eduzhai > Applied Sciences > Engineering >

Non-rigid 3D motion estimation at high temporal resolution from prospectively undersampled k-space data using low-rank MR-MOTUS

  • king
  • (0) Download
  • 20210505
  • Save

... pages left unread,continue reading

Document pages: 37 pages

Abstract: With the recent introduction of the MR-LINAC, an MR-scanner combined with aradiotherapy LINAC, MR-based motion estimation has become of increasinginterest to (retrospectively) characterize tumor and organs-at-risk motionduring radiotherapy. To this extent, we introduce low-rank MR-MOTUS, aframework to retrospectively reconstruct time-resolved non-rigid 3D+tmotion-fields from a single low-resolution reference image and prospectivelyundersampled k-space data acquired during motion. Low-rank MR-MOTUS exploitsspatio-temporal correlations in internal body motion with a low-rank motionmodel, and inverts a signal model that relates motion-fields directly to areference image and k-space data. The low-rank model reduces thedegrees-of-freedom, memory consumption and reconstruction times by assuming afactorization of space-time motion-fields in spatial and temporal components.Low-rank MR-MOTUS was employed to estimate motion in 2D 3D abdominothoracicscans and 3D head scans. Data were acquired using golden-ratio radial readouts.Reconstructed 2D and 3D respiratory motion-fields were respectively validatedagainst time-resolved and respiratory-resolved image reconstructions, and thehead motion against static image reconstructions from fully-sampled dataacquired right before and right after the motion. Results show that 2D+trespiratory motion can be estimated retrospectively at 40.8motion-fields-per-second, 3D+t respiratory motion at 7.6motion-fields-per-second and 3D+t head-neck motion at 9.3motion-fields-per-second. The validations show good consistency with imagereconstructions. The proposed framework can estimate time-resolved non-rigid 3Dmotion-fields, which allows to characterize drifts and intra and inter-cyclepatterns in breathing motion during radiotherapy, and could form the basis forreal-time MR-guided radiotherapy.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×