Spinal Cord Toolbox (SCT) - New release v7.3

Announcements
1

Dear neuroimaging community,

We are happy to announce the release of SCT v7.3.

The most notable changes for this release include:

  • Feature: Add deep learning-based motion correction algorithm (-dl flag) to sct_dmri_moco and sct_fmri_moco. View pull request
  • Feature: Implement length_anterior and length_posterior metrics in sct_process_segmentation. View pull request
  • Enhancement: Reduce memory usage in sct_apply_transfo when warping 4D images by avoiding unnecessary data copies. View pull request
  • Enhancement: Handle missing centerline slices in sct_compute_ascor and sct_process_segmentation by limiting segmentation metrics to only overlapping slices. View pull request
  • Bug: Fix excessive RAM usage and slow processing in sct_process_segmentation by resampling after cropping rather than before. View pull request
  • Bug: Fix blank QC report page by improving rendering performance for large reports. View pull request
  • Bug: Fix bug in sct_analyze_lesion where the vert_level column would sometimes be incorrectly included in the total % all_tracts sum when using -f -perslice 1. View pull request
  • Bug: Fix several bugs with the find_angle_hog function in sct_register_to_template to make computed HOG values consistent. View pull request
  • Bug: Avoid crash in sct_deepseg during inference if spine is present in the $SCT_DIR path. View pull request
  • Installation: Fix several bugs in sct.def to improve Apptainer installation (install directory, task installation). View pull request
  • Documentation: Update web tutorials to include changes to the slides from the 2025 SCT Course. View pull request

For the full list of changes included in this release, click here.

For installation instructions, click here.

If you have any questions or feature requests, please post in the SCT forum.

Happy processing!

The Spinal Cord Toolbox Team