Fairness-Aware Fine-Tuning of Vision-Language Models for Medical Glaucoma Diagnosis
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910047824510976 |
|---|---|
| author | Gu, Zijian Liu, Yuxi Zhang, Zhenhao Wang, Song |
| author_facet | Gu, Zijian Liu, Yuxi Zhang, Zhenhao Wang, Song |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_03477 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Fairness-Aware Fine-Tuning of Vision-Language Models for Medical Glaucoma Diagnosis Gu, Zijian Liu, Yuxi Zhang, Zhenhao Wang, Song Computer Vision and Pattern Recognition Machine Learning 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. |
| title | Fairness-Aware Fine-Tuning of Vision-Language Models for Medical Glaucoma Diagnosis |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2512.03477 |