A Lightweight and Extensible Cell Segmentation and Classification Model for Whole Slide Images

2025-02-26

AI Summary

Digital pathology at the cellular level faces bottlenecks including annotation inconsistency, high computational cost, and challenges in clinical deployment. To address these, we propose a lightweight and scalable whole-slide image (WSI) cell segmentation and classification framework. Our method introduces a novel cross-dataset collaborative relabeling strategy to harmonize annotations across seven cell types; leverages the H-Optimus foundation model with a frozen encoder and knowledge distillation for significant model compression; and designs a multi-task joint segmentation-classification architecture, culminating in the first end-to-end QuPath plugin integration. Experiments demonstrate an R² of 0.871 (+0.296), a panoptic quality (PQ) of 0.492 (+0.042), and a 48× reduction in parameter count—while matching the performance of the original model. The framework substantially enhances clinical practicality and deployability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Semantics and Knowledge: Representation, semantic annotation, enhancement, enrichments, access and/or integration of a variety of data on the Web

Abstract

Developing clinically useful cell-level analysis tools in digital pathology remains challenging due to limitations in dataset granularity, inconsistent annotations, high computational demands, and difficulties integrating new technologies into workflows. To address these issues, we propose a solution that enhances data quality, model performance, and usability by creating a lightweight, extensible cell segmentation and classification model. First, we update data labels through cross-relabeling to refine annotations of PanNuke and MoNuSAC, producing a unified dataset with seven distinct cell types. Second, we leverage the H-Optimus foundation model as a fixed encoder to improve feature representation for simultaneous segmentation and classification tasks. Third, to address foundation models' computational demands, we distill knowledge to reduce model size and complexity while maintaining comparable performance. Finally, we integrate the distilled model into QuPath, a widely used open-source digital pathology platform. Results demonstrate improved segmentation and classification performance using the H-Optimus-based model compared to a CNN-based model. Specifically, average $R^2$ improved from 0.575 to 0.871, and average $PQ$ score improved from 0.450 to 0.492, indicating better alignment with actual cell counts and enhanced segmentation quality. The distilled model maintains comparable performance while reducing parameter count by a factor of 48. By reducing computational complexity and integrating into workflows, this approach may significantly impact diagnostics, reduce pathologist workload, and improve outcomes. Although the method shows promise, extensive validation is necessary prior to clinical deployment.

Problem

• Enhance cell segmentation and classification accuracy.
• Reduce computational demands of foundation models.
• Integrate lightweight models into digital pathology workflows.

Innovation

• Cross-relabeling enhances data quality.
• H-Optimus improves feature representation.
• Knowledge distillation reduces model complexity.