A deep learning framework for glomeruli segmentation with boundary attention

2026-04-15

AI Summary

This study addresses the challenge of accurately distinguishing individual glomeruli in kidney histopathology images, where blurred boundaries between adjacent glomeruli hinder precise instance-level segmentation using conventional semantic segmentation methods, thereby compromising pathological diagnosis. To overcome this limitation, the authors propose a U-Net architecture that integrates a foundation model for pathology with a boundary-aware attention mechanism. Specifically, a dedicated attention decoder is designed to enhance feature representation in boundary regions, effectively disentangling adherent glomeruli. The proposed method transcends the inherent constraints of semantic segmentation in instance-level tasks and achieves significant improvements over state-of-the-art approaches in key metrics such as Dice coefficient and Intersection over Union (IoU), thereby enhancing both the accuracy and robustness of glomerular segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

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.

Problem

• glomeruli segmentation
• boundary separation
• instance-level segmentation
• kidney tissue
• semantic segmentation

Innovation

• boundary attention
• glomeruli segmentation
• instance-level segmentation
• pathology foundation models
• U-Net