Transformer-based multiclass segmentation pipeline for basic kidney histology

Current applications of deep learning in renal pathology focused on anatomical structures with morphology, yet little research has focused on the performance of models, such as versatility, in regions with severe kidney damage. In this study, we explored the difference in modal/domain shift capabili...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Scientific reports Ročník 15; číslo 1; s. 38957 - 14
Hlavní autoři:	He, Junling, Valkema, Pieter A., Long, Jingmin, Li, Jia, Florquin, Sandrine, Naesens, Maarten, Koshy, Priyanka, Nguyen, Tri Q., Meziyerh, Soufian, de Vries, Aiko P. J., De Boer, Onno J., Verbeek, Fons J., Xiong, Zhan, Kers, Jesper
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	London Nature Publishing Group UK 06.11.2025 Nature Publishing Group Nature Portfolio
Témata:	639/166 692/699/1585 Architecture Arteries Artificial intelligence Biopsy Classification Deep Learning Fibrosis Histology Humanities and Social Sciences Humans Image Processing, Computer-Assisted - methods Inflammation Innovations Kidney - diagnostic imaging Kidney - pathology Kidney diseases Kidney histology Kidneys Lesions Medical imaging Multiclass segmentation multidisciplinary Neural Networks, Computer Pathology Science Science (multidisciplinary) Segmentation Semantic segmentation Semantics Sensitivity analysis Splitting Transformer Deep learning Kidney histology Transformer Semantic segmentation Multiclass segmentation
ISSN:	2045-2322, 2045-2322
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Current applications of deep learning in renal pathology focused on anatomical structures with morphology, yet little research has focused on the performance of models, such as versatility, in regions with severe kidney damage. In this study, we explored the difference in modal/domain shift capabilities between CNN-based and Transformer-based models. Firstly, we adopted two splitting strategies—WSI-level and patch-level—to stimulate sampling on multiple modal data distribution (i.e., renal WSIs collected from multi-centers). Then, we trained multiple CNN- and Transformer-based models on each splitting scheme respectively. We compared cross-splitting performance and analyzed the effective factors of results. For further validation, all models were tested on an independent external dataset for sensitivity analysis on the degree of fibrosis and inflammation. In conclusion, at both the patch- and WSI-level, M2F-Swin-B substantially outperformed UNet-ResNet18 with an average Intersection over Union (A-IoU) and per-class IoU. Notably, M2F-Swin-B outperformed UNet-ResNet18 in areas of a higher degree of fibrosis and inflammation and, a higher IoU score of arteries. In this study, we developed a robust multi-class segmentation pipeline for kidney histology. Moreover, we showed that the attention mechanism in Mask2Former enables visibly crisper and more uniform segmentation, particularly when the data is inadequate.
Bibliografie:	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23
ISSN:	2045-2322 2045-2322
DOI:	10.1038/s41598-025-22814-5