AI Body Composition Analysis from CT Predicts Mortality and Cardiovascular Risk

An AI tool measuring muscle, fat, and bone on abdominal CT scans predicted 5-year mortality with an AUC of 0.72 for all-cause death and 0.75 for major adverse cardiovascular events, outperforming traditional risk factors in a retrospective cohort analysis.

• Retrospective study using AI to extract body composition metrics from abdominal CT scans; exact sample size not specified in the provided source snippet.
• Key secondary finding: The AI-derived body composition measures added incremental prognostic value beyond standard clinical risk assessment (exact figures not reported in source).
• Limitation: Retrospective design; findings may not yet be externally validated, and generalizability to non-CT populations is unproven.

RadPigeon summaries are original and for information only. They are not clinical advice.