CoFi: A Fast Coarse-to-Fine Few-Shot Pipeline for Glomerular Basement Membrane Segmentation

2025-08-15

AI Summary

To address the challenge of glomerular basement membrane (GBM) segmentation in electron microscopy images—where conventional pixel-level annotation is labor-intensive and existing few-shot methods fail to preserve fine structural details—this paper proposes a coarse-to-fine two-stage few-shot segmentation framework. In the first stage, a lightweight network generates initial segmentation masks; in the second stage, a morphology-aware pruning strategy automatically identifies high-quality point prompts to guide the Segment Anything Model (SAM) for refined segmentation. Crucially, the method eliminates manual prompt engineering. Evaluated with only three annotated images, it achieves a Dice coefficient of 74.54% and an inference speed of 1.9 FPS. By substantially reducing annotation burden and computational overhead, the framework establishes an efficient, clinically viable paradigm for quantitative GBM analysis.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Prompt Engineering / Prompting

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Accurate segmentation of the glomerular basement membrane (GBM) in electron microscopy (EM) images is fundamental for quantifying membrane thickness and supporting the diagnosis of various kidney diseases. While supervised deep learning approaches achieve high segmentation accuracy, their reliance on extensive pixel-level annotation renders them impractical for clinical workflows. Few-shot learning can reduce this annotation burden but often struggles to capture the fine structural details necessary for GBM analysis. In this study, we introduce CoFi, a fast and efficient coarse-to-fine few-shot segmentation pipeline designed for GBM delineation in EM images. CoFi first trains a lightweight neural network using only three annotated images to produce an initial coarse segmentation mask. This mask is then automatically processed to generate high-quality point prompts with morphology-aware pruning, which are subsequently used to guide SAM in refining the segmentation. The proposed method achieved exceptional GBM segmentation performance, with a Dice coefficient of 74.54% and an inference speed of 1.9 FPS. We demonstrate that CoFi not only alleviates the annotation and computational burdens associated with conventional methods, but also achieves accurate and reliable segmentation results. The pipeline's speed and annotation efficiency make it well-suited for research and hold strong potential for clinical applications in renal pathology. The pipeline is publicly available at: https://github.com/ddrrnn123/CoFi.

Problem

• Accurate GBM segmentation in EM images for kidney disease diagnosis
• Reducing annotation burden in supervised deep learning methods
• Capturing fine structural details in few-shot GBM analysis

Innovation

• Coarse-to-fine few-shot segmentation pipeline
• Lightweight network with three annotated images
• Morphology-aware pruning for SAM prompts