Clinical Feasibility of Free-Breathing Dynamic T1-Weighted Imaging With Gadoxetic Acid–Enhanced Liver Magnetic Resonance Imaging Using a Combination of Variable Density Sampling and Compressed Sensing
The purpose of the study was to investigate the clinical feasibility of free-breathing dynamic T1-weighted imaging (T1WI) using Cartesian sampling, compressed sensing, and iterative reconstruction in gadoxetic acid–enhanced liver magnetic resonance imaging (MRI).Materials and Methods
This retrospective study was approved by our institutional review board, and the requirement for informed consent was waived. A total of 51 patients at high risk of breath-holding failure underwent dynamic T1WI in a free-breathing manner using volumetric interpolated breath-hold (BH) examination with compressed sensing reconstruction (CS-VIBE) and hard gating. Timing, motion artifacts, and image quality were evaluated by 4 radiologists on a 4-point scale. For patients with low image quality scores (<3) on the late arterial phase, respiratory motion-resolved (extradimension [XD]) reconstruction was additionally performed and reviewed in the same manner. In addition, in 68.6% (35/51) patients who had previously undergone liver MRI, image quality and motion artifacts on dynamic phases using CS-VIBE were compared with previous BH-T1WIs.Results
In all patients, adequate arterial-phase timing was obtained at least once. Overall image quality of free-breathing T1WI was 3.30 ± 0.59 on precontrast and 2.68 ± 0.70, 2.93 ± 0.65, and 3.30 ± 0.49 on early arterial, late arterial, and portal venous phases, respectively. In 13 patients with lower than average image quality (<3) on the late arterial phase, motion-resolved reconstructed T1WI (XD-reconstructed CS-VIBE) significantly reduced motion artifacts (P < 0.002–0.021) and improved image quality (P < 0.0001–0.002). In comparison with previous BH-T1WI, CS-VIBE with hard gating or XD reconstruction showed less motion artifacts and better image quality on precontrast, arterial, and portal venous phases (P < 0.0001–0.013).Conclusions
Volumetric interpolated breath-hold examination with compressed sensing has the potential to provide consistent, motion-corrected free-breathing dynamic T1WI for liver MRI in patients at high risk of breath-holding failure.