportunities exist to utilize the co-acquired MRI data to improve the quantitative accuracy of the PET component of the scanner. In this presentation, I will introduce several MR-guided methods for PET attenuation and motion correction focusing on cardiovascular and liver imaging applications.

Special Seminar: January 13th at noon

Koen Michielsen

Department of Radiology
University Hospitals Leuven
Maximum Likelihood Reconstruction for Breast Tomosynthesis and Model Observer Evaluation

Abstract: Digital breast tomosynthesis is a recent imaging modality gaining acceptance as valuable diagnostic tool. Clinical evaluations have shown that when combined with digital mammography it improves diagnostic accuracy and reduces recall rate. When looking closer at the different lesion types, evidence points to improved visualization of masses and distortions, but potentially worse visualization of microcalcification clusters. Therefore, we improved the visualization of these microcalcification by expanding the forward model of the maximum likelihood for transmission tomography reconstruction to include an exposure specific resolution model and modified the update sequence to obtain faster convergence. Concurrently, we developed a channelized Hotelling observer that can predict human observer performance when evaluating detectability of microcalcification targets in a structured phantom background.

About the speaker: Koen Michielsen obtained the degree of Bachelor of Science at the University of Hasselt in 2003, and continued his education at KU Leuven where he graduated cum laude in 2005 with the degree of Master of Science in Physics, with a thesis on "Determining the time delays of lensed quasar J1155+635 from a series of CCD images". After receiving a second Master degree in 2007, this time in the field of Medical Radiation Physics and with a thesis titled "Automated data collection strategies and results for patient dosimetry in mammography", he started working as a certified medical physics expert at the department of radiology of the University Hospitals in Leuven. He worked there until December 2010, when he started his PhD project at the department of Imaging and Pathology at KU Leuven on the topic "Maximum a Posteriori Reconstruction of Limited Angle Tomography".

January 12th at noon

Antonios Papaioannou

Doctoral Candidate
Research Assistant
City University of New York
Probing structural disorder and permeability of porous media with diffusion NMR

Abstract: Characterizing the most relevant geometric structure of complex systems with a single transport measurement is central in many fields. Such characterization of the underlying structural complexity may contribute to early detection of cerebral ischemia, optimization of oil production from rock formations or characterization of complex biological networks such as protein–interaction networks. The increased structural complexity of such systems—or disorder—makes the establishment of relations between dynamic parameters, such as the diffusion coefficient, and the underlying geometric structure a challenging problem. Disorder may be categorized in a "handful" of universality classes which lead to distinct long time power law behaviors of the diffusion coefficient, characterized by the dynamical exponent θ. In this seminar, an introduction of the theory of classical transport in disordered media will be discussed. A direct experimental validation of the universal scaling of the diffusion coefficient of H2O diffusing through a homemade phantom of polycarbonate permeable films in a well defined geometry will be presented. In addition, structural parameters such as the diffusive permeability and structural disorder class of the phantom are experimentally determined. Other topics of the seminar include Pulsed Field Gradient NMR techniques and NMR probe development techniques for spin diffusion measurements.

About the speaker: Antonios received his Bs in Physics from University of Ioannina, Greece and is currently a Ph.D. candidate at the physics department of the City University of New York, The Graduate Center under the supervision of Gregory Boutis. His Ph.D. thesis focuses on classical transport in disordered systems. During his Ph.D he also had collaboration with Ravinath Kausik and Yi-Qiao Song (Schlumberger Doll Research-Boston) working on methane gas adsorption in disordered media. He is also interested in the statistical mechanics of complex networks (collaboration with Hernan Makse-CUNY).

January 8th at noon

Christopher Kroenke, PhD

Associate Professor
Oregon Health & Science University
MRI methods to assess fetal brain development and placental function: Application to fetal alcohol exposure

Abstract: The efficacy of therapeutic interventions for neurodevelopmental disorders improves when the disorder is detected early central nervous system development. We have developed MRI strategies for characterizing neural maturation in the fetal cerebral cortex, and for monitoring placental function, throughout the second half of the gestational period. To assess the sensitivity of these fetal MRI methods, we have developed a nonhuman primate model of fetal alcohol spectrum disorders. In this context we demonstrate the utility of MRI for precise characterization of perturbations to normal fetal development.

About the speaker: Dr. Kroenke's research group focuses on developing MRI strategies for characterizing the biological bases of neurodevelopmental disorders. Dr. Kroenke received his PhD in molecular biophysics and biochemistry at Columbia University. He then completed postdoctoral studies in the Washington University Department of Radiology. Dr. Kroenke is currently Associate Professor of Behavioral Neuroscience, and Associate Scientist in the Oregon Health & Science University Advanced Imaging Research Center and Oregon National Primate Research Center.

Special Seminar: December 17th at noon

Garry Gold, MD

Professor of Radiology
Stanford University
Vice President, ISMRM
December 15th at noon

Arvind P. Pathak, PhD

Division of Cancer Imaging Research
Russell H. Morgan Department of Radiology and Radiological Science
The Sidney Kimmel Comprehensive Cancer Center
The Johns Hopkins University School of Medicine
Imaging the Tumor ‘Vasculome’

Abstract: Angiogenesis or new blood vessel formation is one of the ‘hallmarks’ of cancer and necessary for tumor progression and metastasis. However, tumor blood vessels are structurally and functionally abnormal compared to vessels in healthy tissue. These abnormalities profoundly affect tumor hemodynamics, metastatic potential, and drug delivery. A recent explosion in imaging technologies has revolutionized our understanding of the role of the tumor vasculature and these phenomena. This lecture will highlight new 3D imaging techniques for visualizing the tumor vasculature; strategies for imaging the vascular phenotype at different spatial scales; and describe how 3D imaging data that quantify tissue morphology and molecular factors can be used in computational models of cancer and image contrast. The integration of preclinical cancer imaging data lays the ground work for systems biologists to map the ‘vasculome’ of a wide array of diseases. Mapping the tumor vasculature using multiscale imaging and modeling also enhances our understanding of the tumor microenvironment. Collectively, these advances enable us to relate the genotype to the vascular phenotype, identify novel drug targets and develop reliable clinical biomarkers of cancer.

About the speaker: Arvind P. Pathak received the BS in Electronics Engineering from the University of Poona, India. He received his PhD from the joint program in Functional Imaging between the Biophysics Department at the Medical College of Wisconsin and the department of Biomedical Engineering at Marquette University, Milwaukee, Wisconsin. During his PhD he was a Whitaker Foundation Fellow. He completed his postdoctoral fellowship at the Johns Hopkins University School of Medicine in the Molecular Imaging Program. He then joined the faculty of the Departments of Radiology and Oncology at Johns Hopkins. His cancer imaging research has been recognized by numerous journal covers and awards including the Bill Negendank Award from the International Society for Magnetic Resonance in Medicine (ISMRM) given to “outstanding young investigators in cancer MRI” and the Career Catalyst Award from the Susan Komen Foundation.

Special Seminar: December 14th at noon

Michael Garwood, PhD

Malcolm B. Hanson Professor of Radiology
Center for Magnetic Resonance Research and Department of Radiology
University of Minnesota
Performing MRI outside the usual technical boundaries
December 4th at noon

Todd Constable, PhD

Professor of Radiology and Biomedical Imaging and of Neurosurgery
Director of MRI Research
Yale University
Functional Connectome Fingerprinting: Connectivity Profiles Are Both Unique and Meaningful

Abstract: This presentation will discuss the uniqueness of individual functional connectivity profiles and how these signatures reflect behavior. The connectivity measures reflect underlying intrinsic connections that are modified only slightly with different task or resting-state conditions. A method for relating connectivity profiles to behavior, building a model, and then testing the predictive capabilities of the model will be shown. It will be shown that the areas that most characterize individual identification include frontal and parietal circuits. It will also be shown that specific connectivity patterns reflect measures of fluid intelligence and attention.

Special Seminar: November 25th at noon

Karen Holst

Doctoral Candidate
Karolinska Institutet
Challenges in free-breathing 3D cardiac magnetic resonance cine imaging

Abstract: Cardiac cine imaging is an important part of the clinical cardiac exam today, used for assessment of wall function and ventricular volume measurements. However, this is still mostly done with a stack of 2D images, each acquired during a breath-hold to avoid motion artifacts from the respiration. This method is both time consuming, inflexible and gives rise to many artifacts from poor or inconsistent breath holding. The desirable solution is a 3D free breathing technique which minimizes patient cooperation and gives high flexibility after acquisition for extracting arbitrary slice positions from the whole heart. This talk will focus on fast 3D k-space acquisition and reconstruction and the specific requirements trajectories need in order to cope with the constant motion of the beating heart and respiration. Furthermore, examples of different respiratory self-gating techniques will be shown and discussed. Finally, some preliminary results from our suggestion of combing a 3D acquisition technique and self-gating method will be given.

About the speaker: Karen Holst is a PhD student at Karolinska Institutet, Stockholm. She received her MSc in biomedical engineering at Technical University of Denmark where she specialized in medical imaging and radiation physics. Her thesis work is focused on free-breathing ventricular volumetric imaging resolved over both the cardiac and the respiratory cycles with magnetic resonance imaging.

Special Seminar: November 24th at noon

David Rigie

Doctoral Candidate
Medical Physics
University of Chicago
Acceleration Methods for Magnetic Resonance Imaging: Algorithms and Modelling

Abstract: Recent advances in x-ray computed tomography (CT) have led to a new imaging paradigm, called “spectral CT,” whereby a plurality of unique energy measurements are acquired, nearly simultaneously, in a single scan. It has been shown that this extra spectral information can be used to determine the entire energy dependence of the x-ray attenuation coefficient. This leads to the elimination of common image artifacts (e.g. beam hardening), a reduction in radiation dose, and improved quantification of contrast materials. In this talk, I will give a brief overview of spectral CT theory and clinical applications. Then, I will discuss how task based, mathematical observer models and image reconstruction algorithms can be generalized to accommodate this extra “spectral” dimension. The primary goal of my research is to improve the accuracy and robustness of spectral CT imaging by (1) developing objective metrics for optimizing imaging parameters and hardware design and (2) combining optimization based reconstruction methods with sparsity exploiting image priors, tailored to multispectral data. I will discuss some applications of these techniques to novel geometries with challenging data conditions.

About the speaker: David Rigie is a Ph.D. candidate in Medical Physics at the University of Chicago. Prior to arriving in Chicago, he studied Applied Physics at Cornell, with a concentration in molecular biophysics. His research interests include model-based image reconstruction, physical modelling, and spectral x-ray CT. He is currently involved in a collaboration with Toshiba Medical Research Institute, USA investigating the use of energy-resolving, photon-counting detectors for diagnostic CT.

October 27th at noon

Helen Benveniste, MD, PhD

Professor and Vice Chair for Research
Department of Anesthesiology
Stony Brook School of Medicine
Anesthesia-induced neurotoxicity investigated by MRI
October 20th at noon

Gwenn Smith, PhD

Richman Family Professor of Alzheimer’s and Related Diseases
Department of Psychiatry and Behavioral Sciences
Johns Hopkins University School of Medicine
Molecular Imaging of the Depression-Dementia Continuum

Abstract: Neuropsychiatric symptoms in late life are a major predictor of cognitive decline and the dementia transition. The pathophysiology underlying these symptoms is poorly understood. Over the past decade, advances in positron emission tomography (PET) instrumentation and radiotracer chemistry have provided an unprecedented opportunity to test mechanistic hypotheses generated from human post-mortem data and transgenic Alzheimer mouse models. Studies have been performed to identify the neural circuitry of affective and cognitive symptoms in late life depression and the role of the serotonin system. Building upon this work, multi-radiotracer PET imaging studies have been performed in late-life depression and mile cognitive impairment. These studies have tested the observation based on transgenic amyloid mouse models; of vulnerability of cortical monoamine projections (serotonin, to a greater extent) may precede beta-amyloid deposition. Understanding of the pathophysiology of late life depression and neuropsychiatric symptoms and targeting these symptoms may represent a strategy for earlier intervention and prevention.

Special Seminar: October 19th at noon

Matthew Muckley

Doctoral Candidate
Biomedical Engineering Department
University of Michigan
Acceleration Methods for Magnetic Resonance Imaging: Algorithms and Modelling

Abstract: In recent years there has been a growing interest in accelerating MRI scans, both from the viewpoint of reducing the computation time necessary to produce images as well as reducing the amount of time the patient spends in the scanner. This presentation will discuss both of these aspects of MRI acceleration. The first will be the development of a fast algorithm for the setting when parallel receive coils are used in conjunction with compressed sensing assumptions to reduce the scan time. This requires solving a complicated optimization problem, which can take more time than the scan itself. I will discuss an algorithm, BARISTA, that significantly reduces this computation time relative to state-of-the-art methods by carefully considering the structure of the sensitivity maps from the parallel receive coils. For the second aspect of MRI acceleration, I will discuss recent advances for the estimation of functional MRI time series of images using low rank modelling. The low rank modelling approach is demonstrated to be effective in simulation results relative to standard acquisition methods, and preliminary results using prospectively undersampled data will also be shown.

About the speaker: Matthew Muckley is a Ph.D. candidate in the biomedical engineering department at the University of Michigan. Matthew started his academic career by earning his B.S. at Purdue University, where he graduated With Distinction. Matthew's research at Michigan focuses on the application of signal processing methods to various imaging modalities, including MRI, X-Ray CT, optical imaging, and atomic force microscopy. For his research Matthew has been awarded first place in a KLA-Tencor Image Processing contest, a Rollin M. Gerstacker Foundation Fellowship, a GAANN Fellowship, and a Rackham Predoctoral Fellowship. and Aside from his research endeavors, Matthew also actively participates in the Biomedical Engineering Graduate Student Council at Michigan, for which he has served as President for the last year.

October 13th at noon

Joseph Ricker, PhD

Professor of Rehabilitation Medicine and Psychiatry
Director of Psychology, Rusk Rehabilitation
NYU Langone Medical Center
Acceleration Methods for Magnetic Resonance Imaging: Algorithms and Modelling

Abstract: Advances in neuroimaging have clearly changed the way that brain function, dysfunction, and rehabilitation may be conceptualized and studied. In addition to complementing existing behavioral and psychometric data, neuroimaging technologies such as fMRI and DTI novel inferences that cannot necessarily be made through other approach to studying brain-behavior relationships. This talk will provide an overview of contemporary functional neuroimaging techniques that have been specifically applied to traumatic brain injury in humans. The evidence-base (and often the lack thereof) of some technologies will be discussed in relation to the clinical appropriateness of these technologies. Finally, future research issues will be addressed through discussion of methodological and technical concerns rehabilitation, as well as how the integration of functional neuroimaging and clinical neuropsychology may inform the assessment and rehabilitation process.

