DeNuC: Decoupling Nuclei Detection and Classification in Histopathology

2026-03-04

AI Summary

This work addresses the representational degradation and high computational cost commonly observed in pathology foundation models when jointly optimizing nucleus detection and classification. To overcome these limitations, we propose DeNuC, a novel approach that first uncovers the mechanism by which joint optimization leads to representation collapse and then mitigates it through task decoupling. Specifically, DeNuC employs a lightweight model for precise nucleus localization, followed by a coordinate-guided feature querying mechanism that leverages a pathology foundation model for accurate classification. Evaluated on the BRCAM2C and PUMA datasets, our method achieves F1-score improvements of over 4.2% and 3.6%, respectively, while using only 16% or fewer trainable parameters compared to existing approaches, demonstrating both superior performance and remarkable parameter efficiency.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Pathology Foundation Models (FMs) have shown strong performance across a wide range of pathology image representation and diagnostic tasks. However, FMs do not exhibit the expected performance advantage over traditional specialized models in Nuclei Detection and Classification (NDC). In this work, we reveal that jointly optimizing nuclei detection and classification leads to severe representation degradation in FMs. Moreover, we identify that the substantial intrinsic disparity in task difficulty between nuclei detection and nuclei classification renders joint NDC optimization unnecessarily computationally burdensome for the detection stage. To address these challenges, we propose DeNuC, a simple yet effective method designed to break through existing bottlenecks by Decoupling Nuclei detection and Classification. DeNuC employs a lightweight model for accurate nuclei localization, subsequently leveraging a pathology FM to encode input images and query nucleus-specific features based on the detected coordinates for classification. Extensive experiments on three widely used benchmarks demonstrate that DeNuC effectively unlocks the representational potential of FMs for NDC and significantly outperforms state-of-the-art methods. Notably, DeNuC improves F1 scores by 4.2% and 3.6% (or higher) on the BRCAM2C and PUMA datasets, respectively, while using only 16% (or fewer) trainable parameters compared to other methods. Code is available at https://github.com/ZijiangY1116/DeNuC.

Problem

• Nuclei Detection and Classification
• Pathology Foundation Models
• Representation Degradation
• Task Disparity
• Joint Optimization

Innovation

• Decoupling
• Nuclei Detection and Classification
• Pathology Foundation Models
• Representation Degradation
• Lightweight Localization