Optimize Difference (OpDiff): Contrast Enhancement Ordinary Differential Equations for Sub-Second Liver DCE-MRI Reconstruction

Motivation: Achieving high temporal resolution in continuous liver DCE-MRI is highly desirable to capture rapid early changes in contrast enhancement. Goal(s): Reconstruct DCE-MRI images with high temporal resolution while minimizing streaking artifacts and noise. Approach: DCE-MRI reconstruction was modeled via solving the inverse problem for the Ordinary Differential Equations (ODEs) that reproduce measured data. Moreover, Fourier basis functions were used to represent the contrast enhancement rate, enforcing low frequency prior and continuity. Results: Our method reconstructs DCE-MRI images with high temporal resolution and reduced streaking artifacts and noise. Impact: This novel image acquisition and reconstruction approach could improve the accuracy of early post-contrast phase timing for liver DCE-MRI.