生物医学工程研究2025，Vol.44Issue(4)：229-237,9.DOI:10.19529/j.cnki.1672-6278.2025.04.05
针对医学大视觉语言模型的去偏见方法
Debiasing method for large vision-language models in the medical domain
摘要
Abstract
To address the bias issues caused by uneven data distribution in large vision-language models(LVLMs)for medical auxiliary diagnosis,we proposed a universal medical debiasing method called bias fairness enhancement(BFE).The effectiveness of BFE in mitigating bias was validated by constructing a benchmark to evaluate bias issues in medical LVLMs.This benchmark coverd bi-nary classification,multi-classification,and open-ended question tasks,while incorporating two parameters that control output ran-domness to ensure the method's generality.Experimental results demonstrated that BFE outperformed other mainstream debiasing meth-ods in the medical LVLMs(LLaVA-Med,SkinGPT).Notably,in open-ended medical question-answering tasks,LLaVA-Med's per-formance improved by 6.3%(cd_beta=0.1)and 6.7%(cd_beta=0.5).These findings indicate that BFE can effectively alleviate bias in medical LVLMs,and provide significant support for improving the accuracy and reliability of medical auxiliary diagnosis.关键词
大视觉语言模型/去偏见算法/医学辅助诊断/偏见问题/对比学习Key words
Large vision-language models/Debiasing algorithms/Medical assisted diagnosis/Bias issues/Contrastive learning分类
医药卫生引用本文复制引用
王文浩,韩忠义,王斌,吴慎京,魏本征..针对医学大视觉语言模型的去偏见方法[J].生物医学工程研究,2025,44(4):229-237,9.基金项目
国家自然科学基金项目(62372280,62402297) （62372280,62402297）
山东省自然科学基金项目(2024MF139,2023QF094) （2024MF139,2023QF094）
青岛市科技惠民示范专项(23-2-8-smjk-2-nsh). （23-2-8-smjk-2-nsh）
