
Heart Failure and Atrial Fibrillation Prediction from Non-Gated Chest CT Scans Using AI-Based Cardiac Chamber Volumetry
A new AI-CVD study shows that routine, non-gated chest CT scans can provide more than anatomic information. Using AI-based cardiac chamber volumetry, the study evaluated whether left ventricular and left atrial size could identify individuals at increased risk of future heart failure and atrial fibrillation.
Study overview
The analysis was conducted within the Multi-Ethnic Study of Atherosclerosis (MESA), a large, multi-ethnic prospective cohort. The investigators used non-gated chest CT scans acquired for lung imaging and applied automated AI measurements of cardiac chamber volumes.
AI-derived left ventricular volume was evaluated in relation to incident heart failure.
AI-derived left atrial volume was evaluated in relation to incident atrial fibrillation.
Risk patterns were compared across quartiles of chamber volume.
Findings from non-gated lung scans were presented alongside cardiac ECG-gated scans.
Key findings
Participants with larger AI-derived left ventricular volumes had progressively higher cumulative incidence of heart failure. The same pattern was observed for left atrial volume and incident atrial fibrillation. The results were directionally consistent across gated cardiac scans and non-gated chest CT scans.
These findings suggest that cardiac chamber enlargement can be detected opportunistically on chest CT examinations that were not originally acquired for dedicated cardiac assessment.
Why this matters
Non-gated chest CT is performed frequently in clinical practice, including lung cancer screening and evaluation of pulmonary disease. Automated chamber volumetry could help extract additional cardiovascular risk information from these existing scans without requiring a separate examination.
The study supports the potential role of AI-enabled opportunistic imaging as an additional risk-stratification layer. Prospective validation and assessment of clinical workflow integration will be important before routine implementation.
Read the full article
Download the complete article using the button below.


























Comments