Adaptive Learning Strategies for Mitotic Figure Classification in MIDOG2025 Challenge
This work addresses the problem of reliable detection of clinically relevant indicators in pathology for medical professionals, but it is incremental as it builds on existing models and techniques.
The paper tackled the challenge of detecting atypical mitotic figures in pathology images by adapting the UNI2 foundation model with methods like Visual Prompt Tuning and stain normalization, achieving a balanced accuracy of 0.8837 and ROC-AUC of 0.9513 in the MIDOG2025 challenge.
Atypical mitotic figures (AMFs) are clinically relevant indicators of abnormal cell division, yet their reliable detection remains challenging due to morphological ambiguity and scanner variability. In this work, we investigated three variants of adapting the pathology foundation model UNI2 for the MIDOG2025 Track 2 challenge: (1) LoRA + UNI2, (2) VPT + UNI2 + Vahadane Normalizer, and (3) VPT + UNI2 + GRL + Stain TTA. We observed that the integration of Visual Prompt Tuning (VPT) with stain normalization techniques contributed to improved generalization. The best robustness was achieved by further incorporating test-time augmentation (TTA) with Vahadane and Macenko stain normalization. Our final submission achieved a balanced accuracy of 0.8837 and an ROC-AUC of 0.9513 on the preliminary leaderboard, ranking within the top 10 teams. These results suggest that prompt-based adaptation combined with stain-normalization TTA offers a promising strategy for atypical mitosis classification under diverse imaging conditions.