About the speaker: Dr. Ricker is a board certified neuropsychologist and rehabilitation psychologist who came to NYU in 2013 as Professor of Rehabilitation Medicine and Director of Psychology for Rusk Rehabilitation at NYU Langone Medical Center. Prior to coming to NYU, he was a tenured Associate Professor and Vice Chair for Neuropsychology & Rehabilitation Psychology in the Department of Physical Medicine & Rehabilitation at the University of Pittsburgh School of Medicine. Dr. Ricker’s research career has been devoted to the study of cognitive impairment, recovery, and rehabilitation following human traumatic brain injury (TBI). He was among the very first investigators in the late 1990s to apply functional neuroimaging to investigate cognition after TBI. His work was honored in 2001 by two separate early career awards from the American Psychological Association, one in clinical neuropsychology and the other in rehabilitation psychology. Dr. Ricker’s current research focuses on the application of anatomic, functional, connectomic, and molecular brain imaging technologies in the investigation of neuropsychological impairment after brain injury. His NIH-funded research has included the use of technologies such as functional MRI, positron emission tomography, diffusion tensor imaging after TBI. He is a member of the editorial boards of four research journals (Journal of Clinical & Experimental Neuropsychology; Journal of Head Trauma Rehabilitation; Clinical Neuropsychologist; and, Rehabilitation Psychology), and serves as a grant reviewer for several U.S. and Canadian agencies, including the National Institutes of Health, the Centers for Disease Control and Injury Prevention, the Department of Veterans Affairs, and the Ontario Neurotrauma Foundation.

September 29th at noon

Thomas Benkert, PhD

Postdoctoral Researcher
Bernard & Irene Schwartz Center for Biomedical Imaging
NYU Langone Medical Center
Novel Steady-State Techniques for Magnetic Resonance Imaging

Abstract: Steady-state sequences are a class of rapid imaging techniques based on gradient-echo acquisitions with short repetition times. This class includes the balanced Steady-State Free Precession (bSSFP, TrueFISP) sequence, which provides the highest signal-to-noise ratio per unit time among all known imaging sequences. However, aside from a few applications such as cardiac imaging, this method is hardly established in the clinical routine. The main reasons are banding artifacts, which are signal voids due to magnetic field inhomogeneities, and the obtained T2/T1-weighted mixed contrast. In this talk, two novel techniques will be presented, which overcome these limitations and could allow for a more widespread use of bSSFP for MR diagnostics.

About the speaker: Dr. Benkert is a postdoctoral researcher at CBI working under the supervision of Dr. Block. During his undergraduate studies he studied to become a teacher for Physics and Mathematics in Wuerzburg, Germany and then focused on MRI, writing his thesis on “Quantification of Relaxation Times in MRI with Steady-State sequences”. During his PhD in Wuerzburg, he continued his MRI work under the supervision of Dr. Felix Breuer. The title of his dissertation was “Novel Steady-State Techniques for Magnetic Resonance Imaging”. His PhD work represents the topic of his Research Forum talk. His current research focuses on further developments for GRASP using fat-water separation.

Special Seminar: September 23rd at noon

Jill Slade, PhD

Department of Radiology
Michigan State University
The influence of age and physical activity on microvascular function

Abstract: The microvasculature is critical for the control of blood flow and tissue perfusion. Compromised microvascular function occurs during aging as well as several disease states and may contribute to compromised muscle performance within these populations. Muscle fMRI using blood-oxygen level-dependent imaging (BOLD) allow noninvasive assessment of peripheral microvascular function. Our findings show age related reductions in lower extremity BOLD and enhancement of muscle BOLD with exercise training in older adults.

September 18th at noon

Georg Schramm, PhD

Postdoctoral Fellow
KU Leuven
Initial results from simulations of joint PET/MRI reconstructions

Abstract: Regularized iterative image reconstruction is used in Positron Emission Tomography (PET) and Magnetic Resonance Imaging (MRI). In combined PET/MRI acquisitions, PET and MR images both suffer from artifacts due to acquisition time constraints. Since these artifacts are fundamentally different, we would like to investigate whether joint iterative reconstruction using joint prior information could improve the image quality in both modalities. In this presentation, Georg Schramm will give a short overview of initial results from simulations with different joint priors.

September 15th at noon

Khegai Oleksandr, PhD

Postdoctoral Fellow
Bernard & Irene Schwartz Center for Biomedical Imaging
NYU Langone Medical Center
Quantification methods for time-resolved metabolic magnetic resonance imaging using hyperpolarized [1-13C]pyruvate

Abstract:Dissolution dynamic nuclear polarization enables real-time non-invasive measurement of metabolic fluxes using magnetic resonance spectroscopy. Quantitative kinetic information of in vivo metabolism is of great interest for medicine as a key characteristic of some diseases, i.e. tumors. In this talk, the developed comprehensive methods for the data acquisition, quantification, interpretation and visualization of dynamic 13C metabolite signals in vitro and in vivo will be presented on the example of hyperpolarized [1-13C]pyruvate.

August 25th at noon

Marta Moreno, PhD

Postdoctoral Research Fellow
Department of Psychiatry
Columbia University
Studying depression with 7T MRI

Abstract:Functional magnetic resonance imaging/spectroscopy (fMRI/MRS) have been used to visualize abnormalities in unipolar depression (MDD) with mixed results. Patients with medication refractory depression (TRD) represent almost one-third of all patients with MDD. There are relatively few treatment options for these patients. Prefrontal repetitive transcranial magnetic stimulation (rTMS) is a non-invasive, well-tolerated alternative technique to pharmacological treatment for MDD. TMS induces stronger electric currents in superficial regions than in deeper structures. However, TMS can modify ongoing neuronal activity within complex neuronal circuits. Effects of TMS can propagate beyond the site of stimulation, impacting a distributed network of brain regions. These observations suggest that TMS may relieve depression by modulating synaptic strength both locally and at distant sites modulating functional connectivity in cortical networks. Some evidence for TMS as an antidepressant points toward cortical excitability increases to normalize abnormal level of activity and distributed modulation of brain activity resulting in network-specific release of neurotransmitters and activity modulation. However, it remains unclear how TMS targeted to Dorsolateral Prefrontal Cortex (DLPFC) exerts its antidepressant effect. The future of TMS relies on identifying its mechanisms of action across the brain. The combination of TMS and BOLD fMRI or MR spectroscopy at lower field strength have shown to be promising. However, the resolution of fMRI and MRSI at lower fields are too low for depiction of node size and temporal resolution. These shortcomings of low field MRI prevents detection of default mode networks (DMN) function with appropriate representation of the strength of FC between the nodes. We have used 7T functional connectivity and MRSI at 0.5cc resolution to demonstrate correlation between glutamate concentration and DMN function. We will discuss how 7T is crucial in visualizing DMN in various brain regions implicated in MDD. A discussion will be presented of technical challenges in realizing 7T advantages in order to use the combined fMRI/MRSI in search for faulty networks. Eliminating 7T RF coil and B0 inhomogeneity in skull base will allow comparing fMRI/MRSI of normal subjects and patients with MDD which, in turn, could make diagnosis and treatment of these patients a quantitative practice. Such tools will reveal further understanding into the impact of TMS on brain function.

Special Seminar: August 4th at noon

Chantal Tax

PhD Candidate
University Medical Center Utrecht
From acquisition to tractography: Some recent advances in diffusion MRI data processing

Abstract:Diffusion MRI (dMRI) has offered exciting new avenues for investigating microstructural and architectural characteristics of tissue in vivo. The growing interest for integrating dMRI in many clinical and scientific studies has triggered the development of different strategies to process dMRI data. These developments include, amongst others, modeling and reconstruction of the dMRI signal beyond diffusion tensor imaging (DTI), and new ways to extract information on the (local) geometry of fiber tractography streamlines. This presentation will focus on our recent work in this area, including robust fitting in diffusion kurtosis imaging (DKI), calibrating the response function for spherical deconvolution (SD), the acquisition of a reference dataset to test processing pipelines, and quantifying whether streamlines locally form a grid-like pattern.

Special Seminar: August 4th at 3pm

Maxime Chamberland

PhD Candidate
Universite de Sherbrooke
Navigating through brain connectivity in real-time: A neurosurgical perspective

Abstract:In the past decade, the fusion between diffusion magnetic resonance imaging (dMRI) and functional magnetic resonance imaging (fMRI) has opened the way for exploring structure-function relationships in-vivo. As it stands, the common approach usually consists of analysing fMRI and dMRI datasets separately or using one to inform the other, such as using fMRI activation sites to reconstruct dMRI streamlines that interconnect them. Also, given the large inter-individual variability of the healthy human brain, it is possible that valuable information is lost when a fixed set of dMRI/fMRI analysis parameters such as threshold values are assumed constant across subjects. By allowing one to modify such parameters while viewing the results in real-time, one can begin to fully explore the sensitivity of structure-function relations and how they differ across brain areas and individuals. This is especially important when interpreting how structure-function relationships are altered in patients with neurological disorders, such as the presence of a tumor. In this study, we present and validate a novel approach to achieve such visualization: First, we present an interactive method to generate and visualize tractography-driven resting-state functional connectivity. Next, we demonstrate how our proposed approach can be used in a neurosurgical planning context. We believe this approach will promote the exploration of structure-function relationships in a subject-specific aspect and will open new opportunities for connectomics.

July 28th at noon

Jakob Asslander, PhD

Research Assistant
Department of Radiology
University Medical Center Freiburg
Spin Echoes in the Regime of Weak Dephasing: From SE-FLASH to MR-Fingerprinting

Abstract: In this talk I will demonstrate the possibility to form spin echoes after a single excitation pulse, where the time between the end of the pulse and the echo is longer than the length of the pulse itself. This stands in contrast to Hahn's theory spin echoes, where the length of a composite pulse is at least equal to the time between the end of the pulse and the echo. A representative spin echo pulse is implemented in an inversion recovery SNAPSHOT-FLASH sequence in order to retrieve quantitative T1- and proton density maps of the lung with increased signal intensity. Last but not least the theoretical concept in translated to a pseudo steady state free precession sequence for MR-fingerprinting.

About the Speaker: Jakob Asslander studied physics in Würzburg (Germany), where he began conducting MRI research. Thereafter, he obtained a PhD under Jürgen Henning in Freiburg (Germany). During doctoral study, Dr. Asslander focused on fast fMRI acquisition techniques with reduced susceptibility to artifacts. More recently, he has pursued research in RF-pulse design and optimal control algorithms.

July 27th at 3:00pm

Sung-Hong Park, PhD

Assistant Professor
Department of Bio and Brain Engineering
Korea Advanced Institute of Science and Technology (KAIST), Daejeon, South Korea
Development of New Anatomical, Physiological, and Functional Magnetic Resonance Imaging Techniques

Abstract: Magnetic resonance imaging (MRI) provides anatomical, physiological, and functional information of our body noninvasively. In this seminar, some new approaches to these imaging modalities will be introduced. The new approaches include techniques for (i) acquisition of time-of-flight MR angiogram and blood oxygenation level dependent (BOLD) MR venogram (often called susceptibility-weighted Imaging, SWI) simultaneously with minimal impacts on the image quality to each other, (ii) imaging blood perfusion and magnetization transfer (MT) asymmetry simultaneously with interslice blood flow and MT effects in 2D sequential multi-slice imaging, and (iii) better understanding of signal sources of high-resolution balanced steady-state free precession functional MRI at high field. Applications of compressed sensing algorithms to these imaging methods will be introduced. Images from humans and animals at various field strengths (3T - 9.4T) will be demonstrated and potential applications of these imaging methods for clinical diagnosis will be discussed.

About the Speaker: Sung-Hong Park received a PhD in bioengineering from the University of Pittsburgh in 2009

July 24, 2015 at noon

Pamela Woodard, PhD

Professor of Radiology and Biomedical Engineering
Washington University
Targeted Molecular Imaging of Atherosclerosis or "Tales in Translation"

Abstract: The talk focuses on PET-MR imaging of atherosclerosis using novel radiopharmaceuticals targeted to specific components suggestive of plaque vulnerability including Natriuretic Peptide Receptor-C and hypoxic macrophages. The speaker will go describe the steps of taking one of these radiotracers from preclinical development and toxicity testing to FDA IND application, approval and clinical trial.

About the Speaker: Pamela Woodard is a Professor of Radiology and Biomedical Engineering at Washington University where she is Radiology Vice Chair of Clinical and Translational Research, Director of the Center for Clinical Imaging Research (CCIR) and Head of Advanced Cardiac Imaging. She is past chair of the Cardiovascular Radiology and Intervention (CVRI) Council and of the American Heart Association and is currently a member of the AHA Operations Committee and past president of the North American Society for Cardiovascular Imaging. She has over 140 peer-reviewed publications and has been PI, on the steering committee or co-investigator on multiple NIH-funded grants and clinical trials.

July 21, 2015 at noon

Tuomo Valkonen, PhD

Research Associate
University of Cambridge
Approaches for dealing with the non-linearity of the Stejskal-Tanner equation

Abstract: The Stejskal-Tanner equation for diffusion tensor imaging (DTI) produces a non-linear correspondence between the DWI measurements and the tensor field. For correct noise modelling, we should in principle include this non-linearity in our DTI reconstruction models. This results in a difficult non-convex optimisation problem. In this talk, I will discuss a primal-dual optimisation method that can effectively handle the Stejskal-Tanner equation—at least when no other complications enter the reconstruction model. In practise, however, our knowledge of the measurement errors and noise is only partial, and accurate noise modelling is not feasible. I will therefore look at the efficacy of foregoing accurate noise modelling, and reducing our knowledge of measurement and model errors and noise to simple bounds, effectively obliterating the Stejskal-Tanner equation from the model (joint work with Yury Korolev and Artur Gorokh).

About the Speaker: Tuomo Valkonen received his Ph.D in scientific computing from the University of Jyväskylä (Finland) in 2008. He has since then worked in well-known research groups in Graz, Cambridge and Quito. Currently in Cambridge, his research concentrates on the mathematical analysis of image processing models, towards studying their reliability, and the development of fast optimisation algorithms for the solution of these models.

April 21, 2015 at noon

Elias Kellner, PhD

Medical Physics
Department of Radiology
University Medical Center Freiburg
Quantitative Cerebral Blood Flow Measurements using Dynamic-Susceptibility-Contrast-MRI: Development of a Measurement Sequence and Comparison with PET

Abstract: Dynamic Susceptibility-Contrast MRI can be used to measure the cerebral blood flow in the brain. The method has successfully been applied in clinical routine for over a decade, particularly in Stroke, but it is currently not exploting its full potential due to several problems concerning the correct quantification. The major problem is related to the measurement of the arterial input function (AIF). The key weakness of the existing, conventional technique is an insufficient consideration of the different physical effects of paramagnetic contrast agent in large blood vessels, and in tissue. In this work, these effects are thoroughly analysed to design an extended measurement sequence with an additional module dedicated to the correct measurement of the blood signal. With this, the AIF can accurately and quantitatively be determined. In a comparison study in the porcine model, the proposed technique is validated against the current gold standard, positron-emission-tomography (PET). This quantitative comparison can for the first time be performed without additional normalisation factors. The results demonstrate a good agreement of both methods. The comparison further reveals that the reasonable interpretation of calulcated maps for both, MRI and PET is not straightforward, and requires consideration of the corresponding kinetic models as well as the physics of the tracers used in the different methods.

About the Speaker: Dr. Kellner's profile at Universtats Klinikum Freiburg.

