WELLSTAR HEALTH SYSTEM

AI Lung Cancer Risk Model Sybil Validated in Predominantly Black Patient Population

- Researchers at the University of Illinois Hospital & Clinics validated the Sybil AI model's accuracy in predicting lung cancer risk within a predominantly Black patient population, addressing critical racial disparities in screening. - The deep learning model achieved an Area Under the Curve of 0.94 for one-year cancer risk prediction, demonstrating remarkable accuracy in a cohort where 62% of participants identified as Non-Hispanic Black. - The study analyzed 2,092 baseline low-dose CT screenings over a decade, with 68 patients subsequently diagnosed with lung cancer during follow-up periods extending up to 10.2 years. - The Sybil Implementation Consortium plans to initiate prospective clinical trials to integrate the AI tool directly into clinical workflows for real-world lung cancer screening programs.