Lesion Instance Segmentation in Multiple Sclerosis: Assessing the Efficacy of Statistical Lesion Splitting
(2024) ISMRM & ISMRT Annual Meeting & Exhibition — Location: Singapore, Singapore (4.May.2024)
Files
ISMRM2023-Comparingsegmentationtoolsforlesioninstancesegmentation-1.pdf
Preprint
Open Access - Adobe PDF
- 236.09 KB
Details
- Authors
- Abstract
- Accurate white matter lesion (WML) counting and delineation are crucial for multiple sclerosis (MS) diagnosis and prognosis. Though being a critical step in clinical research and automated tools relying on lesion-centered patches, no previous work studied the effectiveness and correctness of current post-processing methods to transform voxel-wise segmentations into lesion instance masks in MS. In this study, we compare the conventional and straightforward connected components (CC) method to a confluent lesion splitting (CLS) method that was used but never validated. CC and CLS's performances are evaluated using three common lesion segmentation tools (LSTs): SPM, SAMSEG, and nnU-Net.
- Affiliations
Citations
Wynen, M., Gordaliza, P. M., Stölting, A., Maggi, P., Bach Cuadra, M., & Macq, B. (2024). Lesion Instance Segmentation in Multiple Sclerosis: Assessing the Efficacy of Statistical Lesion Splitting. ISMRM & ISMRT Annual Meeting & Exhibition, Singapore, Singapore. https://hdl.handle.net/2078.5/31663