Computational framework for structuring and analyzing clinical trial criteria for AI-guided fine-grained matching

Download PDF here

This study introduces a computational framework to structure and analyze eligibility criteria across three real-world clinical trial protocols, supporting more fine-grained AI-driven trial matching. Criteria were decomposed into individual variables and evaluated by data type, scope, and interdependency. Trials varied widely, containing 22–160 eligibility variables, with 4–22% dependent on other criteria. Reading grade levels ranged from sixth grade to college level, and complexity scores differed substantially. These findings highlight the need for standardized, computable approaches to improve transparency and scalability in AI-driven clinical trial recruitment.

Recommended citation: Habib DRS, Mahajan I, Evancha B, Micheel C, Fabbri D. Computational framework for structuring and analyzing clinical trial criteria for AI-guided fine-grained matching. J Med Sys. 2025;49:168. doi:10.1007/s10916-025-02303-y