Efficient Fine-Tuning of DINOv3 Pretrained on Natural Images for Atypical Mitotic Figure Classification (MIDOG 2025 Task 2 Winner)
This work addresses the challenge of detecting abnormal cell divisions in medical imaging for improved prognosis, though it is incremental as it applies existing methods to a new domain.
The paper tackled the problem of classifying atypical mitotic figures in histopathology images, which is difficult due to low prevalence and subtle morphology, by fine-tuning DINOv3-H+ with LoRA and achieving first place on the MIDOG 2025 benchmark.
Atypical mitotic figures (AMFs) represent abnormal cell division associated with poor prognosis. Yet their detection remains difficult due to low prevalence, subtle morphology, and inter-observer variability. The MIDOG 2025 challenge introduces a benchmark for AMF classification across multiple domains. In this work, we fine-tuned the recently published DINOv3-H+ vision transformer, pretrained on natural images, using low-rank adaptation (LoRA), training only ~1.3M parameters in combination with extensive augmentation and a domain-weighted Focal Loss to handle domain heterogeneity. Despite the domain gap, our fine-tuned DINOv3 transfers effectively to histopathology, reaching first place on the final test set. These results highlight the advantages of DINOv3 pretraining and underline the efficiency and robustness of our fine-tuning strategy, yielding state-of-the-art results for the atypical mitosis classification challenge in MIDOG 2025.