基于报告衍生放射学观察的可审计CT表型分析

中文摘要 AI 辅助

医学图像基础模型可从计算机断层扫描（CT）预测临床表型，但其优异性能无法明确模型是读取疾病特异性发现还是与诊断相关的捷径。本研究采用基于报告衍生放射学观察的可审计CT表型分析（ACT）方法，在221项电子健康记录（EHR）表型中验证上述问题。研究以38317名患者训练ACT，挖掘376194项观察结果，并在25183名保留患者中进行评估。ACT在零样本标注上优于5个视觉语言基线，在零样本评分（0.651 vs 0.572）和线性探测（0.709 vs 0.662）两项指标上，均优于CT-CLIP在221项未见过的CT肺血管造影表型上的表现。读取每个探测结果可揭示准确率所掩盖的信息：221个排名第一的位置仅由97项观察结果占据，描述主动脉和冠状动脉钙化的一个短语在20项表型中排名第一，包括骨质疏松、尿路感染和重度抑郁症。将观察库限制为临床医生指定的证据，可将这些探测结果重定向到86项表型的相关观察结果，且准确率未受影响（0.751 vs 0.741）。因此，基于CT的准确EHR表型分析可能依赖于编码表型的无效证据，而ACT可识别并干预此类情况。

英文摘要

Medical image foundation models can predict clinical phenotypes from computed tomography (CT), but strong performance leaves open whether they read disease-specific findings or shortcuts that correlate with the diagnosis. We tested this in 221 electronic-health-record (EHR) phenotypes using Auditable CT phenotyping (ACT), built on report-derived radiological observations. We trained ACT on 38,317 patients, mined 376,194 observations and evaluated it in 25,183 held-out patients. ACT exceeded five vision-language baselines on zero-shot annotation, and CT-CLIP across 221 phenotypes from unseen CT pulmonary angiography, both under zero-shot scoring (0.651 versus 0.572) and under linear probing (0.709 versus 0.662). Reading each probe exposes what accuracy conceals: only 97 observations occupy the 221 rank-1 positions, and one phrase describing aortic and coronary calcification ranks first for 20 phenotypes, including osteoporosis, urinary tract infection and major depressive disorder. Restricting the bank to clinician-specified evidence redirects those probes onto phenotype-related observations in 86 phenotypes at no accuracy cost (0.751 versus 0.741). Accurate CT-based EHR phenotyping can therefore rest on observations that are not valid evidence for the coded phenotype and that ACT can identify and intervene on.

发表机构

• University of Pennsylvania（宾夕法尼亚大学）
• Technical University of Munich（慕尼黑工业大学）
• University Hospital Essen（埃森大学医院）

机构由 AI 辅助整理，请以论文原文为准。