March 31st, 2015 at noon

Rama Jayasundar, PhD

Department of NMR
All India Institute of Medical Sciences
New Delhi, India
Applications of NMR in the Indian traditional medicine of ayurveda

Abstract: The growing interest in systems perspective is not only revolutionising cell biology but also providing the impetus for clinical medicine to shift from a reductionistic to a holistic approach for efficient disease management. This inevitably brings into focus one of the longest unbroken healthcare system in the world, i.e. ayurveda, indigenous to Indian subcontinent. The unique ability of NMR to study whole systems (in vitro and in vivo) and generate a wide range of information non-invasively makes it ideally suited to study holistic medicine like ayurveda. It offers a powerful non-invasive means to not only validate ayurveda but also to gain understanding of its concepts and translate them for use in modern healthcare. Different areas ranging from ayurveda’s therapeutic use of medicinal plants to diagnosis, treatment efficacy and concepts of preventive healthcare can be studied and validated effectively through NMR, opening new vistas for expanding the role of NMR in healthcare. This presentation, while outlining the various potential applications of MR in ayurveda will also elaborate on the systems approach of ayurveda.

About the Speaker: Dr. Rama Jayasundar, after her initial training in Physics, obtained her PhD in NMR from Cambridge University, UK. In addition to her main training as a physicist, she is also a qualified doctor trained in both ayurveda (the indigenous Indian medical system) and modern medicine. She holds a Bachelor’s degree in Ayurvedic Medicine (BAMS - Bachelor of Ayurvedic Medicine and Surgery). She is currently a faculty in the Department of NMR, All India Institute of Medical Sciences (AIIMS), New Delhi, India. Her area of specialization is Biomedical MR - RF coil designing and building, RF pulse sequence programming, clinical imaging and spectroscopy. She developed indigenously a low cost MR coil for clinical use, for which she received the Young Scientist Award. During her stint as a visiting Professor at the Max Planck Institute of Biophysical Chemistry, Gottingen, Germany (1997-1998), she worked on the development of functional MR spectroscopy techniques. She has authored a number of research publications in peer reviewed journals and has also won many awards and honors. Using her dual qualification as an NMR scientist and a professionally qualified ayurvedic doctor, she is currently involved in scientific research in Ayurveda. Her research interests range from applications of NMR, MRI and other analytical techniques in basic and clinical ayurvedic research.

March 30th, 2015 at noon

Dr. Elmar Merkle

Professor and Chairman
Department of Radiology
University Hospital Basel
A hard look at MR: is it simple and fast enough to fill the gap?

Abstract:This talk was initially presented as a plenary during the annual ISMRM meeting in Montreal in 2011. It focuses on the lack of speed and simplicity as well as the lack of robustness of MR imaging in comparison to other cross sectional imaging modalities. Now, almost 4 years later, this talk will be repeated in its original form to challenge NYU’s research group to answer the simple question – what has changed since?

About the Speaker: Dr. Merkle's profile at Universitatsspital Basel.

March 24th, 2014 at 11:00am

Dmitri “Mitya” Chklovskii, PhD

Group Leader for Neuroscience
Simons Center for Data Analysis
Simons Foundation, New York City
How do animals see motion? Insights from fly connectomics.

Abstract: Animal behaviour arises from computations in neuronal circuits, but our understanding of these computations has been frustrated by the lack of detailed synaptic connection maps, or connectomes. For example, despite intensive investigations over half a century, the neuronal implementation of local motion detection in the visual system remains elusive. By developing a semi-automated pipeline using electron microscopy we were able to reconstruct the biggest connectome to-date within the Drosophila visual system and identify neurons and synapses comprising the motion detection circuit motif. Electrophysiological recordings from the identified neurons have confirmed our predictions. More recently, a similar motif has been identified in the vertebrate retina suggesting that the principles of neural computation are shared across species.  

About the Speaker: Before coming to the Simons Foundation in 2014, Mitya Chklovskii was a group leader at the Howard Hughes Medical Institute’s (HHMI) Janelia Farm Research Campus in Ashburn, Virginia. Chklovskii also initiated and led a collaborative project at HHMI that assembled the largest-ever connectome, a comprehensive map of neural connections in the brain. Before that, he worked at Cold Spring Harbor Laboratory in New York, where he founded the first theoretical neuroscience group, having worked there as a first assistant, and later an associate professor. As group leader for neuroscience, Chklovskii leads an effort to understand how the brain analyzes complex datasets streamed by sensory organs, in an attempt to create artificial neural systems. He holds a Ph.D. in physics from the Massachusetts Institute of Technology.

August 26th, 2014 at 12:00pm

Lior Weizman, PhD

Postdoctoral Fellow
Department of Electrical Engineering
Technion – Israel Institute of Technology, Haifa
The application of compressed sensing for longitudinal MRI

Abstract: Magnetic Resonance Imaging (MRI) is the method of choice for diagnosis, evaluation and follow-up of brain pathologies. In the common treatment scheme, patients are repeatedly scanned every few weeks or months to assess disease progression and treatment response. Although the important information for clinical evaluation lies in the change between the follow-up MRI and the former one, every follow-up scan is acquired anew. This makes most of the data in the later scan redundant. In MRI, data is acquired in a spatial frequency domain, called "k-space". In my talk I'll discuss the application of compressed sensing (CS) for MRI and the mutual similarity of follow-up scans in longitudinal MRI studies. I'll present a sampling and reconstruction framework that exploits the redundancy of the acquired data in longitudinal studies. This would rely on two extensions of compressed sensing, adaptive-CS and weighted-CS. In adaptive CS, k-space sampling locations are optimized such that the acquired data is focused on the change between the follow-up MRI and the former one. Weighted CS uses the locations of the nonzero coefficients in the sparse domains as a prior in the recovery process.Results are presented on MRI scans of patients with brain tumors, and demonstrate improved spatial resolution and accelerated acquisition for 2D and 3D brain imaging at 10-fold k-space undersampling.

About the Speaker: Lior Weizman received the B.Sc. and M.Sc. degrees in Electrical Engineering from Ben-Gurion University of the Negev, Beer-Sheva, Israel, in 2002 and 2004, respectively, and the Ph.D. degree in computer science in 2013 from the Hebrew University of Jerusalem, Israel. From 2005 through 2008 he was with RAFAEL, Advanced Defense Systems LTD. During 2011 he was a visiting student at Stanford University, CA. He is currently a post-doctoral fellow at the Department of Electrical Engineering, Technion - Israel Institute of Technology, Haifa. His research interests are in the general areas of sampling theory, statistical signal processing and their applications to medical image processing and medical imaging.

August 25th, 2014 at 12:00pm

Afonso Silva, PhD

Chief, Cerebral Microcirculation Unit
Laboratory of Functional and Molecular Imaging
National Institute of Neurological Disorders and Stroke
National Institutes of Health
Anatomical and Functional MRI of Conscious, Awake Marmosets
August 20th, 2014 at 12:00pm

James M. Balter PhD

Department of Radiation Oncology and Biomedical Engineering
University of Michigan Medical School
Integration of MRI in the Radiation therapy environment

Yue Cao, PhD

Department of Radiation Oncology, Radiology and Biomedical Engineering
University of Michigan Medical School

MRI-based perfusion measurements for assessment and individualization of patients undergoing radiation therapy for intrahepatic cancer
August 19th, 2014 at 12:00pm

Eric T. Ahrens, PhD

Professor and Director of Stem Cell Molecular Imaging
Department of Radiology
University of California, San Diego
MRI-based approaches to quantitatively study cell trafficking and function in vivo

Abstract:An unmet challenge to successful clinical development of stem cell therapies is the development of non-invasive methods to image the behavior and movement of cells following transplant into patients. Moreover, imaging is needed to improve safety surveillance of cell therapies to help overcome regulatory hurdles. MRI is experiencing a rapid expansion in its ability to visualize specific cell populations in vivo. These capabilities are facilitated by the development of new imaging probes that tag cells prior to transfer or alter a cell’s proteome to facilitate MRI detection. This talk will first cover a new approach for cell tracking developed in our lab called ‘in vivo cytometry.’ In this approach, cell populations of interest, such as stem cells, are tracked and quantified in vivo. We formulate novel perfluorocarbon (PFC) emulsions to label cells ex vivo. The labeled cells are then introduced into the subject and their migration can be monitored using fluorine-19 (19F) MRI. The 19F images are extremely selective for the labeled cells, with no background signal from the host’s tissues. Moreover, the absolute number of labeled cells in regions of interest can be estimated directly from the in vivo 19F images. Additionally, the PFC emulsion reagents have bio-sensing properties that report on the absolute level of intracellular oxygen and can potentially be used to monitor cell differentiation or apoptosis in vivo. Looking ahead, MRI will be able to harvest the power of molecular biological tools to impart exogenous image contrast to living tissue in a cell-specific or event-related manner. This will be accomplished using transgenic and vector technologies to express reporter genes coding for paramagnetic metalloproteins. Towards this goal, I will describe efforts to develop and characterize new generations of nucleic-acid based MRI reporters that render cells paramagnetic and detectable in vivo. For example, MRI reporters can be used for labeling stem cells for long-term tracking in vivo.

About the Speaker:Eric T. Ahrens, Ph.D., is a Professor and Director of Stem Cell Molecular Imaging in the Department of Radiology at the University of California, San Diego. Formally, he was a Professor of Biological Sciences at Carnegie Mellon University and the Director of the Pittsburgh NMR Center for Biomedical Research. Prior to this, he served as a Senior Research Fellow in the Department of Biology at the California Institute of Technology. He holds a Ph.D. in physics from the University of California at Los Angeles and was a graduate fellow at Los Alamos National Laboratory. Ahrens’ research investigates in vivo biological processes using unique molecular, cellular and anatomical MRI and NMR methods.

August 12th, 2014 at 12:00pm

Charles Watson, PhD

Siemens Healthcare Molecular Imaging
A Sparse Transmission Method for PET Attenuation Correction in the Head

Abstract: In positron emission tomography (PET), attenuation of the annihilation radiation in the body is the largest physical effect confounding the quantitative interpretation of the emission data. Traditional g-ray transmission (TX) measurements for attenuation correction in clinical PET were largely abandoned 12 years ago with the advent of PET/CT. Recently, however, several technological developments have converged to significantly enhance the power of TX measurements, sparking renewed interest. These include new joint reconstruction algorithms for simultaneously acquired emission and transmission data; time-of-flight (TOF) measurement capability for discriminating attenuation effects in emission data; and the positron beam technique for injecting transmission sources into the field of view of integrated PET/MR systems. In this talk we will describe a novel solution for PET attenuation correction in the head based on the joint reconstruction of simultaneously acquired emission and sparse transmission (sTX) data corresponding to 20 fixed line sources placed in a ring around the head. Simulations of an 18FDG study show that the sTX data effectively constrain cross-talk. Bone, soft tissue and voids are approximately represented in the estimated attenuation image. The results are compared to a standard MLEM reconstruction of emission-only data, and to joint reconstruction of simultaneous emission-transmission data using a full-ring source. We find that 10 to 20% underestimation of activity in the peripheral regions of the brain in the latter two images is reduced to < 5% on average in the sTX case. We thus demonstrate that an sTX array can provide better cross-talk reduction than a conventional full-ring transmission source, and will offer a qualitative explanation of why this occurs. We will also examine the impact of TOF information on the joint reconstruction in the noise-free case. We estimate that such an sTX technique would increase patient radiation dose in a typical 18FDG clinical study by < 4%.

About the Speaker:Dr Watson earned a PhD in physics from Yale University in 1980. Following post-doctoral study at the California Institute of Technology in planetary science, he joined the corporate research staff of Schlumberger in Ridgefield, Connecticut in 1982, where he developed Monte Carlo simulations of g-ray and neutron transport for the design and interpretation of nuclear well-logging instruments. In 1993, he joined CTI PET Systems in Knoxville, Tennessee, which subsequently merged into Siemens Healthcare. At CTI/Siemens Dr Watson has been involved in nearly all aspects of the physics of PET, PET/CT and PET/MR scanners. He is the author of a widely used 3D scatter simulation algorithm for the correction of positron emission data. From 1999 to 2002 he was the project leader for the development of the first commercial PET/CT scanner. He served as the chief PET physicist for the development of the first integrated whole-body PET/MR, Siemens’ mMR. His current research interests include applications of positron beams in PET/MR systems, and the development of next-generation transmission systems for the attenuation correction of PET data. He is the author of numerous scientific publications and patents in the field of PET instrumentation, and serves on the editorial board of EJNMMI-Physics.

August 11th, 2014 at 10:00am

Chunlei Liu, PhD

Assistant Professor of Radiology
Durham, NC
Quantitative Susceptibility Mapping and Susceptibility Tensor Imaging

When the brain is situated in a magnetic field, it creates a small field of its own in response to the presence of the external field. This interaction, though extremely weak, becomes measurable under the strong field provided by MRI scanners. With MRI, this small perturbation field can also be spatially localized and quantified. The strength and direction of the perturbation is influenced by a number of physiologically important factors including molecular composition, cellular organization and neuronal connectivity. By imaging this field perturbation, one may then be able to infer a wealth of information about brain microstructure. Such information include, for example, iron deposit in aging and Parkinson’s disease, myelination in brain development, demyelination in multiple sclerosis, and neuronal connectivity. Besides brain, this magnetic interaction is also significant in many other organs including kidney and heart. I will present some recent methodological developments and discuss potential applications.

August 4th, 2014 at 9:30am

Colin Studholme, PhD

Professor Pediatrics and Bioengineering
Adjunct Professor of Radiology
Department of Pediatrics, University of Washington, Seattle, WA
MR Imaging the Moving, Growing Human Fetal Brain

Recent work that combines computer vision with fast MR imaging techniques is beginning to allow the collection of full 3D MRI scans of the human fetal brain in-utero without sedation. The basic ideas behind the engineering approaches to these techniques will be reviewed with examples on typical clinical structural and diffusion imaging studies. Results of the application of these imaging techniques to study human fetal brain development by constructing spatio-temporal growth models will then be covered.

About the speaker: Dr. Studholme is a Professor of Pediatrics and Bioengineering, and Adjunct Professor of Radiology at University of Washington, Seattle. He completed his Ph.D. in medical physics and biophysics from the University of London and a postdoctoral fellowship in diagnostic radiology at Yale University. Dr. Studholme’s research focuses on the development of new mathematical and computational algorithms to manipulate and analyze biomedical image data. His work is currently motivated by the study of brain anatomy and the patterns of its change over time in two broad clinical areas: fetal and pre-term infant brain development, and neurodegenerative processes in adults.

July 29th, 2014 at 12:00pm

Joseph Alukal, M.D..

Assistant Professor; Dir Reproductive Health & Benign Disorders of Prostate
Departments of Obstetrics and Gynecology (Obs/Gyn) and Urology (Urology)
NYU Urology Associates
Management of non-obstructive azoospermia: is there a role for imaging?

For information on Timothy Duong's current research, please click here.

July 25th, 2014 at 12:00pm

Timothy Q. Duong, Ph.D.

SI Glickman MD Endowed Chair, Professor
MRI Division Chief, RII
Assistant Director for Research, RII
University of Texas Health Science Center
San Antonio, TX
MRI of experimental stroke and TBI

For information on Timothy Duong's current research, please click here.

