It has been reported that a new sequence, epi-SPEEDI, can offer sub-millisecond temporal resolution, but its scan time is relatively long due to phase-encoding. In this study we propose an accelerated version of epi-SPEEDI (named epi-SPEEDI-kt) that improves the acquisition efficiency of epi-SPEEDI by randomly undersampling the k-space, followed by image reconstruction using joint spatiotemporal partial separability and sparsity constraints. Through a human imaging example for visualizing the dynamics of aortic valve, we demonstrated that epi-SPEEDI-kt can provide comparable image quality to epi-SPEEDI and reduce the scan time by ~50% concurrently. This acceleration is expected to enhance SPEEDI applications.
This abstract and the presentation materials are available to members only; a login is required.