A UNet pipeline for segmentation of new MS lesions
| dc.contributor.author | Efird, Cory | |
| dc.contributor.author | Miller, Dylan | |
| dc.contributor.author | Cobzas, Dana | |
| dc.date.accessioned | 2023-11-14T18:18:26Z | |
| dc.date.available | 2023-11-14T18:18:26Z | |
| dc.date.issued | 2021 | |
| dc.description.abstract | A pipeline for the second multiple sclerosis segmentation challenge (MSSEG-2) hosted by MICCAI is proposed. Two FLAIR images taken at different time-points are used as a multi-channel input to a 3D CNN to detect new lesions. Patch sampling strategies are adopted to keep the input volume shape manageable in terms of memory requirements. To further improve results, multiple models and patch orientations are ensembled. Performance is evaluated against nn-UNet. | |
| dc.identifier.citation | Cory Efird, Dylan Miller, Dana Cobzas. A UNet Pipeline for Segmentation of New MS Lesions. In MSSEG-2 challenge proceedings: Multiple sclerosis new lesions segmentation challenge using a data management and processing infrastructure. MICCAI 2021 - 24th International Conference on Medical Image Computing and Computer Assisted Intervention, Sep 2021, Strasbourg, France, pp. 53-56, https://hal.inria.fr/hal-03358968v3 | |
| dc.identifier.uri | https://hdl.handle.net/20.500.14078/3246 | |
| dc.language.iso | en | |
| dc.rights | All Rights Reserved | |
| dc.subject | multiple sclerosis | |
| dc.subject | segmentation | |
| dc.subject | deep learning | |
| dc.title | A UNet pipeline for segmentation of new MS lesions | en |
| dc.type | Presentation |
Files
Original bundle
1 - 1 of 1