AI Enhancements in Coronary Inflammation Detection from CT Scans

Recent advancements in artificial intelligence (AI) technology have demonstrated promising capabilities in identifying high cardiovascular risk among patients with low or zero coronary artery calcium (CAC) scores. A study showcased at the Society of Cardiovascular Computed Tomography (SCCT) conference highlighted the potential benefits of using AI to analyze coronary computed tomography angiography (CCTA) scans.

The study centered around the AI software CaRi-Heart from Caristo Diagnostics, which automates the measurement of coronary inflammation using a Fat Attenuation Index (FAI) Score. The research involved a review of paired CCTA and non-contrast CT data from 19,092 patients with a mean age of 56 over a median follow-up of 6.7 years. The AI tool calculates the average FAI Score from the evaluation of three coronary arteries, thereby offering a metric for assessing coronary inflammation.

Findings from the study indicated that among patients with CAC scores below 10, 127 individuals (1.2%) experienced cardiovascular death. Those within the top decile for FAI Score on CCTA exhibited a nearly 13-fold heightened risk of cardiovascular mortality compared to those below the 25th percentile. Similarly, patients in the top decile for the FAI Score on non-contrast CT had an eightfold increased risk.

Kenneth Chan, MBBS, a Clinical Research Fellow in Cardiology at the University of Oxford, led the study. Dr. Chan emphasized the importance of identifying patients at high risk despite having low or zero CAC scores, stating that coronary inflammation can serve as a significant marker. He noted that patients with CAC scores under 100 account for a considerable portion of cardiovascular events, highlighting the need for refined risk stratification.

The study also marks a novel achievement in obtaining an automated assessment of coronary inflammation from non-contrast CT scans, which could extend the scope of early screening and risk assessment. According to Dr. Chan, non-contrast scans offer an economically viable primary prevention strategy, eliminating the need for contrast administration and potentially broadening access to cardiac diagnostics.

Dr. Chan, who received the Young Investigator Award at the conference, sees this technological advancement as enhancing the role of cardiac CT in evaluating cardiovascular risk. By providing additional data alongside CAC scores, the AI tool offers a more comprehensive view of patient risk profiles.

The study underscores the need for improved tools and strategies in cardiovascular risk identification and supports the ongoing development of AI applications in medical imaging for better patient outcomes. The potential for AI-driven insights from standard CT procedures could revolutionize how cardiovascular risks are assessed, particularly in asymptomatic patients, elevating both preventive and management strategies in clinical practice.