Mitosis detection in domain shift scenarios: a Mamba-based approach

2025-08-28

AI Summary

To address domain shift in histopathological images caused by staining variability and equipment differences, this work proposes a robust cross-domain mitosis detection framework. Methodologically, we introduce (1) VM-UNet—a segmentation model built upon the Mamba architecture—to enhance long-range dependency modeling in high-resolution pathology images; and (2) a targeted stain-style augmentation strategy to improve generalization and stability on unseen domains. Preliminary evaluation on the MIDOG++ dataset demonstrates substantial mitigation of cross-domain performance degradation, with superior detection robustness over baseline models. The approach has been submitted to Track 1 of the MIDOG 2024 Challenge. By enabling reliable mitosis detection across multi-center, multi-platform histopathology data, this work provides a scalable and transferable solution for automated pathological image analysis.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Mitosis detection in histopathology images plays a key role in tumor assessment. Although machine learning algorithms could be exploited for aiding physicians in accurately performing such a task, these algorithms suffer from significative performance drop when evaluated on images coming from domains that are different from the training ones. In this work, we propose a Mamba-based approach for mitosis detection under domain shift, inspired by the promising performance demonstrated by Mamba in medical imaging segmentation tasks. Specifically, our approach exploits a VM-UNet architecture for carrying out the addressed task, as well as stain augmentation operations for further improving model robustness against domain shift. Our approach has been submitted to the track 1 of the MItosis DOmain Generalization (MIDOG) challenge. Preliminary experiments, conducted on the MIDOG++ dataset, show large room for improvement for the proposed method.

Problem

• Detecting mitosis in histopathology images for tumor assessment
• Addressing performance drop in machine learning under domain shift
• Improving model robustness with Mamba-based approach and stain augmentation

Innovation

• Mamba-based approach for mitosis detection
• VM-UNet architecture for domain shift
• Stain augmentation for model robustness