Towards Automated Banff Lesion Scoring: Tissue Segmentation in Kidney Transplant Biopsies using Deep Learning

Sebastiaan Ram, Dominique van Midden, Jeroen van der Laak, Linda Studer
Proceedings of the MICCAI Workshop on Computational Pathology, PMLR 316:248-265, 2026.

Abstract

Inflammation and chronic changes in the different tissue structures (e.g., glomeruli, tubuli, interstitium) are major contributors to kidney transplant failure. Kidney transplant biopsy diagnostics is based on the Banff classification system, in which pathologists assess these changes. However, many of these factors have suboptimal reproducibility and the scoring is labor-intensive. To address this, we developed a multi-class segmentation approach that covers all tissue structures relevant for diagnostics. Our dataset comprises 99 Periodic-acid Schiff (PAS)-stained kidney transplant biopsy slides from two pathology departments. An expert pathologist manually annotated >17,000 structures across eight classes (glomeruli, sclerotic glomeruli, empty Bowman space, proximal tubuli, distal tubuli, atrophic tubuli, capsule, arteries/arterioles, and interstitium). We compared two segmentation approaches: (1) a combination of two nnU-Nets (one for tissue segmentation and one specialized for structure boundary detection) and (2) the SAM-Path foundation model. For the peritubular capillary segmentation, we used a previously developed U-Net. The nnU-Nets achieved a per-class average Dice score of 0.80, outperforming SAM-Path (0.69) and providing a reliable solution for all tissue structures relevant for kidney transplant biopsy diagnostics. Next, the nnU-Nets will be used in a reader study aimed at investigating the impact of AI on pathologists’ performance in Banff lesion scoring. The algorithm is publicly available on Grand Challenge.

Cite this Paper

BibTeX
@InProceedings{pmlr-v316-ram26a, title = {Towards Automated Banff Lesion Scoring: Tissue Segmentation in Kidney Transplant Biopsies using Deep Learning}, author = {Ram, Sebastiaan and Midden, Dominique van and Laak, Jeroen van der and Studer, Linda}, booktitle = {Proceedings of the MICCAI Workshop on Computational Pathology}, pages = {248--265}, year = {2026}, editor = {Studer, Linda and Ciompi, Francesco and Khalili, Nadieh and Faryna, Khrystyna and Faryna, Khrystyna and Yeong, Joe and Lau, Mai Chan and Chen, Hao and Liu, Ziyi and Brattoli, Biagio}, volume = {316}, series = {Proceedings of Machine Learning Research}, month = {27 Sep}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v316/main/assets/ram26a/ram26a.pdf}, url = {https://proceedings.mlr.press/v316/ram26a.html}, abstract = {Inflammation and chronic changes in the different tissue structures (e.g., glomeruli, tubuli, interstitium) are major contributors to kidney transplant failure. Kidney transplant biopsy diagnostics is based on the Banff classification system, in which pathologists assess these changes. However, many of these factors have suboptimal reproducibility and the scoring is labor-intensive. To address this, we developed a multi-class segmentation approach that covers all tissue structures relevant for diagnostics. Our dataset comprises 99 Periodic-acid Schiff (PAS)-stained kidney transplant biopsy slides from two pathology departments. An expert pathologist manually annotated >17,000 structures across eight classes (glomeruli, sclerotic glomeruli, empty Bowman space, proximal tubuli, distal tubuli, atrophic tubuli, capsule, arteries/arterioles, and interstitium). We compared two segmentation approaches: (1) a combination of two nnU-Nets (one for tissue segmentation and one specialized for structure boundary detection) and (2) the SAM-Path foundation model. For the peritubular capillary segmentation, we used a previously developed U-Net. The nnU-Nets achieved a per-class average Dice score of 0.80, outperforming SAM-Path (0.69) and providing a reliable solution for all tissue structures relevant for kidney transplant biopsy diagnostics. Next, the nnU-Nets will be used in a reader study aimed at investigating the impact of AI on pathologists’ performance in Banff lesion scoring. The algorithm is publicly available on Grand Challenge.} }
Endnote
%0 Conference Paper %T Towards Automated Banff Lesion Scoring: Tissue Segmentation in Kidney Transplant Biopsies using Deep Learning %A Sebastiaan Ram %A Dominique van Midden %A Jeroen van der Laak %A Linda Studer %B Proceedings of the MICCAI Workshop on Computational Pathology %C Proceedings of Machine Learning Research %D 2026 %E Linda Studer %E Francesco Ciompi %E Nadieh Khalili %E Khrystyna Faryna %E Khrystyna Faryna %E Joe Yeong %E Mai Chan Lau %E Hao Chen %E Ziyi Liu %E Biagio Brattoli %F pmlr-v316-ram26a %I PMLR %P 248--265 %U https://proceedings.mlr.press/v316/ram26a.html %V 316 %X Inflammation and chronic changes in the different tissue structures (e.g., glomeruli, tubuli, interstitium) are major contributors to kidney transplant failure. Kidney transplant biopsy diagnostics is based on the Banff classification system, in which pathologists assess these changes. However, many of these factors have suboptimal reproducibility and the scoring is labor-intensive. To address this, we developed a multi-class segmentation approach that covers all tissue structures relevant for diagnostics. Our dataset comprises 99 Periodic-acid Schiff (PAS)-stained kidney transplant biopsy slides from two pathology departments. An expert pathologist manually annotated >17,000 structures across eight classes (glomeruli, sclerotic glomeruli, empty Bowman space, proximal tubuli, distal tubuli, atrophic tubuli, capsule, arteries/arterioles, and interstitium). We compared two segmentation approaches: (1) a combination of two nnU-Nets (one for tissue segmentation and one specialized for structure boundary detection) and (2) the SAM-Path foundation model. For the peritubular capillary segmentation, we used a previously developed U-Net. The nnU-Nets achieved a per-class average Dice score of 0.80, outperforming SAM-Path (0.69) and providing a reliable solution for all tissue structures relevant for kidney transplant biopsy diagnostics. Next, the nnU-Nets will be used in a reader study aimed at investigating the impact of AI on pathologists’ performance in Banff lesion scoring. The algorithm is publicly available on Grand Challenge.
APA
Ram, S., Midden, D.v., Laak, J.v.d. & Studer, L.. (2026). Towards Automated Banff Lesion Scoring: Tissue Segmentation in Kidney Transplant Biopsies using Deep Learning. Proceedings of the MICCAI Workshop on Computational Pathology, in Proceedings of Machine Learning Research 316:248-265 Available from https://proceedings.mlr.press/v316/ram26a.html.

Related Material