Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914286468595712 |
|---|---|
| author | Phitidis, Jesse Smithard, Alison Q. Whiteley, William N. Wardlaw, Joanna M. Bernabeu, Miguel O. Hernández, Maria Valdés |
| author_facet | Phitidis, Jesse Smithard, Alison Q. Whiteley, William N. Wardlaw, Joanna M. Bernabeu, Miguel O. Hernández, Maria Valdés |
| contents | White matter hyperintensities (WMH) and ischaemic stroke lesions (ISL) are imaging features associated with cerebral small vessel disease (SVD) that are visible on brain magnetic resonance imaging (MRI) scans. The development and validation of deep learning models to segment and differentiate these features is difficult because they visually confound each other in the fluid-attenuated inversion recovery (FLAIR) sequence and often appear in the same subject. We investigated six strategies for training a combined WMH and ISL segmentation model using partially labelled data. We combined privately held fully and partially labelled datasets with publicly available partially labelled datasets to yield a total of 2052 MRI volumes, with 1341 and 1152 containing ground truth annotations for WMH and ISL respectively. We found that several methods were able to effectively leverage the partially labelled data to improve model performance, with the use of pseudolabels yielding the best result. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_20503 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI Phitidis, Jesse Smithard, Alison Q. Whiteley, William N. Wardlaw, Joanna M. Bernabeu, Miguel O. Hernández, Maria Valdés Computer Vision and Pattern Recognition Artificial Intelligence White matter hyperintensities (WMH) and ischaemic stroke lesions (ISL) are imaging features associated with cerebral small vessel disease (SVD) that are visible on brain magnetic resonance imaging (MRI) scans. The development and validation of deep learning models to segment and differentiate these features is difficult because they visually confound each other in the fluid-attenuated inversion recovery (FLAIR) sequence and often appear in the same subject. We investigated six strategies for training a combined WMH and ISL segmentation model using partially labelled data. We combined privately held fully and partially labelled datasets with publicly available partially labelled datasets to yield a total of 2052 MRI volumes, with 1341 and 1152 containing ground truth annotations for WMH and ISL respectively. We found that several methods were able to effectively leverage the partially labelled data to improve model performance, with the use of pseudolabels yielding the best result. |
| title | Comparative evaluation of training strategies using partially labelled datasets for segmentation of white matter hyperintensities and stroke lesions in FLAIR MRI |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2601.20503 |