Single Blood Test Predicts Heart Diseases 15 Years Before Onset

Cardiovascular diseases remain the leading cause of death worldwide, claiming nearly 19.8 million lives in 2022 alone. Early identification of individuals at risk is critical for prevention, yet conventional screening tools often fail to detect subtle biological changes before disease develops. Researchers have developed an artificial intelligence (AI)–driven blood test that predicts the risk of six major cardiovascular diseases up to 15 years before symptoms appear.

Researchers from the Department of Pharmacology and Pharmacy at the LKS Faculty of Medicine of The University of Hong Kong (Hong Kong, China have developed an AI-based cardiovascular risk prediction tool called CardiOmicScore, which integrates multiomics data from a single blood sample to forecast the likelihood of developing several major cardiovascular conditions. The CardiOmicScore model was built using deep learning methods that integrate large-scale multiomics datasets, including genomics, proteomics, and metabolomics.

The researchers analyzed population data from the UK Biobank, examining 2,920 circulating proteins and 168 metabolites obtained from blood samples. These molecular signals act as dynamic indicators of the body’s biological status, capturing real-time changes in immune function, metabolism, and vascular health. By combining these measurements with clinical information such as age and gender, the system converts complex molecular patterns into personalized risk scores.

The analysis demonstrated that CardiOmicScore significantly improved prediction accuracy compared with traditional polygenic risk scores. The model was able to forecast the risk of six major cardiovascular diseases, including coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease, and venous thromboembolism. In high-risk individuals, the system was able to provide warning signals up to 15 years before clinical symptoms appeared.

Conventional cardiovascular risk assessments typically rely on factors such as age, blood pressure, and smoking status. While these indicators remain important, they often fail to detect early biological changes that occur long before symptoms develop. Polygenic risk scores have also gained attention in recent years, but these tools reflect inherited genetic risk that remains fixed throughout life. They do not account for dynamic influences such as lifestyle, environmental exposure, or evolving physiological conditions. By contrast, proteins and metabolites circulating in the bloodstream continuously reflect the body’s current health status.

The CardiOmicScore approach captures these dynamic biological signals, potentially enabling physicians to identify disease risk much earlier and guide preventive strategies. The findings, published in Nature Communications, highlight a shift in precision medicine toward dynamic, multiomics-based disease prediction. In the future, a small blood sample could be sufficient to generate a comprehensive cardiovascular risk profile across multiple diseases. Such predictive tools may help clinicians identify individuals who would benefit from earlier lifestyle interventions, closer monitoring, or preventive therapies, potentially reducing the global burden of cardiovascular disease.

“Genes determine where we start—they define our baseline health risk. However, proteins and metabolites reflect our current physical health,” said HKUMed Professor Zhang Qingpeng. “Our AI tool is designed to decode these complex molecular signals, enabling doctors and patients to identify risks much earlier, which can potentially change the trajectory of disease through timely lifestyle modifications and early prevention.”

Related Links:
LKS Faculty of Medicine