New AI Blood Test Predicts Stroke, Heart Failure, and More up to 15 Years in Advance

Key Points:

• Researchers at the University of Hong Kong have developed CardiOmicScore, an AI-powered blood test that predicts the risk of six major cardiovascular diseases up to 15 years before clinical symptoms appear by analyzing multiomics data.
• The test integrates genomics, metabolomics, and proteomics using deep learning to capture real-time molecular signals reflecting a person's current biological condition, improving early detection beyond traditional risk factors and fixed genetic scores.
• CardiOmicScore demonstrated significantly higher accuracy in forecasting coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease, and venous thromboembolism compared to conventional polygenic risk scores.
• This advancement represents a shift in precision medicine from static genetic risk models to dynamic, multiomics-based approaches, enabling earlier intervention and potentially transforming cardiovascular disease prevention and management.
• The study was published in Nature Communications and supported by multiple research grants from Hong Kong and China.