The application of Compressed Sensing for Longitudinal MRI

Purpose: The mutual similarity of the follow-up scans in longitudinal studies is exploited on top of the well known sparse transform domains for rapid MRI by reducing the number of k-space measurements. Theory and Methods: A framework for adaptive Compressed Sensing (CS) MRI that exploits the redundancy of the acquired data in longitudinal studies is proposed. The baseline MR scan is utilized both in the sampling stage, with adaptive CS, and in the reconstruction stage, with 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: Experiments demonstrate that our longitudinal adaptive CS MRI (LACS-MRI) scheme provides reconstruction quality which outperforms traditional CS MRI for rapid MRI. Examples are shown on patients with brain tumors and demonstrate improved spatial resolution and accelerated acquisition for 2D and 3D brain imaging. Conclusions: Our method provides high quality MRI reconstruction, at 10-fold k-space undersampling. The proposed approach can play a major part and significantly reduce scanning time in many applications that consist of disease follow-up and monitoring of changes.