Manuscript received June 12, 2024; revised January 10, 2025; accepted January 14, 2025; published March 17, 2025
Abstract—Diffusion Magnetic Resonance Imaging (dMRI) is
being increasingly used to study neural connectivity of brain
regions. High Angular Resolution Diffusion Imaging (HARDI)
reliably estimates the local orientations of white matter tracts
for neural connectivity analysis in the brain. However, HARDI
suffers from long acquisition times, limiting its clinical usage. An
effective way to reduce the acquisition time is to acquire an
undersampled signal followed by its compressive reconstruction.
In general, due to various calibration inaccuracies in the MR
scanners, acquisition trajectories in the k-space get perturbed.
Although radial sampling allows significant undersampling, the
trajectory errors are more pronounced with it. Hence, if the
trajectory errors are not corrected, compressive reconstruction
of the signal would be adversely impacted. In this work, we
propose ARTEC, a joint framework of accelerated
reconstruction of the HARDI signal undersampled in the joint
(k-q)-space, while incorporating trajectory error corrections.
Simulation results on both phantom and real data demonstrate
the superior performance of the proposed method over the
existing state-of-the-art methods.
Keywords—Diffusion MRI, HARDI compressed sensing,
trajectory errors, Spherical Ridgelets
Cite: Ashutosh Vaish, Anubha Gupta, and Ajit Rajwade, "ARTEC: Accelerated Reconstruction of High Angular Resolution Diffusion Imaging with Trajectory Error Correction," International Journal of Machine Learning vol. 15, no. 1, pp. 29-33, 2025.
Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).