A new AI-driven blood test, CardiOmicScore from HKU Medicine, can predict risk for six major cardiovascular diseases up to 15 years before symptoms appear in high-risk individuals.

The findings come from analyses of UK Biobank population data and were published in Nature Communications.

The test captures dynamic signals from immune activity, metabolism, and vascular health, enabling earlier lifestyle changes and preventive interventions.

CardiOmicScore outperforms traditional polygenic risk scores and shows additional gains when combined with basic clinical information like age and gender.

In clinical practice, the approach could offer patients and doctors early, multi-disease risk profiles from a simple blood test to guide prevention and monitoring.

Key researchers include Professor Zhang Qingpeng and first author Luo Yan from HKU IDS, with the HKU Medical researchers leading the study.

The test analyzes 2,920 circulating proteins and 168 metabolites from a single blood sample using deep learning to assess the current biological state beyond traditional risk factors.

This work marks a shift from fixed genetic risk estimates to proactive, prevention-focused precision medicine, aiming to generate comprehensive cardiovascular risk profiles from one blood test.

The six diseases covered are coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease, and venous thromboembolism.