RSNA spotlights HeartLung.AI study tying CT muscle quality to future COPD risk

RSNA featured a HeartLung.AI study showing that AI-measured myosteatosis on coronary CT scans predicted later COPD more strongly than an emphysema-like lung measure. The 20-year MESA analysis suggests routine cardiac CT could double as an opportunistic screening tool for earlier disease risk detection.

Why it matters: - The study suggests existing coronary artery calcium CT scans may reveal future COPD risk years before diagnosis. - AI-derived muscle-quality measurements could add a new screening signal without ordering a separate exam. - The findings point to COPD as a broader cardiopulmonary and metabolic disease, not only a lung disorder.

What happened: - RSNA featured HeartLung.AI’s study in an August 6 news article. - The research appeared in Radiology: Cardiothoracic Imaging, an RSNA journal. - Investigators analyzed baseline coronary artery calcium CT scans from 5,535 participants in the Multi-Ethnic Study of Atherosclerosis. - The cohort was followed for about 20 years. - During follow-up, 396 participants, or 7.1%, were diagnosed with COPD.

The details: - HeartLung.AI’s AI-CVD platform automatically quantified thoracic skeletal muscle attenuation to identify myosteatosis, a CT marker of fatty infiltration and reduced muscle quality. - The same platform also measured an emphysema-like lung signal from the same CT scans. - AI-CVD analyzed visible thoracic muscles across the full scan volume, rather than a single manually selected image or region of interest. - Participants in the lowest quartile of muscle quality had a 2.74-fold higher adjusted risk of developing clinically diagnosed COPD than those in the highest quartile. - That risk estimate held after adjustment for age, sex, smoking status, body mass index, race, asthma, physical activity, inflammatory markers and insulin resistance. - The hazard ratio was 2.74 for myosteatosis and 1.50 for the emphysema-like measurement after multivariable adjustment. - The myosteatosis association stayed consistent across subgroups defined by age, sex, obesity, smoking history and physical activity. - Coronary calcium CT is usually performed to assess coronary atherosclerosis and cardiovascular risk. - The same scan can also surface quantitative information about other organs and disease pathways when analyzed with AI. - RSNA’s coverage highlights the potential of quantitative AI analysis to identify disease vulnerability before symptoms or advanced disease appear.

Between the lines: - The result strengthens the case for opportunistic screening, where one scan can support multiple preventive insights. - The muscle-quality signal may capture systemic disease biology that lung-only imaging misses. - The comparison is not perfect, because coronary calcium CT does not fully image the upper lungs where emphysema can begin. - That limitation helps explain why the muscle measure and emphysema-like measure may not perform the same way. - HeartLung.AI founder and president Morteza Naghavi said AI can measure subtle findings such as muscle fat infiltration reproducibly from CT images even when routine visual review is impractical.

What's next: - The investigators said additional validation is needed before myosteatosis can be used as a clinical COPD biomarker. - Future studies are expected to test the finding in additional populations. - Researchers also plan to examine whether muscle-quality changes precede declines in pulmonary function. - Another open question is whether interventions that improve muscle quality can reduce future COPD risk.

The bottom line: - A routine heart CT may carry early warning signs for COPD, and AI could make those hidden signals measurable at scale. - More information: RSNA News Feature and Radiology: Cardiothoracic Imaging study