July 22nd, 2014 at 12:00pm

Paul Vaska, PhD

Department of Biomedical Engineering
Stony Brook University
Novel PET and multimodal imaging technologies, from neuroscience to oncology

Our research encompasses the development of new detector materials and concepts, low-noise microelectronic signal processing, high-throughput data acquisition methods, Monte Carlo simulation, and new data processing techniques to optimize the extraction of quantitative information from the PET data. This talk will present an overview of our unique imaging technologies - RatCAP, small-animal PET-MRI, human breast PET-MRI, wrist scanner for input function, and future human brain imagers.

July 16th, 2014 at 2:00pm

Theresa Bachschmidt

PhD Candidate
Siemens MR Erlangen, MSK Team
Metal artifact correction using pTX
July 15th, 2014 at 12:00pm

Ricardo Otazo, PhD

Assistant Professor Radiology
New York University School of Medicine
Stretching the limits of compressed sensing dynamic MRI: Low-rank plus sparse reconstruction and self discovery of motion

Abstract: Extensive spatiotemporal correlations in dynamic MRI enable the application of compressed sensing techniques to accelerate data acquisition. Low-rank plus sparse (L+S) matrix decomposition or robust principal component analysis (RPCA) can be employed to represent dynamic images as a superposition of a background component (L) and a dynamic component (S). The dynamic component can include for example organ motion or contrast-enhancement information. The L+S model increases the compressibility of dynamic images with respect to L- or S-only models and performs automatic background suppression in the S component. This talk will describe how the L+S model can be employed to reconstruct undersampled dynamic MRI data with automatic separation of background and dynamic components. An extension of the L+S approach that incorporates a motion model to improve the performance in the presence of organ motion will be also discussed. Reconstruction of highly-accelerated dynamic MRI data corresponding to cardiac perfusion, cardiac cine, time-resolved peripheral angiography, and abdominal perfusion using Cartesian and golden-angle radial sampling will be presented to show feasibility and general applicability of the L+S method.

Bio: Ricardo Otazo is an Assistant Professor of Radiology at New York University School of Medicine. He received his B.Sc. in Electronics Engineering from Universidad Catolica de Asuncion, Paraguay in 2001, and his M.Sc. and Ph.D. in Electrical Engineering from the University of New Mexico in 2005 and 2007 respectively. His research interests include the development of rapid MRI and low-dose CT techniques using compressed sensing, image reconstruction algorithms, application of MRI techniques to clinical studies and signal processing methods in general.

June 25th, 2014 at 2:00pm

Rafael O’ Halloran, PhD

Assistant Professor Radiology
Mount Sinai Hospital
Diffusion-Weighted MRI in the (Not-So-) Steady State

Abstract: Diffusion-weighted (DW) MRI is mostly done with single-shot EPI because multi-shot DW MRI is sensitive to motion-induced phase. Solving the multi-shot problem would open up DW MRI to other pulse sequences, allowing gains in resolution and geometric fidelity. In this talk we’ll look at some solutions to this problem in the context of a 3D (massively multi-shot) DW steady-state free precession sequence.

Bio: Rafael O’Halloran started his career in MRI at the University of Wisconsin in Madison where he worked in Sean Fain’s group on hyper-polarized Helium-3 MRI of the lung, fast radial imaging, and diffusion. After graduation, he moved to sunny Stanford, California to work with Roland Bammer on DW MRI of the brain with steady state free precession sequences. In January of this year Rafael joined Mount Sinai as Assistant Professor of Radiology and continues to work on diffusion and other interesting contrasts in the brain. He lives with his wife and 22-month-old son on the upper East side and is slowly adjusting to life in New York.

June 24th, 2014 at 12:00pm

Jiangyang Zhang, PhD

Associate Professor
Johns Hopkins University
Imaging Brain Structures and Injuries with Oscillating Gradient Diffusion MRI

Diffusion MRI utilizes water molecule diffusion to probe brain microstructures and is an important tool to visualize a wide spectrum of pathologies. In recent years, a special diffusion MRI technique, the oscillating gradient diffusion MRI, has shown promise in providing additional information on tissue microstructures. Our research focus on using oscillating gradient diffusion MRI to bring novel imaging contrasts to visualize structures and pathology in the brain. We have demonstrated that the technique can reveal densely packed neuronal layers in the mouse hippocampus and cerebellum. In a mouse model of neonatal hypoxia ischemia, our results suggest that the technique can detect swelling of glial cells and their processes at 24 hours after insult.

June 20th, 2014 at 12:00pm

Jack Poulson, PhD

Assistant Professor
Department of Computational Science and Engineering
Georgia Institute of Technology
High-performance Low-rank Plus Sparse Matrix Decompositions for Undersampled Dynamic MRI

Low-rank plus Sparse matrix decompositions were recently proposed (by Otazo et al.) as a means of separating the background and dynamic components of undersampled dynamic MRI. For typical model sizes, a sequential plane-by-plane 4D reconstruction using an Alternating Direction Method of Multipliers (ADMM) requires a few hours of computation, most of which is spent within 2D non-uniform Fourier transforms (NUFTs) and the proximal map for the nuclear norm, so-called singular-value soft-thresholding (SVT).Due to the structure of the acquisition operator, within each plane, the NUFTs are embarrassingly parallel over both the channels and timesteps, whereas the SVT primarily consists of a QR decomposition of a tall-skinny matrix and is best decomposed within the image domain. It is shown that, with a careful redistribution of the data at each iteration, both the NUFTs and SVTs can be effectively parallelized on thousands of cores and reconstruction times have been observed to reduce from several hours to roughly one minute. Furthermore, the preliminary implementation is made available as part of an open source package, currently named Real Time Low-Rank Plus Sparse MRI (RT-LPS-MRI).

Jack Poulson is an Assistant Professor in the Department of Computational Science and Engineering at the Georgia Institute of Technology. Jack completed his PhD in Computational and Applied Mathematics at UT Austin at the end of 2012 and spent a brief postdoc in Stanford's Department of Mathematics before moving to Georgia Tech in November of 2013.

June 11th, 2014 at 12:00pm

Shalom Michaeli, PhD

Center for Magnetic Resonance Research
Department of Radiology
University of Minnesota
"Probing biological systems using frequency swept pulses: relaxation in high rank rotating frames, n ≥ 2."

NMR offers a plethora of tools for investigating tissue properties in vivo. The present presentation aims to describe novel MRI and MRS approaches that have been recently developed in our laboratory, based on the implementation of frequency swept (FS) pulses operating in adiabatic and non-adiabatic regimes. The tissue contrasts generated by such techniques will be explained within the context of rotating frame relaxation mechanisms, magnetization transfer effects [6], relaxation along a fictitious field (RAFF) in the rotating frames of rank n ≥2 (RAFFn), and MRI with RAFFn preparations using no echo time SWIFT readout. Frequency swept pulses offer unique capabilities to investigate biological systems for both in vivo and high-resolution NMR. Applications to glioma gene therapy, Parkinson diseases and multiple sclerosis, and to quantification of protein dynamics will be presented.

June 10th, 2014 at 1:00pm

Silvia Mangia, PhD

Assistant Professor
Department of Radiology
University of Minnesota
"Functional 1H MRS at ultra-high field"

Proton magnetic resonance spectroscopy (1H MRS) allows the non-invasive measurement of metabolite concentrations, and is a powerful tool to investigate brain biochemistry and metabolism in health and disease. Similar to MRI, 1H MRS benefits from the gain in signal-to-noise ratio which originates in the increased polarization at higher magnetic field. High magnetic fields also increase chemical shift dispersion, thus emphasizing the characteristic spectral patterns of metabolites and decreasing spectral overlaps. Greater spectral dispersion additionally improves water suppression and spectral editing. The improved sensitivity achieved at high magnetic fields ultimately results in gains in spatial resolution, temporal resolution and/or reliability of quantification of an increased number of metabolites. Magnetic fields higher than 4 T are widely employed in 1H MRS studies of animal models. Recent progresses in magnet technology, gradient system performance, RF coil and pulse sequence design enabled localized in vivo 1H MRS also in humans at ultra-high magnetic fields up to 7 T. Exciting applications of 1H MRS in humans involve the functional studies of the metabolic events occurring during various stimuli. Our group (Mangia et al, 2007a; 2007b) and others (Lin et al, 2012; Schaller et al, 2013) have measured the concentrations of multiple metabolites with unprecedented sensitivity and temporal resolution at 7 T in the human primary visual cortex during paradigms of visual stimulation. These studies provided critical insights into the metabolic events of increased neuronal activity, and shed light into the neurometabolic coupling of astrocytes and neurons.

May 6th, 2014 at 12:00pm

Steven Baete, Ph.D.

Post-doctoral fellow
New York University School Of Medicine
Langone Medical Center
"Improved Diffusion Spectrum Imaging by Radial q-space sampling using a multi-echo stimulated echo diffusion sequence"

Diffusion Spectrum Imaging (DSI) is a powerful means for robustly and non-invasively imaging long-range neuronal architecture in the human brain. This robustness is rooted in DSIs model-independent determination of the Orientation Distribution Function (ODF) through the sampling of the ODFs Fourier transform in q-space. The large number of q-space samples needed for accurate measurements of the ODF lead to long acquisition times, hindering practical implementation. These long acquisition times can be partially mitigated by multi-slice or multiband techniques where several slices are encoded at the same time. A second hindrance is that practically feasible b-values (e.g. 4000 s/mm2) limit the achievable angular resolution as the angular resolution is proportional to the inverse of the largest distance sampled in q-space when sampling q-space on a Cartesian grid.
In this talk we will show that these limitations to the practical implementation of DSI can be overcome by radially sampling q-space (RDSI) using a multi-echo stimulated echo diffusion sequence. When sampling q-space along radial lines, each radial line in q-space is directly connected by the Fourier slice theorem to the value of the radial ODF at the same angular location. This has the advantage that the angular resolution depends on the number of radial lines sampled rather than on the maximum b-value. Hence, Radial q-space sampling for DSI results in an improved angular resolution at lower b-values compared to Cartesian q-space sampling for a similar number of samples. In addition, the radial sampling lends itself to using a multiple echo stimulated echo diffusion sequence, accelerating the acquisition almost fourfold. The higher diffusion times of the stimulated echoes are also expected to lead to increased anisotropy and better fiber tracking. The findings which will be presented in this talk suggest that radial acquisition of q-space can be favorable for the practical implementation of DSI.

April 29th, 2014 at 12:00pm

Florian Knoll
Postdoctoral Researcher
NYU Langone Medical Center

"Joint Reconstruction of MR-PET data with multi-sensor compressed sensing"

Integrated MR-PET systems like the Siemens Biograph mMR allow simultaneous acquisition of PET and MR data. However, image reconstruction is performed separately and results are only combined at the visualization stage. PET images are reconstructed using a variant of Expectation Maximization while MR data are reconstructed with an inverse Fourier transform or iterative algorithms for parallel imaging or compressed sensing. We propose an integrated joint reconstruction framework based on multi-sensor compressed sensing. This approach uses MR and PET data simultaneously during image reconstruction and exploits anatomical correlations between the two modalities. Results will be shown for numerical simulations and in-vivo imaging that demonstrate improvements in image quality of both MR and PET images. We expect that joint reconstruction can provide additional enhancements to the information content of multimodality studies in the future.

April 25th, 2014 at 3:00pm

Beatriz Luna, PhD
Staunton Professor of Psychiatry and Pediatrics
University of Pittsburgh

"Functional Specificity and Integration of Brain Processes underlying Cognitive Development"

The adolescent period incurs vulnerabilities that undermine survival (risk-taking behaviors) and importantly increase the risk for the emergence of psychopathology. These vulnerabilities have been associated with a protracted maturation of prefrontal executive and striatal motivational systems. The contribution of each of these systems and importantly systems-level processing to cognitive development are not well understood. I will present a set of fMRI and DTI studies that identify developmental changes in functional specificity including prefrontal systems underlying inhibitory control and striatal neurophysiology and function in reward processing. In addition, studies characterizing changes in functional and structural connectivity through adolescence will be discussed. Together, these findings indicate that adolescents have access to executive systems supporting decision-making but in the context of a reactive motivational system underlied by an established though specializing network connectivity.

April 22nd, 2014 at 12:00pm

Yiğitcan Eryaman, PhD
Associate Prof. Neurobiology
Post-Doctoral Fellow at Research Laboratory of Electronics,MIT
Martinos Center for Biomedical Imaging, MGH

"Improving RF Safety in High Field MRI"

Magnetic Resonance Imaging (MRI) is a safe imaging technology that provides various clinical benefits. Basically, MRI is performed by exciting magnetic spins with radiofrequency (RF) pulses and receiving the response generated by these spins as they relax into their original state. This response is spatially encoded by using the gradient fields and converted to an actual image. Although it is not desirable, the body is exposed to an electric field during the RF excitation of the spins. The electric field distribution may cause heat dissipation in the conductive medium of body tissues. Safety problems related to such local heating arise when patients with medical implants are to be imaged using MRI. Currently, there are more than 1.5 million patients in US who have active implants (e.g., pacemakers and deep brain stimulators (DBS) ) in their bodies. 50 to 75 percent of these patients will need an MRI scan during the lifetime of their devices. Every 5 minutes, a patient is denied an MRI scan because of the safety issues related to an active implanted medical device. A solution towards improving the safety of patients with implants under MRI is crucial.

April 9th, 2014 at 12:00pm

Alayar Kangarlu, PhD
Associate Prof. Neurobiology
Department of Psychiatry, Radiology and Biomedical Engineering
Columbia University
Head of MRI Physics and Engineering
MRI Research Center
New York State Psychiatric Institute

"Magnetic Resonance in Psychiatry"

Magnetic resonance (MR) imaging has recently shown unique capabilities in characterization of psychiatric disorders. MR technologies such as voxel based morphometry (VBM), functional MRI (fMRI), magnetic resonance spectroscopy (MRS), and diffusion imaging (DTI) have shown to be capable of visualizing structural and functional manifestation of neural abnormalities and potential for characterizing their expression. MRI provides tools for in vivo examination of neuroanatomy with potential to differentiate among psychiatric and healthy subject groups. Finding the neural substrates of some psychiatric disorders is now within the reach of structural MRI. In addition, structural MRI is more potent when combined with functional and MRS studies. For example, two metabolites, GABA and glutamate have been found to be most prevalent in schizophrenia. Contrary to the early use of MRS, today’s scanners are capable of resolving glutamate-glutamine levels which sheds light on glutametergic biosynthetic pathway in schizophrenia. The great potential of fMRI lies in its ability to detect the BOLD signal in specific brain regions to identify differences of activity between brains of clinical, subclinical and healthy subjects. Arterial spin labeling has shown promise in revealing subtle brain perfusion changes occurring in psychiatric illnesses. DTI has visualized abnormalities in structural connectivity of the brain regions which in their comparison with functional connectivity maps offer a great tool for assessment of the etiology of psychiatric disorders. This talk will offer a brief discussion about MRI applications and their associated perils and payoffs in psychiatry research. In this context, the challenges in the development of the biomarkers for such use of MRI will be discussed. Potentials of MRI in providing new insight into the etiology and pathophysiology of psychiatric disorders will also be discussed.

