Search papers, labs, and topics across Lattice.
This study evaluates five automated strategies for longitudinal tracking of multiple sclerosis lesions in spinal cord MRI data, addressing the limitations of conventional segmentation methods that fail to capture complex temporal patterns of lesions. By comparing approaches that utilize deformable registration and anatomical reference systems, the research quantifies tracking accuracy through instance-level metrics, revealing that the registration-based overlap method outperforms others. This work represents the first systematic analysis of lesion-instance correspondence in spinal cord MS, highlighting both the strengths and weaknesses of different tracking paradigms.
Registration-based overlap methods significantly outperform other strategies in tracking multiple sclerosis lesions over time, revealing critical insights into lesion dynamics.
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 .
