在检索增强的医疗问答系统中评估和缓解偏差
Yuelyu Ji1, Hang Zhang2, Yanshan Wang3,4,5
1Dept. of Information Science, University of Pittsburgh, Pittsburgh, PA, USA.
AMIA ... Annual Symposium proceedings. AMIA Symposium
|February 23, 2026
概括
医疗RAG系统显示人口偏差. 多数投票聚合提高了准确性和公平性,强调了在医疗保健中需要公平的AI.
科学领域:
- 人工智能在医学中的应用
- 自然语言处理自然语言处理.
- 医疗信息学 医疗信息学
背景情况:
- 检索增强生成 (RAG) 通过整合外部知识来增强医疗问答系统 (QA).
- 独立的大型语言模型 (LLM) 可能会出现不准确性,而RAG系统可能会无意中延续人口偏见 (种族,性别,社会经济因素).
研究的目的:
- 系统地评估医疗RAG管道中的人口偏见.
- 评估偏见缓解策略在提高公平性和准确性的有效性.
主要方法:
- 在多个质量保证基准 (MedQA,MedMCQA,MMLU,EquityMedQA) 中使用人口敏感查询进行评估.
- 实施和比较偏见缓解技术:思维链,反事实过,对立即提炼和多数投票聚合.
主要成果:
- 在检索一致性和答案正确性方面发现了显著的人口差异.
- 大多数选票聚合显示了准确性和公平性指标的改进.
结论:
- 医疗RAG系统表现出显著的人口偏差.
- 公平意识的检索和快速的工程对于开发公平的医疗质量保证系统至关重要.
相关概念视频
The Availability Heuristic
7.1K
A heuristic is a general problem-solving framework (Tversky & Kahneman, 1974). You can think of these as mental shortcuts that are used to solve problems. Different types of heuristics are used in different types of situations, and the impulse to use a heuristic occurs when one of five conditions is met (Pratkanis, 1989):
7.1K
Pharmacovigilance
1.8K
Post-marketing surveillance is a critical component of pharmaceutical regulation, often uncovering unanticipated adverse drug reactions (ADRs) once a drug is widely used over an extended period.
This process, termed pharmacovigilance, aims to detect, evaluate, and minimize harmful effects related to medication use. The data collection for pharmacovigilance depends on spontaneous reporting systems, where healthcare professionals or patients voluntarily report suspected ADRs.
In some cases, there...
This process, termed pharmacovigilance, aims to detect, evaluate, and minimize harmful effects related to medication use. The data collection for pharmacovigilance depends on spontaneous reporting systems, where healthcare professionals or patients voluntarily report suspected ADRs.
In some cases, there...
1.8K
Health Information Technology and Healthcare Information System
3.5K
Health Information Technology (HIT)
Health Information Technology, commonly called HIT, integrates advanced information systems and technology in healthcare settings. Its primary functions include:
Health Information Technology, commonly called HIT, integrates advanced information systems and technology in healthcare settings. Its primary functions include:
3.5K
Methods of Documentation V: CBE
1.5K
Charting by Exception, or CBE, is a method of documentation used in healthcare, particularly in nursing, that focuses on documenting only significant or abnormal findings rather than recording every detail. This approach aims to streamline the documentation process, improve efficiency, and ensure that healthcare providers can quickly identify deviations from normalcy in patient assessments.
In CBE, healthcare professionals establish predefined standards of practice that define what constitutes...
In CBE, healthcare professionals establish predefined standards of practice that define what constitutes...
1.5K
Errors occurring during blood pressure monitoring
1.6K
Blood pressure monitoring is a crucial clinical procedure in diagnosing and managing various cardiovascular conditions. Despite its significance, the accuracy of blood pressure measurements can be compromised by multiple factors, potentially leading to either falsely high or low readings. These inaccuracies are critical as they can significantly impact patient care. So, it is vital to understand these challenges deeply and adopt strategic approaches to minimize errors.
Several factors...
Several factors...
1.6K
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
500
Biopharmaceutical studies constitute a vital field aiming to enhance drug delivery methods and refine therapeutic approaches, drawing upon diverse interdisciplinary knowledge. In research methodologies, the choice between controlled and non-controlled studies significantly influences the study's reliability and accuracy.
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
500

