让医学视觉语言模型更公平：降低不同人群诊断差异
Fairness-Aware Fine-Tuning of Vision-Language Models for Medical Glaucoma Diagnosis
- 用可微分的极值准确率差距损失，端到端优化不同群体诊断公平性
- 在1万张眼底图上，差距减少69%，整体准确率达53.15%
- 仅需0.24%可训练参数，适合资源有限的医疗场景
视觉语言模型在医学影像任务中已达到专家水平，但在不同人口群体间存在显著诊断准确率差异。本文提出面向医疗VLM的公平性感知低秩适应方法，结合参数效率与显式公平性优化。核心算法是可微分的MaxAccGap损失，实现跨群体准确率均等的端到端优化。提出三种方法：FR-LoRA将MaxAccGap正则化融入训练目标，GR-LoRA采用逆频率加权平衡梯度贡献，Hybrid-LoRA融合两者机制。在10,000张青光眼眼底图像上评估，GR-LoRA使诊断准确率差异降低69%，同时保持53.15%的整体准确率。消融实验表明，强正则化强度可实现最优公平性且精度损失最小，种族特异性优化可降低60%差异。该方法仅需0.24%可训练参数，适用于资源受限的医疗AI部署。
原文摘要 · Abstract (English)
Vision-language models achieve expert-level performance on medical imaging tasks but exhibit significant diagnostic accuracy disparities across demographic groups. We introduce fairness-aware Low-Rank Adaptation for medical VLMs, combining parameter efficiency with explicit fairness optimization. Our key algorithmic contribution is a differentiable MaxAccGap loss that enables end-to-end optimization of accuracy parity across demographic groups. We propose three methods: FR-LoRA integrates MaxAccGap regularization into the training objective, GR-LoRA applies inverse frequency weighting to balance gradient contributions, and Hybrid-LoRA combines both mechanisms. Evaluated on 10,000 glaucoma fundus images, GR-LoRA reduces diagnostic accuracy disparities by 69% while maintaining 53.15% overall accuracy. Ablation studies reveal that strong regularization strength achieves optimal fairness with minimal accuracy trade-off, and race-specific optimization yields 60% disparity reduction. Our approach requires only 0.24% trainable parameters, enabling practical deployment of fair medical AI in resource-constrained healthcare settings.
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。