Tracking multiple sclerosis lesions in spinal cord MRI over time
Longitudinal tracking of multiple sclerosis lesions in the spinal cord: A validation study
Computer Vision and Pattern Recognition
Summary
Multiple sclerosis lesions in the spinal cord can change over time by appearing, disappearing, splitting, or merging, making it hard to track them with usual methods. The authors compared five ways to automatically track individual lesions using MRI scans taken at different times. One method, based on aligning images and measuring how much lesions overlap, worked best at correctly identifying which lesions matched across scans. This study helps improve understanding of lesion changes over time and compares methods to do this more reliably.
What this means in practice
- •For clinical radiologists: Use improved lesion tracking methods to monitor spinal cord MS lesion changes over time for patient assessment.
- •For medical imaging software developers: Incorporate the best-performing registration-based overlap technique to enhance longitudinal lesion tracking tools for MS in spinal cord MRI.
Authors
Pierre-Louis Benveniste, Julian McGinnis, Shannon Kolind, Larry D. Lynd, Sarah A. Morrow, Jiwon Oh, Alexandre Prat, Alice Schabas, Penelope Smyth, Roger Tam, Anthony Traboulsee, Mark Mühlau, Herve Lombaert, Julien Cohen-Adad
Abstract
Longitudinal characterization of multiple sclerosis (MS) lesions remains constrained by the lack of frameworks capable of establishing consistent instance-level correspondences across time. Conventional segmentation approaches produce semantic lesion masks at each visit and therefore fail to capture the complex instance temporal patterns associated with lesion appearance, disappearance, splitting, or merging. This study presents a comparative evaluation of five strategies for automated tracking of spinal cord MS lesions in longitudinal MRI data from a multi-site cohort. The investigated strategies rely either on deformable registration or on a spinal anatomical reference system, and encompass overlap-based matching, coordinate-based Hungarian algorithm, gradient-boosted classification, and Siamese model classification. Tracking accuracy is quantified using instance-level true positives, false positives, and false negatives, allowing to assess the presence of one-to-many and many-to-one associations. Results show best performance for the registration-based overlap method. This study provides the first systematic analysis of lesion-instance correspondence in the spinal cord and outlines the strengths and limitations of registration-based and registration-free paradigms for longitudinal MS assessment. The code is available at http://github.com/ivadomed/longitudinal-sc-ms-lesion-tracking .