A General Pipeline for Glomerulus Whole-Slide Image Segmentation

TL;DR

This paper introduces a versatile pipeline for glomerulus segmentation in whole-slide images, improving detection accuracy and setting new benchmarks through comprehensive evaluation on large datasets.

Contribution

The proposed pipeline enhances patch-level and WSI-level segmentation by stitching overlapping patches, and is validated across diverse datasets with superior performance.

Findings

Outperforms previous state-of-the-art methods.

Achieves higher detection coverage near patch borders.

Sets new benchmark for glomerulus segmentation.

Abstract

Whole-slide images (WSI) glomerulus segmentation is essential for accurately diagnosing kidney diseases. In this work, we propose a general and practical pipeline for glomerulus segmentation that effectively enhances both patch-level and WSI-level segmentation tasks. Our approach leverages stitching on overlapping patches, increasing the detection coverage, especially when glomeruli are located near patch image borders. In addition, we conduct comprehensive evaluations from different segmentation models across two large and diverse datasets with over 30K glomerulus annotations. Experimental results demonstrate that models using our pipeline outperform the previous state-of-the-art method, achieving superior results across both datasets and setting a new benchmark for glomerulus segmentation in WSIs. The code and pre-trained models are available at https://github.com/huuquan1994/wsi_glomerulus_seg.