Figure 1
From: Accelerated first-pass perfusion CMR using compressed sensing with regional spatiotemporal sparsity
Example images at different time points and temporal profiles from in vivo patient perfusion CMR. Fully sampled data (a-d) serve as a reference and time points are pointed out on profiles(d). The proposed regional sparsity method (i-l) outperformed kt-SLR (e-h) at rate 4 acceleration. More residual artifacts were found on kt-SLR, where the difference was most obvious at t3 when large motion occurred (k v.s. g). The temporal profiles also show that using regional sparsity better recovered the motion than kt-SLR, as highlighted by red arrows.