评估AI辅助乳腺X线摄影判读中的认知偏差：一项跨经验水平放射科医生的可解释AI模拟阅片研究
该单中心模拟阅片研究纳入6名乳腺放射科医生，在无辅助、AI辅助及AI辅助联合基于显著性的可解释AI热图三种条件下独立判读200例乳腺X线摄影。在30%检查中AI的BI-RADS样建议被故意扰动一个类别。结果显示可解释AI可降低自动化偏差和锚定偏差，并提高非扰动子集的诊断准确率，但未消除偏差。
要点
- AI辅助可导致自动化偏差和锚定偏差，低年资放射科医生更易受影响。
- 基于显著性的可解释AI热图可显著降低偏差决策，并小幅提升诊断准确率。
- 将可解释AI与针对性培训嵌入AI辅助乳腺X线摄影工作流程可能提升患者安全。
核心结论
- 自动化和锚定偏差影响AI辅助乳腺X线摄影判读，尤其在经验较少的放射科医生中更明显；基于显著性的可解释AI减少但未消除这些效应。
研究设计与样本
- 单中心、全交叉模拟阅片研究，于2024年3月至6月开展，6名乳腺放射科医生按经验分层，独立判读200例乳腺X线摄影。
- 三种序贯条件：无辅助、AI辅助、AI辅助联合基于显著性的可解释AI热图；采用广义线性混合效应模型分析，考虑阅片者和病例层面聚类。
诊断标准与关键数值
- 在30%检查中，BI-RADS样AI建议被故意扰动一个类别以量化对误导性AI建议的易感性，其余70%保留原生AI输出。
- 无解释的AI辅助条件下，操纵病例中自动化偏差发生65/180（36.1%），锚定偏差61/180（33.9%）；联合可解释AI时分别降至32/180（17.8%）和31/180（17.2%）。
- 混合效应模型中，可解释AI与自动化偏差较低几率相关（aOR 0.56，95% CI 0.44-0.71；p < 0.001），与锚定相关修正偏差较低几率相关（aOR 0.61，95% CI 0.48-0.78；p < 0.001）。
- 非操纵子集诊断准确率从无辅助期的724/840（86.2%）提高至AI联合可解释AI期的757/840（90.1%）。
临床应用与阅片要点
- 日常AI辅助乳腺X线摄影判读中应警惕自动化偏差和锚定偏差，尤其低年资医生。
- 可解释AI热图有助于减少对AI建议的过度依赖和锚定修正偏差，但不应完全替代独立判读。
- 将可解释AI与针对性培训结合可能提升AI辅助判读的安全性和准确性。
局限性
- 单中心模拟阅片研究，样本为6名放射科医生和200例乳腺X线摄影，结论外推需谨慎。
- 研究采用模拟阅读场景，非真实临床工作流程，可能影响偏差表现。
以上解读依据论文摘要（全文未开放获取或许可不允许翻译转载）。
英文原文（摘要）· 点击展开
OBJECTIVES: To evaluate the impact of automation and anchoring bias in artificial intelligence (AI)-assisted mammography interpretation and to assess whether saliency-based explainable AI (XAI) mitigates these biases across radiologists of varying experience.
MATERIALS AND METHODS: In this monocentric, fully crossed simulation reader study conducted between March and June 2024, six breast radiologists stratified by experience independently reviewed 200 mammograms under three sequential conditions: unassisted, AI-assisted, and AI-assisted with saliency-based XAI heatmaps. To quantify susceptibility to misleading AI advice under controlled discordance conditions, BI-RADS-like AI recommendations were deliberately perturbed by one category in 30% of examinations, whereas the remaining 70% retained the native AI output. Bias outcomes were analyzed using generalized linear mixed-effects models accounting for reader- and case-level clustering.
RESULTS: In the AI-assisted condition without explanations, automation bias occurred in 65/180 (36.1%) and anchoring bias in 61/180 (33.9%) of manipulated cases. With XAI, these rates decreased to 32/180 (17.8%) and 31/180 (17.2%), respectively. In mixed-effects models, XAI was associated with lower odds of automation bias (aOR 0.56, 95% CI 0.44-0.71; p < 0.001) and anchoring-related revision bias (aOR 0.61, 95% CI 0.48-0.78; p < 0.001). On the non-manipulated subset, diagnostic accuracy improved from 724/840 (86.2%) in the unaided phase to 757/840 (90.1%) in the AI + XAI phase.
CONCLUSION: Automation and anchoring bias affected AI-assisted mammography interpretation, particularly among less experienced radiologists. Saliency-based explainable AI reduced, but did not eliminate, these effects.
KEY POINTS: Question AI assistance can systematically influence BI-RADS decisions in mammography, particularly among less experienced radiologists, through automation and anchoring biases. Findings Saliency-based explainable AI (XAI) substantially reduces biased decisions while modestly improving overall diagnostic accuracy compared with standard AI support alone. Clinical relevance Embedding XAI and targeted training into AI-assisted mammography workflows may enhance patient safety and support safer clinical integration of mammography AI tools.
引用
- Pesapane F, Latronico A, Abbate F, et al. Evaluating cognitive biases in AI-assisted mammography interpretation: a simulation reader study of explainable AI across radiologist experience levels. Eur Radiol. 2026. PMID: 42213113. doi:10.1007/s00330-026-12666-6
供执业医师工作参考，不替代临床判断；数值与分级以指南原文为准。