Accelerated first-pass perfusion MRI using BLOSM: Evaluation using dynamic simulations and patient datasets with prominent respiratory motion

k-t FOCUSS with ME/MC and k-t SLR. MSE for 26 magnitude-valued datasets (A) and for 12 complexvalued datasets (B) showed that BLOSM achieved the minimum error. BLOSM also achieved the lowest error at all acceleration rates (C). (* P<0.01 BLOSM vs. the other methods, ANOVA; ** P<0.01 BLOSM vs. all the other methods, paired Student’s t-test) Figure 2. Example images at two time points (rows 1 and 2) from one patient’s perfusion CMR. The heart moved at the 2nd row (E-H) due to breathing. BLOSM (B,F) outperformed k-t FOCUSS with ME/MC (C,G) and k-t SLR (D,H), closely matching the fully-sample images (A,E) with the presence or absence of motion. X-t profiles demonstrating similar results are shown on the bottom row, with important dynamic features highlighted by red arrows. Figure 1. Reconstruction of retrospectively rate-4 undersampled images and x-t profiles from computersimulated phantoms. Phantom 1 (P1) undergoes rigid translational shifts along the phase-encoding direction. P2 has an abrupt change in size as well as appearance and disappearance of features to mimic through plane motion combined with translational shifts in the readout direction. P3 undergoes rigid translational shifts along the readout direction. P4 undergoes a gradual change in size which can be interpreted as either cardiac contraction or through-plane motion. P5 rotates counterclockwise. BLOSM (column B) provided the most accurate recovery of the fully sampled images. For k-t FOUCSS with ME/MC (column D) and kt SLR (column E) residual artifacts and motion blurring can be observed. 4108