GPT-4 showed comparable performance to traditional models for 10-year cardiovascular risk prediction, achieving an AUROC of 0.725 vs 0.733 (ACC/AHA) in UK Biobank and 0.664 vs 0.674 in KoGES.
Observational (n=53,186)
Yes
Does GPT-4 provide comparable 10-year cardiovascular risk prediction to traditional models in diverse populations?
GPT-4 demonstrates comparable performance to established ACC/AHA and Framingham risk scores for 10-year cardiovascular risk prediction across diverse cohorts, even with missing variables.
Effect estimate: AUROC 0.725 (UKB), 0.664 (KoGES)
Cardiovascular disease (CVD) remains a pressing global health concern. While traditional risk prediction methods such as the Framingham and American College of Cardiology/American Heart Association (ACC/AHA) risk scores have been widely used in the practice, artificial intelligence (AI), especially GPT-4, offers new opportunities. Utilizing large scale of multi-center data from 47,468 UK Biobank participants and 5,718 KoGES participants, this study quantitatively evaluated the predictive capabilities of GPT-4 in comparison with traditional models. Our results suggest that the GPT-based score showed commendably comparable performance in CVD prediction when compared to traditional models (AUROC on UKB: 0.725 for GPT-4, 0.733 for ACC/AHA, 0.728 for Framingham; KoGES: 0.664 for GPT-4, 0.674 for ACC/AHA, 0.675 for Framingham). Even with omission of certain variables, GPT-4's performance was robust, demonstrating its adaptability to data-scarce situations. In conclusion, this study emphasizes the promising role of GPT-4 in predicting CVD risks across varied ethnic datasets, pointing toward its expansive future applications in the medical practice.
Han et al. (Wed,) conducted a observational in Cardiovascular disease (n=53,186). GPT-4 vs. ACC/AHA and Framingham risk scores was evaluated on 10-year cardiovascular risk prediction (AUROC) (AUROC 0.725 (UKB), 0.664 (KoGES)). GPT-4 showed comparable performance to traditional models for 10-year cardiovascular risk prediction, achieving an AUROC of 0.725 vs 0.733 (ACC/AHA) in UK Biobank and 0.664 vs 0.674 in KoGES.