March 27th, 2014 at 9:00am

Thomas O’Donnell, PhD
Senior Staff Scientist
CT Collaborations R&D
Siemens Healthcare

"Modelling the physics in the iterative reconstruction for transmission computed tomography,” by Johan Nuyts et al.

There is an increasing interest in iterative reconstruction (IR) as a key tool to improve quality and increase applicability of x-ray CT imaging. IR has the ability to significantly reduce patient dose; it provides the flexibility to reconstruct images from arbitrary x-ray system geometries and allows one to include detailed models of photon transport and detection physics to accurately correct for a wide variety of image degrading effects. This paper reviews discretization issues and modelling of finite spatial resolution, Compton scatter in the scanned object, data noise and the energy spectrum. The widespread implementation of IR with a highly accurate model-based correction, however, still requires significant effort. In addition, new hardware will provide new opportunities and challenges to improve CT with new modeling.

March 26th, 2014 at 12:00pm

Leo Tam, Ph.D.
Postdoctoral Associate
Magnetic Resonance Research Center
Department of Diagnostic Radiology
Yale University School of Medicine

"Nonlinear Gradient Encoding: Design and Implementation on a 3T Human Scanner"

Imaging with several detectors concurrently, known as parallel imaging, is a method to accelerate scans. However, parallel imaging encounters diminishing returns when increasing the number of detectors. Nonlinear gradient encoding allows encoding fields to complement the receiver coil detection to reconstruct equivalent images from less data. Nonlinear gradient encoding expands the encoding functions available to efficiently encode MR images. From the literature, nonlinear encoding has been shown to increase resolution in regions of the image, reduce peripheral nerve stimulation, and localize the field of view during radiofrequency excitation. Nonlinear gradient encoding and its optimization for faster images with equivalent image quality are examined. O-space imaging using the Z2 field has previously reported dispersed artifacts during accelerated scans. The inherent incoherence (distributed artifacts) of O-space imaging is explored and optimized within a compressed sensing framework. Null Space Imaging is a generalization of O-space imaging and uses an algebraic method of determining encoding fields from coil receiver profiles. Gradient hardware to perform nonlinear encoding is featured, including a 12 cm ID Z2 gradient wrist imaging insert, a 38 cm ID Z2 neuroimaging insert, and a 10 channel 20 cm ID gradient insert. The resulting body of work suggests nonlinear gradient imaging is a flexible and advantageous improvement on traditional parallel MR imaging.

March 25th, 2014 at 12:00pm

Riccardo Lattanzi, PhD
Assistant Professor of Radiology
New York University School of Medicine

"Novel tools for rational design and absolute performance assessment of MR coils"

At high and ultra-high magnetic field strengths, understanding interactions between tissues and the electromagnetic fields generated by radiofrequency coils becomes crucial for safe and effective coil design as well as for insight into limits of performance. In this work, we present a rigorous electrodynamic modeling framework, using dyadic Green’s functions, to derive the electromagnetic field in homogeneous spherical and cylindrical samples resulting from arbitrary surface currents. We show how to calculate ideal current patterns that result in the highest possible signal-to-noise ratio (ultimate intrinsic signal-to-noise ratio) compatible with electrodynamic principles. We identify familiar coil designs within optimal current patterns at low to moderate field strength, thereby establishing and explaining graphically the near-optimality of traditional surface and volume quadrature designs. We also document the emergence of less familiar patterns, e.g., involving substantial electric- as well as magnetic-dipole contributions, at high field strength. Performance comparisons with particular coil array configurations demonstrate that optimal performance may be approached with finite arrays if ideal current patterns are used as a guide for coil design.

March 18th, 2014 at 12:00pm

Alexander Leemans, Ph.D.
Associate Professor
Image Sciences Institute
University Medical Center Utrecht
The Netherlands

"The do’s and don’ts of processing and analyzing diffusion MRI data"

Abstract: "With its unique way of characterizing tissue organization, diffusion MRI (dMRI) has been used in a wide range of clinical and biomedical applications. In addition to a brief introduction to the basic concepts of dMRI, I will cover practical guidelines on quality and processing of dMRI data for subsequent analysis. Several considerations regarding dMRI limitations and data interpretation will also be presented. For relevant background information, see for instance and"

Bio: "Alexander Leemans is a physicist who received his PhD in 2006 at the University of Antwerp, Belgium. From 2007 to 2009, he worked as a postdoctoral researcher at the Cardiff University Brain Research Imaging Center (CUBRIC), Cardiff University, Wales, United Kingdom. In 2009, he joined the Image Sciences Institute (ISI), University Medical Center Utrecht, the Netherlands, where he currently holds a tenured faculty position as Associate Professor. He heads the PROVIDI Lab and is the developer of ExploreDTI, which is a graphical toolbox for investigating diffusion MRI data."

February 28th, 2014 at 12:00pm

Jan Paška, PhD student
Institute for Biomedical Engineering
University and ETH Zurich

"Modeling, Design, and Safety, of RF-Transmitters for High Field MRI"
February 25th, 2014 at 3:00pm

J. Thomas Vaughan, Ph.D.
Departments of Radiology, Electrical Engineering and Biomedical Engineering
University of Minnesota


UHF MRI requires new RF technology and methods to realize the full high field benefit to biomedical science and clinical diagnositcs. New multi-channel transmitters, receivers, and safety monitoring methods for 3T, 7T, and 10.5T are included. New approaches to these UHF challenges are being developed in Minnesota, at NYU, and at other luminary labs around the world. This talk will present some of Minnesota's work, will invite sharing from NYU's experience, and will provide a forum for mutual discussion of approaches taken by other labs. The results of this presentation and following discussions will lead to formulation of future collaborations between NYU and the UMN.

February 25th, 2014 at 12:00pm

Jeffrey Berman, Ph.D.
Assistant Professor of Radiology
The Children's Hospital of Philadelphia (CHOP) and the University of Pennsylvania, Perelman School of Medicine

"DTI and HARDI Tractography for Presurgical White Matter Mapping"

Abstract: A goal of neurosurgery is to preserve both functionally important cortices and the underlying white matter tracts. Diffusion MR tractography is a non-invasive method of visualizing the 3D course of white matter tracts. Traditional DTI fiber tracking is widely used for surgical planning, but fails to accurately represent the microstructure of crossing white matter tracts. The insufficiencies of DTI have motivated the application of high angular resolution diffusion imaging (HARDI) tractography to neurosurgical planning. This talk will describe the development, validation, and clinical utility of diffusion MR tractography for surgical planning, with an emphasis on recent HARDI techniques.

Bio: Jeffrey Berman is an Assistant Professor of Radiology at the Children's Hospital of Philadelphia (CHOP) and the University of Pennsylvania, Perelman School of Medicine. He received his PhD from the joint UC Berkeley - UC San Francisco bioengineering graduate program where he developed and validated diffusion MR techniques for surgical planning. During a postdoc at UC San Francisco, he used diffusion MR to study the developing brain of premature and term infants. At CHOP since 2010, his research interests include combining diffusion MR with MEG to study neuropsychiatric disorders such as autism and developing advanced diffusion MR tools for surgical planning.

February 12th, 2014 at 12:00pm

Mehmet Akcakaya, PhD
Instructor of Medicine, Harvard Medical School
Senior Research Scientist, Beth Israel Deaconess Medical Center

"Acceleration Methods for High-Resolution Cardiac MRI Using Compressed Sensing"

Abstract: One of the major challenges of cardiac MRI is its lengthy acquisition, which limits the achievable spatial and temporal resolutions, and volumetric coverage. In this talk, we will discuss novel compressed sensing (CS) based image reconstruction techniques used for accelerating data acquisition in cardiac MRI. We will introduce techniques that utilize patient and anatomy-specific information to improve reconstruction quality with respect to standard CS methods, as well as to state-of-the-art parallel imaging techniques. We will validate these techniques in accelerated high-resolution coronary artery and late gadolinium enhancement imaging. We will also extend these acceleration techniques to accelerated perfusion cardiac MRI with free-breathing applicability. We will conclude with a brief overview of ongoing research, as well as future research directions.

February 11th, 2014 at 12:00pm

Hersh Chandarana, MD
Assistant Professor
Department of Radiology
New York University School of Medicine

"Advanced Imaging of the Renal Cancer and Renal Function"

Renal cancers are being increasingly diagnosed incidentally. Some of these tumors are aggressive whereas others have relatively indolent course. Current morphologic imaging is limited in assessment of tumor aggressiveness. Promising MR techniques such as intravoxel incoherent diffusion weighted imaging (IVIM) and dynamic contrast enhanced (DCE) imaging will be discussed. Unsolved problems and clinical need will be highlighted. There is also need to develop and validate better techniques to assess renal function. MR techniques such as DCE, DTI, and BOLD have shown considerable early promise.

Short Bio: Dr. Chandarana is an abdominal radiologist and clinical scientist in the Department of Radiology, New York University School of Medicine, with interest in advanced oncologic and functional imaging.

February 4th, 2014 at 12:00pm

Lauren Burcaw, Ph.D.
Postdoctoral Fellow
Department of Radiology
New York University School of Medicine

"Time Dependent Diffusion in White Matter"

The sensitivity of time-dependent diffusion to the overall structure of its environment makes it appealing tool in the study of white matter fibers. Previous studies have mainly focused on increasing the q value as much as possible under a clinical system. In contrast, we vary the diffusion time, t, which allows us to probe the structure by increasing the diffusion length.
We observe via DTI measurements on a fiber phantom that the long time diffusion exhibits a unique (log t)/t dependence transverse to fibers as a result of disordered packing. This has implications in a variety of diffusion experiments such as oscillating gradients and axon diameter estimation.
This leads to the question of whether or not time-dependent diffusion is even observable on a clinical scanner. So far, the literature is inconclusive. We scan five healthy volunteers using a DTI protocol with diffusion times ranging from 26 to 400 ms and find that we indeed do see time dependence parallel and perpendicular to the axons. The effect is strongest along the axonal direction possibly indicative of heterogeneities within axonal fibers.

January 28, 2014 at 12:00pm

Dr. N. Jon Shah
Director of the Institute of Neurosciences and Medicine
Forschungszentrum Julich in Germany

"Simultaneous Multimodal Imaging: MR-PET-EEG at 3T and 9.4T"
January 21, 2014 at 12:00pm

Olivier Reynaud, Ph.D.
Postdoctoral Fellow
Department of Radiology
New York University School of Medicine

"Characterization of microvascular flow dynamics using flow-enhanced MRI"

In clinical neuroimaging, perfusion MRI is of spectacular importance to study cerebrovascular diseases and cancer. However, at the moment, there is no perfusion MRI sequence that allows for a complete, non-invasive and precise quantification of microvascular flow dynamics. This work focuses on the use of the recently introduced Flow Enhanced Signal Intensity method (FENSI) to characterize and quantify vasculature at capillary level, at high magnetic field (7T). For that purpose, the possible quantification of blood flux with FENSI is explored in vivo. The combination of flux quantification and flow-enhanced signal (compared to Arterial Spin Labeling) can make of FENSI an ideal method to characterize in a complete non-invasive way the brain microvasculature. After removal of magnetization transfer (MT) effects, the blood flow dynamics are studied with FENSI in a very aggressive and propagative rat brain tumor model: the 9L gliosarcoma. The objective is to assess whether FENSI is suitable for a longitudinal non-invasive characterization of microvascular changes associated with tumor growth. The results obtained with FENSI are compared with literature on 9L perfusion and immuno-histochemistry. functional MRI might also benefit from the development of flow enhanced MRI. With the implementation of a new MT-free FENSI technique, the possibility to map the brain cerebral functioning based on a quantitative physiological parameter (CBFlux) more directly related to neuronal activity than the usual BOLD signal is within reach. Preliminary results on rats and human brain are presented.

January 14, 2014 at 12:00pm

Jean-Christophe Brisset, PhD
Postdoctoral Fellow
Department of Radiology
Center for Biomedical Imaging
New York University Langone Medical Center

"Susceptibility Blooming Effects on High Field MR"

In recent years, there is increasing recognition of cerebral microbleeds (MCBs) in patients with cerebrovascular diseases and dementia with MRI, in particular at high-field-strength. The detection of CMBs or lesions small iron component (i.e. amyloid plaques) depends on several MRI characteristics including field strength, pulse sequence, imaging parameters, spatial resolution, iron concentration, and image post-processing. We hypothesized that using optimal imaging sequence/parameters, there is a significantly enhanced blooming effect (ie. larger area than the actual object size) at high field MR, which has potential to detect much smaller iron containing lesions or structures. In this study, we used 3D gradient-echo imaging to quantify the susceptibility blooming factor (i.e. detected size/real size of object) based on a tube phantom with different iron concentrations and post-mortem brain slices. Ultra-high-field MR (e.g. 7T) provides superb susceptibility contrast (i.e. marked blooming effects) to enhance the capability of the detection of small lesions that contain iron component. We have characterized and demonstrated the actual degree, enhanced visibility and imaging optimization of the blooming effects based on the results of phantom, simulation, and clinical images on 7T as compared to standard 3T and 1.5T MR. We investigate the use of available iron contrast agent to found the optimal parameters in a clinical perspective at high field (7T). Our results suggest that a 3D gradient echo with optimal TE and voxel size help to detect even small quantity of iron. We establish each blooming factor (measured size/actual size) for specific TE, iron concentration, or spatial resolution on 7T as compared to 3T and 1.5T. The blooming factor may provide a tool to approximate the actual size of structure even with a size even smaller than a voxel.

December 19th, 2013 2:00pm

Miriam Bredella, M.D
Associate Professor of Radiology
Musculoskeletal Imaging and Intervention Division Department of Radiology Massachusetts General Hospital
Harvard Medical School

"The Bone-Fat Connection"

Dr. Bredella's research is at the interface of radiology and endocrinology. In this talk, she will describe her work investigating the effects of different kinds of fat depots on bone density, structure, strength, and marrow fat in obesity and anorexia nervosa. She is also investigating the role of growth hormone in improving bone health and decreasing cardiovascular risk in obesity. Dr. Bredella attended medical school at the University of Hamburg in Germany. She subsequently worked for 2 years at the Osteopororsis and Arthritis Research Group at UCSF under Harry Genant. She completed her residency at UCSF followed by a musculolskeletal radiology fellowship at MGH, where she has been on staff since 2005. She is currently an Associate Professor of Radiology at Harvard Medical School. She was previously awarded an NIH K23 grant and most recently received an R01 grant focusing on skeletal dysregulation in obesity. She is also a co-investigator on an R24 grant examining the role of marrow fat.

December 13th, 2013 11:00am

BJ Casey, PhD.
Director of Sackler Institute for Developmental Psychobiology
Professor of Developmental Psychobiology
Weill Medical College of Cornell University

"Combining human imaging with mouse genetics to understand fear regulation and development"

About the speaker: BJ Casey is the Sackler Professor and Director of the Sackler Institute at Weill Medical College of Cornell University. She is a pioneer in novel uses of neuroimaging methodologies to examine behavioral and brain development. Her program of research focuses on attention and affect regulation, particularly their development, disruption and neurobiological basis. She has been examining the normal development of brain circuitry involved in attention and behavioral regulation and how disruptions in these brain systems (prefrontal cortex, basal ganglia and cerebellum) can give rise to a number of developmental disorders. Using a mechanistic approach she has dissociated attentional deficits observed across the disorders of Attention Deficit Hyperactivity Disorder, Obsessive Compulsive Disorder, Tourette Syndrome and Childhood Onset Schizophrenia.

