HeartLung.AI study on AI-predicted AF and stroke featured on Stroke Alert Podcast

HeartLung.AI says its research on AI-derived heart chamber measurements from routine coronary calcium scans was featured on the June 2026 Stroke Alert Podcast. The study found those measurements improved long-term prediction of atrial fibrillation and stroke in 6,812 participants from MESA and Framingham.

Why it matters: - HeartLung.AI’s work suggests routine, noncontrast coronary artery calcium scans may do more than measure calcium burden. - AI-extracted chamber measurements from the same scan could help identify people at higher long-term risk for atrial fibrillation and stroke. - The findings point to a lower-friction screening approach because the imaging test is already widely used and does not require contrast or a separate scan.

What happened: - HeartLung.AI’s article, "AI-Derived LA Volume Index, LA/RA and LA/LV Volume Ratios From Coronary Artery Calcium Scans Predict Long-Term Atrial Fibrillation and Stroke," was featured on the June 2026 episode of the Stroke Alert Podcast. - Stroke Alert Podcast is the podcast of Stroke, the journal of the American Heart Association/American Stroke Association. - The podcast segment summarized the study for the stroke and vascular neurology community.

The details: - Investigators pooled participant-level data from the Multi-Ethnic Study of Atherosclerosis and the Framingham Heart Study Offspring cohort. - The analysis included 6,812 participants total: 5,670 from MESA and 1,142 from Framingham. - Participants were followed for a median of about 17 years. - During follow-up, 1,302 participants developed atrial fibrillation and 365 experienced stroke. - Participants in the highest 5% of AI-derived chamber metrics had significantly higher long-term risk of both outcomes. - Left atrial volume index was the strongest AI-derived predictor of future atrial fibrillation. - The left atrial-to-right atrial ratio was the strongest AI-derived predictor of stroke. - Adding these imaging biomarkers improved risk reclassification beyond CHARGE-AF for atrial fibrillation and the Framingham Stroke Risk Profile for stroke. - The article was published in Stroke. - The full article DOI is available here. - The podcast episode is available here.

Between the lines: - The study fits a broader push toward opportunistic imaging, where scans done for one reason yield additional risk information. - HeartLung.AI’s core thesis is that AI can unlock clinical value from chest CT and CAC scans beyond traditional single-purpose reporting. - Morteza Naghavi, founder of HeartLung.AI and senior author of the study, said the podcast feature recognizes the work and the broader opportunity to use AI to reveal hidden clinical value from scans already being performed. - Naghavi said a CAC scan should not be viewed only as a calcium score and can also reveal chamber remodeling linked to future atrial fibrillation and stroke. - The authors said the findings are meant to support improved risk prediction and hypothesis generation, not to define a new clinical standard on their own.

What’s next: - The authors said more research is needed before AI-derived chamber measurements from CAC scans can be added to clinical workflows. - Future work will need to test how the metrics fit into preventive care, rhythm monitoring strategies and stroke prevention programs. - The podcast feature may help drive broader clinical and research interest in AI-enabled opportunistic imaging.

The bottom line: - A standard CAC scan may contain more prognostic information than the calcium score alone, and AI may be able to extract part of that value for AF and stroke risk prediction.