CP #4: Comprehensive Quantitative Ultrafast 3D liver MRI

Comprehensive quantitative ultrafast 3D liver MRI


Approximately 7% of MR exams performed are abdominal MRIs, and the liver is by far the most common abdominal organ imaged using MR. When successful, lesion characterization by MR is exquisite and elegant, and the definitive nature of the information can aid treatment follow-up, stop unnecessary workups and prevent invasive biopsies. All too often, however, liver MR is unsuccessful because of patient difficulties with breath-holds, problems with timing or acquisition in the critical (and non-repeatable) timed post-contrast image series, operator dependence, image artifacts, or differences in certainty between various radiologists arising from inherently qualitative interpretations. Weighted images are at best surrogates of the underlying parameter and often poorly reflect the parameters. Moreover, when quantitative information is available, it can outperform even expert readers of clinical contrasts. If the liver examination could be performed quickly during free-breathing, without relying on operator expertise, and could provide quantitative parameters which can be used to definitively diagnose disease, the impact on the diagnosis and treatment of liver diseases would be significant. 


In this work, we will leverage rapid imaging, parameter quantitation, and body MRI expertise in the MR research group at Case Western Reserve University (CWRU), amplified by the methods pioneered by the CAI2R team in New York, to provide a rapid, quantitative, high quality 3D exam in under 10 minutes of scan time. Figure 1: Liver perfusion images for a subject with a biopsy proven sclerosing hemangioma. (a) Clinical standard images acquired with long breath-hold of ~21 sec/volume. (b) Free-breathing images acquired with 1.9 sec/volume. (c) Liver perfusion maps. Arterial fraction for this subject was 99%, in agreement with literature on hemangiomas.


A standard liver MR exam consists of multiple scans highlighting different contrastmechanisms: T2 weighted scans without and with fat saturation, T1-weighted in- and opposed-phase gradient echo, diffusion images, and T1-weighted fat saturated 3D gradient echo scans pre- and at least 3 timed phases post-contrast. We will develop, optimize and validate methods to generate quantitative measurements of each mechanism, by devel¬oping quanti¬tative 4D high resolution DCE perfusion, fat fraction, T1 and T2 mapping, and improved 3D diffusion mapping. In the case of perfusion and fat fraction mapping, the goal is to achieve full 3D volumetric coverage in 0.5-2 s/volume, allowing accurate characterization of contrast dynamics and quantitative modeling of perfusion related parameters. Acceleration will be provided by combining undersampled non-Cartesian trajectories (radial and spiral) with non-Cartesian parallel imaging and compressed sens¬ing methods. The diagnostic value of the new protocol will then be extensively tested on patients with biopsy proven pathologies, and compared to the current clinical standard.

Push-Pull Relationship with TR&D Projects 1 & 2:

TR&D#1: This project has a natural and indeed nearly ideal fit with TR&D#1. A number of advanced acquisition and reconstruction methods developed at CWRU will be implemented in this CP [1-4]. Meanwhile, during visits to NYU and in various communications, we have been highly impressed by the performance of GRASP [5,6], L+S [7] , and related dynamic imaging approaches under development in New York, which we expect to provide complementary advantages. Thus, in addition to benefitting from advances in motion-robustness, etc as a result of CAI2R technologies, we will provide ongoing feedback to BTRC staff regarding comparative performance in our patient populations. Moreover, we are particularly interested in the combination of multiparametric imaging approaches such as our MR Fingerprinting (MRF) technique (4) with the paradigm of rapid continuous imaging, and we will push the BTRC to develop robust combinations in Specific Aims 2 and 3 of TR&D #1. TR&D#2: We are also quite interested in the multicoil extensions of MRF developed by BTRC staff member Dr. Cloos, and will stay apprised of body applications of plug and play pTx explored at NYU as part of TR&D #2. Finally, we are in a good position to take advantage of 3T body arrays developed by the BTRC RF engineering core group, since our acceleration methods rely in part on encoding by RF coil elements.

Principal Investigator: 


Latest Updates

05/10/2019 - 21:50
05/10/2019 - 21:23

Philanthropic Support

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

Go to top