用少量参数微调大模型，高效识别罕见异常有丝分裂图像。
Efficient Fine-Tuning of DINOv3 Pretrained on Natural Images for Atypical Mitotic Figure Classification
- 仅训练130万参数，通过低秩适配微调自然图像预训练的DINOv3。
- 在多域数据上表现最优，准确率超越挑战赛其他方案。
- 适合病理图像少样本、跨域差异大的医学分析场景。
异常有丝分裂图像（AMFs）提示细胞异常分裂，与不良预后相关。其检测困难源于出现率低、形态细微及观察者间差异大。MItosis DOmain Generalization（MIDOG）2025挑战赛为此提供了跨域分类基准。本文对最近发布的、在自然图像上预训练的DINOv3-H+视觉变换器，采用低秩适配（LoRA）进行微调，仅训练130万参数。结合大规模数据增强与领域加权焦点损失，有效应对数据集中强烈的域异质性。尽管自然图像与组织病理学之间存在显著分布差异，微调后的DINOv3仍表现出良好迁移能力，在最终测试集上取得第一名。结果凸显了DINOv3预训练的优势，并验证了该微调策略在效率与鲁棒性上的优越性，实现了MIDOG 2025异常有丝分裂分类挑战的最先进性能。代码已公开于GitHub。
原文摘要 · Abstract (English)
Atypical mitotic figures (AMFs) indicate abnormal cell division associated with poor prognosis. Their detection remains difficult due to low prevalence, subtle morphology, and inter-observer variability. The MItosis DOmain Generalization (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. We combine this with extensive augmentation and a domain-weighted Focal Loss to better handle the strong domain heterogeneity in the dataset. Despite the large shift between natural images and histopathology, our fine-tuned DINOv3 transfers effectively, 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. Our code is publicly available on GitHub.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。