December 10th, 2013 12:00pm

Thomas Koesters, Ph.D.
Research Scientist
Department of Radiology
Center for Biomedical Imaging
NYU School of Medicine

"Advanced PET Image Reconstruction for the Siemens Biograph mMR "

Abstract: The Siemens Biograph mMR installed in the CBI (first floor of 660 First Ave) allows simultaneous acquisition of MR and PET data. Although spatially and temporally aligned raw data is available, both modalities are often treated separately and corresponding images are only fused after independent reconstruction. This talk gives an overview of current research projects in the CBI where MR information is used to improve the PET image reconstruction. These projects include motion detection, motion correction and finally joint reconstruction of PET and MR data.

2012 - 2013: Development Engineer at GEA
2010 - 2012: PostDoc at European Institute for Molecular Imaging (Muenster, Germany)
2006 - 2010: PhD in Mathematics at WWU (Muenster, Germany)
2001 - 2006: Mathematics at WWU (Muenster, Germany)

December 3rd, 2013 12:00pm

Ileana Jelescu, Ph.D.
Postdoctoral Fellow
Department of Radiology
NYU School of Medicine

" Magnetic resonance microscopy of Aplysia neurons: studying neurotransmitter-modulated transport and response to stress"

Recent progress in MRI has opened the way for micron-scale resolution, and thus for imaging biological cells. The goal of my thesis work was to perform magnetic resonance microscopy (MRM) on the nervous system of Aplysia californica, a model particularly suited due to its simplicity and to its very large neuronal cell bodies, in the aim of studying cellular-scale processes with various MR contrasts. Experiments were performed on a 17.2 T horizontal magnet, at resolutions down to 25 µm isotropic. Initial work consisted in conceiving and building radiofrequency microcoils adapted to the size of single neurons and ganglia. The first major part of the project consisted in using the manganese ion (Mn2+) as neural tract tracer in the nervous system of Aplysia. We performed the mapping of axonal projections from motor neurons into the peripheral nerves of the buccal ganglia. We also confirmed the existence of active Mn2+ transport inside the neural network upon activation with the neurotransmitter dopamine. In the second major part of the project, studied the changes in water ADC at different scales in the nervous system, triggered by cellular challenges. A 3D Diffusion-Prepared FISP sequence was first implemented, which met criteria for high resolution in a short acquisition time, with minimal artifacts. Using this sequence, ADC measurements were performed on single isolated neuronal bodies and on ganglia tissue, before and after two types of challenge (hypotonic shock and ouabain). Both types of stress produced an ADC increase inside the cell and an ADC decrease at tissue level. The results favor the hypothesis that the increase in membrane surface area associated with cell swelling is responsible for the decrease of water ADC in tissue, typically measured in ischemia or other conditions associated with cell swelling.

November 26th, 2013 12:00pm

Andrew Maudsley, Ph.D.
Professor of Radiology
Miller School of Medicine
University of Miami

"Whole Brain Metabolite Imaging"

MR spectroscopic measurements of human brain are commonly limited to small regions to minimize difficulties associated with magnetic field inhomogeneities and lipid contamination; however, several clinical applications could greatly benefit from obtaining MRS measurements over larger brain volumes, including for example, measurement of diffuse tissue injury with traumatic brain injury, characterization of tumor volumes for therapy planning, and localization of neocortical epilepsy. This presentation will review some of the experimental approaches that can be used to extend the measurement volume for MR spectroscopic imaging, and show examples of clinical applications of these methods.

November 19th, 2013 12:00pm

Valentin Riedl, MD, PhD
Department of Neuroradiology and Nuclear Medicine
TUM-Neuroimaging Center
Klinikum Rechts der Isar der Technischen Universität München (TUM), Germany

"How does multimodal brain imaging advance our understanding of large-scale brain networks?"

Recent functional magnetic resonance imaging (fMRI) revealed organized activity in the brain at rest which gained enormous relevance for systems and clinical neuroscience. Particularly, this organized activity is defined by synchronous, low frequency (<0.1Hz) fluctuations of the blood-oxygenation-level-dependent (BOLD) fMRI signal between remote brain areas, termed resting-state functional connectivity (rs-FC). However, the neurophysiological and metabolic underpinnings of rs-FC are still incompletely understood. In this talk I will summarize recent findings of rs-fMRI and present first data of simultaneous PET/MR imaging in humans indicating a neuronal basis of resting state FC.

November 5th, 2013 12:00pm

Ryan Brown
Assistant Professor
Department of Radiology
Center for Biomedical Imaging
New York University Langone Medical Center

"Adventures in 3D Printing"

CBI's RF lab (second floor of 660 First Ave) houses a 3D printer that provides the capability to build a wide range of MRI compatible fixtures. This talk is aimed at educating potential users on the printer's general capabilities for rapid prototyping. Specifications on CAD software, build-size, resolution, and print speed will be reviewed in the context of objects designed at CBI during the past year. While many of the examples are hardware-related fixtures such as anatomically-correct and aesthetically-pleasing RF coil formers, it is anticipated that the lecture will spark interest in a more expansive range of applications.


October 30, 2013 12:00pm

Rainer Schneider
Institute of Biomedical Engineering and Informatics
Ilmenau University of Technology
Ilimenau, Germany

"Multi-slice pTX pulse design for local signal recovery"

For compensating the signal loss in GRE-based sequences induced by through-plane susceptibility, two state-of-the-art techniques using the parallel transmit technology (pTX) were analyzed. Both approaches, the tailored 3-dimensional RF pulses (3DTRF) and time-shifted spokes excitation, were implemented on the 3T Skyra system with two integrated whole-body transmit channels. The methods were extended and evaluated with human in-vivo experiments.


September 24, 2013 12:00pm

Jelle Veraart
Vision Lab, University of Antwerp
Antwerp, Belgium

"Accurate estimation of diffusion MRI parameters"

Diffusion magnetic resonance imaging (dMRI) is currently the method of choice for the in vivo and non-invasive quantification of water self-diffusion in biological tissue. Several diffusion models have been proposed to obtain quantitative diffusion parameters. Those parameters might provide novel information on the structural and organizational features of biological tissue, the brain white matter in particular. However, an accurate and precise estimation of those diffusion parameters remains challenging because of the non-Gaussian MR data distributions. Indeed, widely used estimator – e.g. the class of least squares estimators – will show systematic errors in the estimation of diffusion measures because the actual data statistics are not taken into account. The squashing of the ADC peanut or the overestimation of the kurtosis metrics are typical examples of such, so called, noise artifacts. During the seminar, an overview of the commonly used parameter estimators will be given. Their strengths and limitations will be discussed. In addition, a comprehensive framework for accurate diffusion MRI parameter estimation will be introduced.


September 17, 2013 12:00pm

Lisa Mosconi, PhD
New York University School of Medicine
New York, New York

"Preclinical Detection of Alzheimer’s Disease Using Brain Positron Emission Tomography Imaging"

The development of biomarkers for the preclinical detection of Alzheimer’s disease (AD) is a vital step in developing prevention therapies. For many years, we and others have been using biological markers of AD pathology and its effects on brain structure and function to characterize early changes in presymptomatic individuals at risk for AD. Such markers include in vivo brain Magnetic Resonance Imaging (MRI); Positron Emission Tomography (PET) imaging using 2-[18F]fluoro-2-Deoxy-D-glucose (FDG) and N-methyl[ 11C]2-(4'-methylaminophenyl)-6-hydroxy-benzothiazole (PiB) as the tracers to measure glucose metabolism and fibrillar amyloid-beta (Aß) deposition, respectively; cerebrospinal fluid levels of Aß1-40 and 1-42, tau pathology (total tau and hyperphosphorylated tau231) and inflammation (F2-isoprostanes); and recently plasma measures of oxidative stress (activity of mitochondria cytochrome oxidase, electron transport chain complex IV, COX).
This lecture will give an overview of biomarker findings in individuals at risk for AD, with the main focus on presymptomatic individuals carrying genetic mutations responsible for early-onset familial AD and cognitively normal (NL) people with a first degree family history of LOAD. Overall, these studies have shown that it is possible to identify and track biomarker changes prior to cognitive impairments arise and along with AD progression. All told there is considerable promise for an early and specific diagnosis of AD by assessing biomarkers in NL individuals at risk for AD.


August 20, 2013 12:00pm

Dikoma C. Shungu, PhD
Professor of Physics in Radiology
Weill Medical College of Cornell University
New York, New York, USA

"In Vivo Proton MRS Measurement of Cortical GABA, Glutamate and Glutathione: Methods and Selected Clinical Research Applications"

The most widely investigated neurochemical hypotheses of major psychiatric disorders now posit neurodevelopmental deficits that involve, among others, dysregulations of the inhibitory and excitatory amino neurotransmitter systems of gamma-Aminobutyric acid (GABA) and glutamate (Glu), respectively. Glutathione (GSH) is a major intracellular antioxidant and redox regulator, whose dysregulations and in vivo deficits have been implicated in various neurological, neuropsychiatric and neurodegenerative disorders. Currently, proton magnetic resonance spectroscopy (1H MRS) is the only noninvasive neuroimaging technology that offers the possibility to investigate abnormalities in GABA, Glu and GSH in the living human brain. In this presentation, our decade-long experience in developing and optimizing the relevant MRS technology will first be described, and then the full power and growing importance of the technology in biomedical and neuroscience research will be illustrated with selected clinical applications in neuropsychiatry and neurology.


July 30, 2013 12:00pm

Chao-Gan Yan, PhD
Nathan Kline Institute for Psychiatric Research,
Child Mind Institute, and
New York University Child Study Center

"Resting-state fMRI: Algorithms, Applications to Brain Disorders and Data Processing"

Resting-state functional magnetic resonance imaging (R-fMRI) has emerged as a mainstream imaging modality with myriad applications in basic, translational and clinical neuroscience. Beyond impressive demonstrations of accuracy, reliability and reproducibility for measures of intrinsic brain function, this approach has gained popularity due to its sensitivity to developmental, aging and pathological processes, ease of data collection in otherwise challenging populations, and amenability to aggregation across studies and sites. In this talk, I would like to introduce the principles, computational algorithms and methodological issues of R-fMRI as well as its clinical application to brain disorders (e.g., Alzheimer's disease). Finally, I would like to demonstrate the data processing of R-fMRI with our convenient pipeline toolbox DPARSF.


July 29, 2013 12:00pm

Kristian Bredies, PhD
Institute for Mathematics and Scientific Computing
University of Graz

"Variational modelling with total generalized variation"

We discuss the recently introduced total generalized variation (TGV) which is a well-suited regularizer for variational imaging problems. In addition to the well-known total variation (TV), it does not only model free discontinuities but is also aware of higher-order smoothness. It can be interpreted as a regularizer which adaptively selects the appropriate smoothness level.
After studying basic properties of the TGV functional, we show how abstract methods for finding convex-concave saddle point problems can be applied to solve variational imaging problems with TGV-regularization. Several applications are presented, ranging from basic imaging problems like denoising and deconvolution to applications in MRI, CT and compressed sensing.
Finally, we show the potential of general measure-based regularization beyond TV and TGV. In particular, convex regularization functionals are discussed which are able to count vertices and edges. Furthermore, their application to the reconstruction of elongated structures is presented.


July 17, 2013 12:00pm

Prof. Haim Azhari
Associate Professor
Department of Biomedical Engineering
Technion-Israel Institute of Technology

"Improving PET Imaging"

PET is a powerful modality in medical imaging. However, its spatial resolution is very poor compared to other major modalities (CT, MRI and Ultrasound). The challenge is to improve PET image quality without inserting any physical changes in the scanner hardware. In this lecture two approaches will be introduced. The first approach is to implement super-resolution strategy. With super-resolution several low resolution images are acquired, where each image is shifted by a sub pixel distance relative to the other. An algorithm is then implemented to combine the information and produce a high resolution image. The second approach is implemented on data acquired by a hybrid PET-CT scanner. The images obtained from the CT are fused with the images obtained from the PET using an algorithm called "Hybrid Computerized Tomography (HCT)". The obtained images depict sharp border PET distribution.


June 19, 2013 9:00am

Nicole Seiberlich, PhD
Assistant Professor
Biomedical Engineering
Case Western Reserve University and Biomedical Engineering

Vikas Gulani, MD, PhD
Assistant Professor
Director of MRI
Departments of Radiology, Urology
Case Western Reserve University

"Pushing the Limits: Novel Acquisition and Reconstruction Strategies for Rapid Quantitative MRI"


June 13, 2013 9:00am

Gregory Metzger, PhD
Associate Professor
Department of Radiology
University of Minnesota

"Developing MRI biomarkers of prostate cancer aggressiveness and UHF body imaging"


May 28, 2013 12:00pm

Assaf Tal, PhD
NYU Langone Medical Center

"Short echo time spectroscopy in the human brain via Hadamard Encoding at 3T"

Abstract: "Magnetic resonance spectroscopy is used routinely to measure metabolite concentrations in the human brain. Due to fast relaxation times and complex J-coupling patterns, many of the most important metabolites observable with spectroscopy - such as GABA and Glutamine/Glutamate - are difficult to discern using standard spectroscopic techniques. In this talk, I will argue why short echo times (<10 ms) offer significant benefits when trying to image such metabolites, why in-vivo spectroscopy has only fairly recently begun exploring these possibilities, and present our own approach for doing so using radiofrequency Hadamard pulses.
I will also briefly discuss two other projects which may be of interest to other researchers at the CBI: our approach to dealing with B0 field drifts, as well as our approach to analyzing global white/grey matter metabolite concentrations."

Short bio: "Assaf Tal obtained his BSc in physics from the Hebrew University in Israel, and his PhD from the Weizmann Institute of Science in Israel with Prof. Lucio Frydman in the field of liquid state NMR, where he has done work on single-scan methods in 2D NMR as well as fast imaging methods based on quadratic spin phase. His current post-doctoral research in the lab of Oded Gonen focuses on developing new sequences and processing methodologies for in-vivo human brain spectroscopic imaging."


May 15, 2013 12:00pm

Sanjeev Chawla, PhD
Research Associate
Department of Radiology, University of Pennsylvania

"MR Imaging and Spectroscopy in Brain Infections, Brain Tumors and Head and Neck Cancers"

Abstract: Non-invasive differentiation of brain abscesses such as pyogenic and tuberculous, anaerobic and aerobic or sterile is essential for facilitating prompt and appropriate treatment of patients. MR spectroscopy and magnetization transfer MR imaging may be used to characterize intracranial cystic lesions with similar features on conventional MR imaging. Precise MR imaging correlation of different stages of neurocysticercosis with histopathology is essential for better understanding of the disease that is usually hampered by complexities in performing such studies on humans. Therefore, detailed correlative MR imaging and histopathological studies on pigs infected with neurocysticercosis are warranted.

Given the heterogeneous nature of neoplastic lesions and inherently different physiological information provided by different MR pulse sequences, multi-parametric data analysis may be a better approach in differential diagnosis, predicting prognosis, monitoring treatment response in brain tumors and head and neck cancers with greater accuracy. Combined use of MR spectroscopy and perfusion weighted imaging may be used to distinguish histological grades, histological subtypes and genetic profiles of the gliomas.


