发表机构
Tsinghua University; Imperial College London; Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University（清华大学; 帝国理工学院; 浙江大学医学院附属邵逸夫医院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文指出纵向临床AI的基于事件评估可能夸大早期检测性能，通过定义识别前转变、独立参考标准和识别代理，使识别前检测声明可检验。
AI 中文摘要
临床上有用的早期检测需要经过验证的识别前提前时间。然而，基于事件的纵向临床AI评估可能将识别介导的护理过程信号视为捷径，并将依赖识别的终点作为参考标准，从而夸大表观性能和提前时间，同时削弱跨中心可移植性。此类结果可能服务于预后，但并未确立识别前的检测。我们定义了一个区间删失的识别前转变、一个独立的截至参考标准和一个预先指定的识别代理，以使该声明可检验。
英文摘要
Clinically useful early detection requires validated pre-recognition lead time. Yet event-based evaluations of longitudinal clinical AI can treat recognition-mediated care-process signals as shortcuts and recognition-dependent endpoints as reference standards, inflating apparent performance and lead time while undermining cross-center transport. Such results may serve prognosis without establishing detection before recognition. We define an interval-censored pre-recognition transition, an independent as-of reference standard, and a prespecified recognition proxy to make the claim testable.
Comments27 pages, 1 figure, 3 tables, 1 box; includes Supplementary Note
