A deep learning framework for glomeruli segmentation with boundary attention
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915939378790400 |
|---|---|
| author | Elhaminia, Behnaz King, Catherine Lv, Jiaqi Harper, Lorraine Moss, Paul Cain, Owen Chanouzas, Dimitrios Raza, Shan E Ahmed |
| author_facet | Elhaminia, Behnaz King, Catherine Lv, Jiaqi Harper, Lorraine Moss, Paul Cain, Owen Chanouzas, Dimitrios Raza, Shan E Ahmed |
| contents | Accurate detection and segmentation of glomeruli in kidney tissue are essential for diagnostic applications. Traditional deep learning methods primarily rely on semantic segmentation, which often fails to precisely delineate adjacent glomeruli. To address this challenge, we propose a novel glomerulus detection and segmentation model that emphasises boundary separation. Leveraging pathology foundation models, the proposed U-Net-based architecture incorporates a specialised attention decoder designed to highlight critical regions and improve instancelevel segmentation. Experimental evaluations demonstrate that our approach surpasses state-of-the-art methods in both Dice score and Intersection over Union, indicating superior performance in glomerular delineation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_14263 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | A deep learning framework for glomeruli segmentation with boundary attention Elhaminia, Behnaz King, Catherine Lv, Jiaqi Harper, Lorraine Moss, Paul Cain, Owen Chanouzas, Dimitrios Raza, Shan E Ahmed Tissues and Organs Computer Vision and Pattern Recognition Machine Learning Accurate detection and segmentation of glomeruli in kidney tissue are essential for diagnostic applications. Traditional deep learning methods primarily rely on semantic segmentation, which often fails to precisely delineate adjacent glomeruli. To address this challenge, we propose a novel glomerulus detection and segmentation model that emphasises boundary separation. Leveraging pathology foundation models, the proposed U-Net-based architecture incorporates a specialised attention decoder designed to highlight critical regions and improve instancelevel segmentation. Experimental evaluations demonstrate that our approach surpasses state-of-the-art methods in both Dice score and Intersection over Union, indicating superior performance in glomerular delineation. |
| title | A deep learning framework for glomeruli segmentation with boundary attention |
| topic | Tissues and Organs Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2604.14263 |