May 10, 2013 12:00pm

Evren Ozarslan, PhD
Lecturer on Radiology
Department of Radiology, Brigham and Women’s Hospital, Boston, MA

"Recent advances in diffusion-weighted MRI: From multi-pulse experiments to new analysis schemes."

Conventional magnetic resonance (MR) imaging scans suffer from limited resolution that prohibits the visualization of individual cells thus providing information at coarse length scales. To obtain information at smaller length scales, the MR signal can be sensitized to self-diffusion of water molecules whose motional history is influenced by the local microstructure. I will present several new developments in the field of diffusion-weighted MRI. Emphasis will be given to the multiple pulsed field gradient techniques, which could be used to characterize the local microstructural features of the medium without the need to employ strong magnetic field gradients. In the second part of the talk, I will describe the recently introduced mean apparent propagator (MAP) MRI technique, which is a comprehensive computational framework that could be employed to address a number of challenges encountered in the analysis of diffusion-weighted MRI data.

About the speaker: Evren Özarslan is a research associate at Brigham and Women's Hospital and holds a concurrent academic appointment at Harvard Medical School (HMS). Before joining HMS, Dr. Özarslan performed research at the Section on Tissue Biophysics and Biomimetics (STBB), NICHD, National Institutes of Health (NIH) first as a postdoctoral fellow, then as a scientist with the Center for Neuroscience and Regenerative Medicine (CNRM) and the Henry M. Jackson Foundation. He graduated with a Bachelor of Science in Physics from the University of Illinois at Urbana-Champaign, and obtained his M.S. degree in Biomedical Engineering and Ph.D. in Physics, both from the University of Florida. His current research is on modeling diffusion in biological tissue and other porous media with the specific aim of characterizing the microstructure of the specimen using noninvasive magnetic resonance techniques.


April 30, 2013 12:00pm

José P. Marques
Department of Radiology, University of Lausanne
Switzerland Functional and Metabolic Imaging Laboratory, EPFL, Switzerland

"Imaging with inhomogeneous RF fields: avoiding, fighting and making the most of them"



April 16, 2013 12:00pm

Christopher M. Collins, PhD
Professor of Radiology
NYU Langone Medical Center

Use of high-permittivity materials to enhance coil performance in MRI

In a growing number of studies and applications, strategic selection and placement of passive high-permittivity materials are shown to improve SNR and/or reduce required transmit power in imaging a select region of interest. We will discuss some basic mechanisms by which high-permittivity materials can improve RF efficiency in MRI and review a variety of cases where they have been demonstrated to do so.


April 5, 2013 12:00pm

Daniel S. Weller, Ph.D.
University of Michigan

Sparse Modeling for Magnetic Resonance Imaging

In this talk, I discuss my research concerning sparse modeling for magnetic resonance imaging. First, I elaborate on three methods for using sparsity to improve upon GRAPPA, an autocalibrating reconstruction method for accelerated parallel imaging. These three methods (1) denoise the reconstructed k-space, (2) regularize the calibration of the GRAPPA kernels, and (3) jointly estimate the full k-space and GRAPPA kernels using prior and likelihood models.

All these methods make use of fixed parameters that control the regularization. While hand-tuning these methods may be possible, we desire an automatic parameter selection method that would work for data-preserving reconstructions. To this end, I introduce Stein's Unbiased Risk Estimate and describe how I extend it to data-preserving regularized parallel imaging reconstructions.

I follow this discussion by outlining my current research exploiting sparsity to prospectively correct for head motion in functional MRI. I demonstrate that this usage of sparsity allows for high-quality time-series correlation analysis in the presence of head motion.


April 2, 2013 12:00pm

Susumu Mori, Ph.D.
Professor, Department of Radiology
Johns Hopkins University School of Medicine, Baltimore, MD

Multi-Modal Image Analysis Based on Atlas-Based Spatial Filtering and Cloud-Based Analysis System

One of the most challenging aspects of image analysis is the overwhelming amount of spatial information. For example, typical T1-weighted image with 1mm resolution contains more than 1 million voxels, each of which carry noisy information. Cross-contrast (e.g. T1 and DTI) and cross-modality (e.g. MRI, MRS, fMRI, PET) data integration have been postulated as potentially a powerful approach to delineate anatomical and functional phenotype of patient populations, which would lead to further increase in spatial information with different coordinate frames and, thus, a systematic reduction of the spatial dimension seems an essential and inevitable requirement. This presentation will introduce our current effort to establish a modern MRI atlas system and associated software tools to perform atlas-based image analysis, in which the entire spatial information is reduced to approximately 200 pre-defined structures. For demonstration, integrative analyses of anatomical MRI, DTI, MRS, and rs-fMRI data and clinical applications will be shown. The automated pipeline for the atlas-based analysis is currently being deployed using a cloud-based architecture for dissemination and future direction of the service model will also be discussed.


March 26, 2013 12:00pm

Pablo Velasco, Ph.D.
Senior Research Scientist/Chief MR Physicist
New York University
Center for Brain Imaging

Real-Time Data-Quality Monitoring of fMRI Data

Functional MRI data quality can be compromised by a series of factors --especially motion and spikes-- which are hard to assess until you start processing your data, well after the scanning session is over. By that time, your subject is gone and you might find you are left with too little data to be able to include that subject in your analysis. I will present the implementation of a real-time data-quality monitoring tool that reconstructs the images, estimates motion parameters and some other statistics on them and displays them on the screen as they are being acquired, so that users can repeat those runs with excessive motion or with spikes, and give feedback to the subject on how well he or she is avoiding motion in the scanner.


March 5, 2013

Dr. Gisele Caseiras, M.D., Ph.D.
PhD in Neuroradiology at University College London-UCL, London England

The use of Conventional and Advanced Magnetic Resonance Technique in the Assessment of Primary Brain Tumours

Low-grade gliomas in adults are diffusively infiltrating tumours that may undergo malignant transformation into high-grade gliomas. This malignant transformation is highly variable and difficult to predict in an individual patient. The purpose of this study was to investigate the value of conventional and advanced magnetic resonance imaging in patients with histology-proven low-grade gliomas and the potential role of these methods as markers of malignant transformation.


February 26, 2013

Leeor Alon
Doctoral Candidate
The Sackler Institute of Graduate Biomedical Sciences
New York University School of Medicine, NY

A New Application for MR - Safety Testing of RF Emitting Devices

By the end of 2012, the number of mobile-connected devices will exceed the number of people on Earth, and by 2016 there will be 1.4 mobile devices per capita." -Cisco VNI Mobile 2012.

Radio frequency (RF) emitting wireless devices such as mobile phones are required to undergo standardized safety testing prior to entering the consumer market. Strict regulations are imposed on the amount of RF energy these devices are allowed to emit to prevent excessive deposition of RF energy into the body. In this presentation, a novel safety evaluation test for wireless devices using magnetic resonance (MR) thermometry is proposed.


February 5, 2013

Uri Nevo, PhD
Senior Lecturer and Director
Laboratory of Cellular Biophysics and Imaging
Tel-Aviv University, Israel

Studying the cellular origins of Diffusion Weighted NMR

Diffusion Weighted NMR (DW-NMR) of tissues characterizes two linked cellular properties: microstructure and viability. DW-NMR in cells is affected by structures that restrict and hinder diffusion. Following brain insults, such as ischemia, water displacement is attenuated and is commonly linked to microstructural changes affecting diffusion. Water displacement is linked not only to microstructure but also to cellular viability and function, as in the case of neuronal activity that is suggested to be correlated with restricted diffusion.

We attempt to quantify the different components of water displacement in cells, in order to obtain an accurate characterization of cells' microstructure and function. In the coming lecture I will first describe our method for quantifying pore size distribution, towards the characterization of cells' sizes. This is done by the use of a double pulsed field gradient experiment, in which gradient pairs are varied by amplitude and direction.

A central hypothesis in our research is that diffusion is not the only component of displacement in cells: we suggest that a significant component of water displacement in neurons is that of actively induced micro-streaming. I will describe our theoretical and experimental work aiming to quantify the relation of function and micro-streaming inside neurons. This is done by using biophysical models and by DW-NMR of isolated and viable neural tissues. I will end by speculating the possible implications of our work on brain function study.


January 29, 2013

Manushka Vaidya
Doctoral Candidate
The Sackler Institute of Graduate Biomedical Sciences
New York University School of Medicine

Understanding and manipulating B1 field distribution inside a dielectric object

The first part of the talk will cover the dependence of the B1 spatial distribution on the electrical properties of the sample and the magnetic field strength. Unanticipated B1 field patterns may be encountered during simulation and experiments, particularly at high operating frequencies. While a distinctive curling of the B1 field is observed at high field strengths, elaborate checkerboard-like patterns may be obtained for certain dielectric samples. In this work, we use full-wave electrodynamic simulations based on dyadic Green's functions to study the effect of the electrical properties of the sample and main magnetic field strength on the B1 field pattern inside a uniform cylindrical object. We show examples of the curling of the field and interference patterns near resonance, providing a conceptual explanation for each case.

In the second half, manipulation of the B1 field distribution inside a sample by placing dielectric pads at a distance from a surface transmit coil will be discussed. The use of dielectric pads between the radiofrequency (RF) coil and sample has been proposed to "focus" the B1 field into the sample to improve transmit efficiency. In this study, we investigated how dielectric pads placed at a distance from the RF coil affect the B1+ spatial distribution inside the sample. We performed numerical simulations of the B1+ distribution inside a uniform cylinder at 7T for various positions of the dielectric pad with and without a surrounding shield. Manipulating B1 spatial distribution with dielectric pads can be advantageous for various MR applications, including improving RF homogeneity at ultra-high fields


January 15, 2013

Dmitry Novikov, PhD
Assistant Professor of Radiology
The Bernard and Irene Schwartz Center for Biomedical Imaging
New York University Langone Medical Center, NY

Characterizing microstructure of living tissues with time-dependent diffusion

A major challenge of in vivo MR is to characterize tissue microstructure at the cellular level, orders of magnitude below the imaging resolution. I will show how a diffusion measurement, taken at a range of diffusion times, can distinguish between different classes of microgeometry. Based on the specific values of the dynamical exponent of a velocity autocorrelator measured with diffusion MRI, we identify the relevant tissue microanatomy in muscles and in brain, quantify cell membrane permeability in muscles, and reveal the microstructural changes driving the diffusivity drop in ischemic stroke. Our framework presents a systematic way to identify the most relevant part of structural complexity with diffusion.

January 10, 2012

Giselle A. Suero-­‐Abreu, MD, MS
Doctoral Candidate
The Slacker Institute of Graduate Biomedical Sciences
New York University School of Medicine, NY


Significant progress has been made in our understanding of the pathogenesis of brain tumors partly due to the development of genetically engineered mouse models that recapitulate the human disease. In this regard, in vivo micro-­‐MRI protocols are powerful tools for the non-­‐invasive, three-­‐dimensional (3D) characterization of these preclinical cancer models and are gradually being recognized as an integral part of basic and translational brain tumor research. In our study, we optimized an in vivo high resolution Manganese-­‐Enhanced MRI protocol (MEMRI) for the characterization of tumor progression in a novel mouse model of Medulloblastoma (MB), the most common malignant pediatric brain tumor originating in the cerebellum (Cb).

In this talk, I will present the characteristics of our tumor model and show that our imaging approach successfully allowed the detection of early tumoral lesions and the longitudinal assessment of their progression into advanced-­‐stage tumors. Furthermore, I will discuss recent results which indicate that these tumors display at least two distinct molecular and imaging features. Ultimately, we are interested in correlating these findings with the clinical and imaging characteristics of human MBs and we expect to draw insights that inform the design of studies to test current and novel drug therapies using this unique pre-­‐clinical model.


February 9, 2012

Tracy Butler, MD
Assistant Professor of Psychiatry and Neurology
Epilespsy Center
New York University Langone Medical Center, NY

Neuroreceptor PET and PET/fMRI: proposed and ongoing projects in epilepsy and neuropsychiatry

In addition to her proposed research, she will also present her ideas as to why simultaneous imaging of PET and MRI is so exciting..


February 13, 2012

Sharmila Majumdar, PhD
Professor of Radiology and Biomedical Imaging
School of Medicine
University of California, San Francisco, CA


Osteoarthritis (OA) is a degenerative disease that is characterized by cartilage thinning and compositional changes, and it is estimated that 20 million individuals in the United States are living with the disease with an annual cost of over 15 billion dollars.

The disease preferentially affects older (> 65 years) individuals, but with traumatic injury such as anterior cruciate ligament injury being a risk-­‐factor, 1 in 20 working age (18-­‐64 years old) adults report activities being limited by arthritis. Despite the recognition that 3D imaging is likely to provide important information regarding joint health, OA, and that biomechanics plays a role in OA and its' progression, the translation and cross-­‐correlation of these metrics have been limited.

The overall objective of this talk is to integrate cutting edge quantitative imaging technologies, link the image derived metrics to joint kinematics, kinetics, patient function, and translate the linkages found to the musculoskeletal clinic, thus affecting patient management and outcome


February 28, 2012

Gadi Goelman, PhD
Associate Professor of Medical Biophysics
The Hebrew University of Jerusalem, Israel

Coherent low frequency fluctuations of the BOLD signal in resting state (rest-fcMRI) were shown to contain functional neuronal network information. Resting-state networks (RSN) exhibit positive correlations between the regions that constitute the network, suggesting a functional link between them. However, several RSNs were shown to have an inverse correlation between each other. The underlying physiological mechanisms and the relevance of negative correlations to neurobiology are not clear and are the subject of this study.

We compared human and rat rest-fcMRI data, making use of both the similarities (e.g., similar organization: cortical vs. non-cortical structures, inter-hemispheric symmetry etc.) and differences (e.g., different hemodynamic characteristics such as cardiac rates and spatial distances) between them. In addition, the fact that the rats' cortex is relatively unfolded, enables to minimize confounding effects of CSF and large blood vessels on the rest-fcMRI correlations.

We show that: (i) Negative correlations observed in rest-fcMRI reflect true physiological traits and are not the mere result of mathematical biases introduced by data analysis. (ii) At least two distinct mechanisms may underlay the appearance of negative correlations, reflecting the actual synchronization between regional neural activities on the one hand and their manifested BOLD signal responses on the other hand. (iii) The variant involvement of CBV in the hemodynamic responses of two different regions may introduce such negative correlations.


March 13, 2012

Tobias Block, PhD
Assistant Professor of Radiology
Center for Advanced Imaging Innovation and Research
New York University Langone Medical Center, NY

Iterative Reconstruction Concepts for Magnetic Resonance Imaging

Iterative reconstruction techniques are currently getting popular in the MRI community because they enable to reconstruct images from highly incomplete data, which can be exploited to skip acquisition steps and, thus, to reduce the scan time. This talk will first give a step-by-step introduction to the reconstruction technique and demonstrate how the technique can be applied for MRI data. The second part will discuss four application examples to illustrate the advantages over conventional methods. These advantages arise from two main components that inherently compensate for incompletely sampled data: First, the ability to incorporate prior knowledge about the object and, second, the ability to extend the signal modelling for advanced pulse sequences and acquisition techniques.

