Can generative artificial intelligence empower target trial emulations?

作者：Zhouyu Guan, Dian Zeng, Huating Li, Tien Yin Wong, Bin Sheng · 发表于：The Lancet Digital Health · 年份：2026 · DOI：10.1016/j.landig.2025.100950 · 被引用次数：8 · 研究领域：Artificial Intelligence in Healthcare and Education、Ethics in Clinical Research、Meta-analysis and systematic reviews

Target trial emulation (TTE) is a pragmatic framework to estimate causal effects from observational data when randomised controlled trials (RCTs) are infeasible.1 While RCTs remain the gold standard for establishing causal inference between an exposure (eg, drug, device, or artificial intelligence [AI] algorithm) and a patient health outcome (eg, disease morbidity or mortality), RCTs are often constrained by high costs, long timelines, narrow eligibility criteria, ethical concerns, and limited generalisability to real-world populations.