New AI Tool Detects Hidden Heart Disease Risk

A groundbreaking AI-driven biomarker, the fat attenuation index-score (FAI), has demonstrated its capability in assessing coronary inflammation, according to new research validated by the University of Oxford. Developed by Caristo Diagnostics, the CaRi-Heart technology analyzes CT scans to measure changes in the fat tissue surrounding coronary arteries. This FAI-Score biomarker can detect coronary inflammation from a non-contrast CT scan, a significant advancement for this imaging modality. The research, presented at the 2026 European Society of Cardiology congress, revealed that elevated FAI-Scores are linked to a substantially higher risk of cardiovascular mortality, even in patients with a zero calcium score.

Further analysis indicated that coronary inflammation, as measured by FAI-Score, predicted cardiovascular death in individuals without typical modifiable risk factors such as diabetes, hypertension, or high cholesterol. The predictive value was particularly strong in this group, highlighting the limitations of traditional risk assessments. Caristo's technology also shows promise in predicting stroke risk. Data from the ORFAN-MAESTRIA study showed that the technology identified signs of left atrial myopathy on routine cardiac CT, predicting future stroke risk independently of atrial fibrillation. This capability allows for the identification of high-risk patients who might otherwise show no warning signs on standard heart rhythm tests.