In the first example, it is shown that the higher sampling requirement for radial k-space sampling can be ameliorated with a constraint on the solution's total variation (TV), based on the assumption that many objects are piece-wise constant to some degree. Further, by extending the signal model to account for varying sensitivities of the receive coils, all channels can be processed simultaneously in a parallel imaging manner. In example 2, the concept is extended for radial fast spin-echo imaging where spokes with increasing T2 weighting are acquired along the echo train. When adding a spatial relaxation component to the signal calculation, the iterative approach is able to model these contrast inconsistencies and renders a proton-density map and a relaxation map directly from k-space, which can be used for fast T2 quantification.

In example 3, the signal model is extended to calculate the coil sensitivities jointly with the image content during the reconstruction, which offers improved parallel-imaging quality. Because in this way all sampled data is included for estimation of the coil profiles instead of only few reference lines, the method yields artifact-free images in conditions where conventional parallel-imaging reconstructions already show spurious aliasing artifacts. Finally, the last example combines the above ideas with a temporal constraint on sequentially acquired time frames. For measurements with an optimized radial real-time sequence, the technique achieves temporal resolutions of up to 20 ms and yields cinematic insight into the human body.


March 20, 2012

Jason P. Lerch, PhD
Assistant Professor of Medical Biophysics
Hospital for Sick Children
University of Toronto, Canada

Why do our brains differ so much? Using mouse imaging to understand variations in neuroanatomy in normal development and autism

A striking feature of any imaging study is just how much variability there is in brain shapes and sizes. This is especially the case in autism spectrum disorders, where the heterogeneity of the disease has resulted in a plethora of conflicting findings. In this talk I will use brain imaging in the mouse, where we have much tighter control over genetics and the environment, to illustrate both how different genetic mutations related to autism can lead to similar behavioural outcomes yet divergent neuroanatomical alterations as well as how the environment, learning, and memory can themselves change local brain shape.


March 27, 2012

Els Fieremans, PhD

Assistant Professor of Radiology
The Bernard and Irene Schwartz Center for Biomedical Imaging
New York University Langone Medical Center, NY


Assessment of white matter microstructural integrity with non Gaussian diffusion MRI

Diffusion MRI is a powerful tool to characterize brain white matter microstructural and architectural tissue organization. Diffusional kurtosis imaging (DKI) is a clinical feasible diffusion MRI method that quantifies the non-Gaussian diffusion properties in biological tissue through estimation of the diffusional kurtosis. In this talk, I will present a specific white matter model that allows for a direct physical interpretation of the non-Gaussian signal in terms of specific white matter microstructural integrity metrics, such as the axonal water fraction and intra- and extra-axonal compartmental diffusivities.

Next, I will discuss how these white matter integrity markers may serve as specific and sensitive biomarkers useful to study both healthy development and a variety of pathological conditions. In particular, our initial findings in human ischemic stroke and Alzheimer's disease illustrate how investigating changes in these white matter metrics reveal new insights in the underlying pathophysiology.


April 23, 2012

Hanzhang Lu, PhD
Associate Professor
Advanced Imaging Center
University of Texas Southwestern Medical Center, TX

A turn-key solution for the measurement of brain oxygen metabolism

We propose a procedure to measure global CMRO2 by combining several non-invasive measures obtained from MRI and pulse oximetry. A key technique of this procedure is a T2-Relaxation-Under-Spin-Tagging (TRUST) technique for the determination of global venous oxygenation. The TRUST MRI technique applies the spin labeling principle on the venous side and acquires control and labeled images, the subtraction of which yields pure venous blood signal. T2 value of the pure venous blood was then determined using non-selective T2-preparation pulses, minimizing the effect of flow on T2 estimation. Further technical considerations were made by using composite RF pulses and RF phase cycling in the T2-preparation.

We have measured Yv in both superior sagittal sinus (SSS) and internal jugular vein (IJV). Both measures yielded results consistent with expected venous values (50-75%) and, furthermore, a strong correlation was observed between them (P=0.0015), which is in agreement with the drainage path of venous blood. CMRO2 was estimated using TRUST and phase-contrast. Studies of intra-session and inter-session reproducibility of the CMRO2 measurement were conducted in seven subjects (26.4±4.0 years, 3 males and 4 females) and each subject underwent 5 sessions on different days. Intra-session and inter-session Coefficient of Variation (CoV) was 2.8±1.3% and 5.9±1.6%, respectively, suggesting a high reproducibility of this technique.

The dependence of CMRO2 on age was evaluated in our recent study. Average CMRO2 of typical 20-year-old subjects is approximately 164.1 µmol/100g/min and it increases with age at a rate of 2.6µmol/100g/min per decade, suggesting a reduced brain energy efficiency with age. We have also studied CMRO2 in an early stage of Alzheimer's Disease (AD) called Mild Cognitive Impairment (MCI) (Clinical Dementia Rating, CDR=0.5). In collaboration with the UTSW Alzheimer's Disease Center, we recruited 18 MCI patients (age 67±7 years) and 19 elderly controls (68±7 years). It was found that CMRO2 in MCI patients was 151.3±26.4 µmol/100g/min (mean±SD), which was significantly lower (P=0.04) than that of the control group (171.2±29.6 µmol/100g/min), suggesting that CMRO2 may be a sensitive marker for Alzheimer's Disease


June 19, 2012

Dung Minh Hoang
Doctoral Candidate
University of Lyon-1, France

The Filling Factor Redefined: Determining the Dominant Sensitivity Driver of a Flat Histology Slide RF Coil

This work investigates the relative gain in sensitivity of a set of five histology coils designed in-house compared to a circularly polarized (CP) mouse head birdcage coil (L=29-mm x ID=28-mm). The dimensions of these coils were tailored to fit tissue sections ranging from 5-µm to 100-µm when mounted on either standard glass slides and/or coverslips. Our simulations and experimental measurements demonstrate that the sensitivity of this flat structure underperforms by a factor of two relative to the CP birdcage coil based on the expected gain in their filling factor ratios. Despite the inevitable dielectric losses attributed to this capacitor-like shape resonator, our results demonstrate that the overall net increase in filling factor overcomes the current leaks inherent to this structure.

Surprisingly, this leads to an enhancement in sensitivity of up to seven-fold for the smallest structure constructed (W=12-mm x L=24-mm x H=0.45-mm). Alternatively, the largest histology coil design (W=52-mm x L=48-mm x H=1.35-mm) enables two times wider radiofrequency flat coverage at equal sensitivity to the CP birdcage. Examples of tissue sections from both mouse organs and human specimens acquired during overnight experiments illustrate the level of detail observed and the near-perfect co-registration with optical microscopy.


September 11, 2012

Steven Baete, PhD
Post-Doctoral Fellow
The Bernard and Irene Schwartz Center for Biomedical Imaging
New York University Langone Medical Center, NY

(Dynamic) Multiple Echo Diffusion Tensor Acquisition Technique (MEDITATE) in a 3T clinical scanner

Diffusion tensor imaging (DTI) provides biomarkers of tissue anisotropy and microstructure (principal diffusivities, mean diffusivity (MD) and fractional anisotropy (FA)), which have many applications in oriented biological tissue (e.g. neural fibers, renal tubules, muscle fibers). One route of acceleration of the multidirectional sampling required for DTI is multiple echoes. This presentation will describe progress in our use of this strategy to construct a dynamic DTI acquisition mode.

The first part of this talk will cover the feasibility of a two-scan Multiple Echo Diffusion Tensor Acquisition Technique (MEDITATE) on a clinical system for muscle DTI. In the MEDITATE-sequence, a pattern of diffusion gradients between the multiple RF-pulses encodes a train of echoes with each a different diffusion weighting and direction sufficient to estimate the 3D diffusion tensor. The work presented in this talk extends the original MEDITATE-approach, previously employed in preclinical settings, by exploiting longitudinal magnetization storage to reduce T2-weighting and optimizing a two-shot full tensor encoding within the clinical scanner hardware regime. Spin-warp phase encoding is used for image encoding. MEDITATE was tested on isotropic (agar gel) and anisotropic diffusion phantoms (asparagus), and in vivo skeletal muscle in healthy volunteers with cardiac-gating. Good quantitative agreement was found between diffusion eigenvalues, mean diffusivity, and fractional anisotropy derived from standard twice-refocused spin echo (TRSE) EPI-DTI and from several varieties of the MEDITATE sequence.

When combined with appropriate k-space trajectories or single voxel acquisition strategies, the accelerated encoding approach of MEDITATE may be used in clinical applications requiring time-sensitive acquisition of DTI parameters such as dynamical DTI in muscle. In that spirit, the second part of this talk will address the measurement of the exercise response of DTI biomarkers in skeletal muscle using dynamic MEDITATE, currently implemented using a line-scan image encoding approach. Finally, future plans and applications of the MEDITATE technique will be discussed.


September 17, 2012

J. Thomas Vaughan, PhD
Professor of Radiology, Electrical and Biomedical Engineering
Center for Magnetic Resonance Research
University of Minnesota

RF Safety (SAR and RF Heating) in MRI

The Radiofrequency (RF) transmit signal which stimulates the MR image signal, also deposits RF energy in the body resulting in heating. Because this RF heating can potentially result in pain, thermogenic tissue damage, and/or thermal stress to the human body, it must be better understood, predicted and monitored. Current MR safety practices however largely ignore tissue temperature as a safety metric in favor of the specific absorption rate (SAR) of RF energy deposition predicted from simple "standard" models of the human anatomy.

The problems with this "SAR" approach to RF safety are: 1. SAR by itself is not the cause of safety concerns, temperature is. 2.) SAR alone indicates neither the location nor the magnitude of thermal hot spots or overall body temperature. SAR based safety models consider only the electrodynamics, but not the thermodynamics or the physiology of humans being scanned. SAR is but one of six or more parameters in bioheat equations needed to predict temperature.

Modeling SAR only is therefore insufficient for predicting RF safety. By basing our safety metric on temperature rather than SAR however, we can not only be more safe, but in many cases we can safely use more RF power in our MRI scan protocols. This presentation will explore and explain SAR, RF Heating, and means to predict, monitor, and control them for MRI.


September 25, 2012

Leslie R. Yan
Doctoral Candidate
The Sackler Institute of Graduate Biomedical Sciences
New York University School of Medicine, NY

Imaging characteristics of posttraumatic stress disorder

Posttraumatic stress disorder is a prevalent psychiatric disorder in civil population and especially among combat veterans. The present study is about imaging characteristics of posttraumatic stress disorder in combat veterans. The neural characteristics of combat veterans who passed the diagnostic criteria of posttraumatic stress disorder were compared with those of combat veterans who did not met the diagnostic criteria. The neural substrates are characterized by MRI in terms of amplitudes of spontaneous activity, temporal synchronization of spontaneous activity, properties and architecture of the neural networks. Results have suggested valuable characteristics such as spontaneous activity in the insula and precuneus, temporal synchronization between the amygdala and prefrontal cortex, disorganization of neural networks etc.


October 23, 2012

Walter Schneider, PhD
Professor of Psychology and Neurosurgery
The Bernard and Irene Schwartz Center for Biomedical Imaging
University of Pittsburgh, PA

Quantifying TBI White Matter Damage in Individual Patients with High Definition Fiber Tracking (HDFT)

High Definition Fiber Tracking (HDFT) enables noninvasive MRI diffusion tracking of millimeter tracts over long distances accurately following from source to destination through tract crossings to detail axon projection fields of white matter tracts. Connection disorders are a major medical problem impacting tens of millions of patients with trauma (TBI), neuro-oncology, neurodegeneration (Alzheimer's) and developmental (autism) pathologies. HDFT involves mapping a million microtracts on a single individual with 3T MRI 257d DSI imaging with novel computation methods calculating directional axonal volume (dAV), tractography, and tract segmentation..

It creates a circuit diagram of the patient quantifying and visualizing the integrity of twenty brain white matter tracts. In a group TBI study the method produced high discriminant validity diagnosis of the anatomical basis of TBI showing nearly all TBI cases have visually and statistically clear damage to multiple tracts in mild TBI that was generally not detectable by previous methods. This provides the potential of definitive anatomically diagnosis of mild TBI and a foundation for a new ecology of personalized care and rehabilitation management.


November 13, 2012

Elfar Adalsteinsson, PhD
Associate Professor of Health Sciences and Technology
Athinoula A. Martinos Center for Biomedical Imaging
New York University Langone Medical Center, NY

Simulation studies of parallel transmit arrays under local and global SAR constraints

Despite intense research in pTx hardware development there has been relatively little theoretical evaluation or optimization of pTx coil arrays, for example determining the benefit of increasing number of transmit channels. We quantify the performance of three pTx body arrays with 4, 8 and 16 channels by incorporating simultaneous constraints on global and local SAR as well as average and maximum forward input power. We analyze RF shimming and 2 spokes excitations in the torso at 3T and compare the tradeoff between excitation fidelity, pulse power metrics and local and global SAR.


November 27, 2012

William Wu
Doctoral Candidate
The Sackler Institute of Graduate Biomedical Sciences
New York University School of Medicine, NY

Quantitative 3D Multivoxel Proton-MR Spectroscopy of In Vivo Cerebral Metabolism in a Rhesus Macaque Model of HIV-Associated Neurocognitive Disorders

Growth in rare and expensive animal models of human disease has increased interest in non-destructive evaluation of tissue injury and/or response to therapy. Proton magnetic resonance spectroscopy (1H-MRS) is a valuable tool because of its unique ability to probe cellular metabolism and bioenergetics noninvasively and nondestructively. However, 1H resonances from metabolites of interest (other than water) typically occur in vivo at 104–105 orders of magnitude lower concentrations than water, leading to much lower sensitivity. To overcome this limitation, we utilize a three dimensional (3D) multivoxel 1H MRS technique, which localizes multiple tissue regions simultaneously, and collect spectra from hundreds of ‹‹1 cm3 voxels. Compared with single-voxel techniques, 3D 1H MRS benefits from improved (~15×) signal-to-noise ratio and higher spectral resolution. Acquiring 3D 1H MRS together with high resolution MRI may provide a quantitative, long-term solution to costly, invasive and destructive histology studies, and improve diagnostic sensitivity and specificity.

Over 50% of the million Americans infected with HIV will suffer milder, long-term HIV associated neurocognitive disorders (HAND). 1H MRS has proven valuable in detecting brain abnormalities in HAND patients, and in simian immunodeficiency virus (SIV) infected rhesus macaques, an excellent model system. Prior histology has demonstrated neuronal dysfunction in (sub)cortical gray and white matter, as well as glial activation. Based on these observations, we test the hypothesis that decreased N acetylaspartate, the MRS-observed marker for neuronal integrity, and increased glial markers: myo-inositol, choline and creatine, can be detected with 3D 1H-MRS both globally and regionally—in subcortical structures—using SIV-infected rhesus macaques.


Latest Updates

05/23/2017 - 14:55
05/19/2017 - 16:45

Philanthropic Support

We gratefully acknowledge generous support for radiology research at NYU Langone Medical Center from:
• The Big George Foundation
• Raymond and Beverly Sackler
• Bernard and Irene Schwartz

Go to top