A groundbreaking blood test enhanced by artificial intelligence could transform the early detection of cardiovascular disease by identifying warning signs years before symptoms appear, according to new research.
Scientists at the LKS Faculty of Medicine of the University of Hong Kong have developed CardiOmicScore, an advanced risk assessment tool that combines artificial intelligence with large-scale biological data analysis to estimate an individual’s future likelihood of developing major cardiovascular conditions.
The findings, published in Nature Communications, indicate that the system can evaluate the risk of six common cardiovascular diseases, including coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease, and venous thromboembolism.
Unlike conventional risk assessment methods, which rely largely on factors such as age, blood pressure, smoking status, and medical history, the new approach analyses biological signals that reflect the body’s current physiological condition. This enables the detection of subtle changes that may occur long before a disease becomes clinically apparent.
To develop the model, researchers applied deep-learning techniques to extensive data from the UK Biobank. The study examined 2,920 circulating proteins and 168 metabolites measured in blood samples, creating detailed molecular profiles capable of capturing complex biological processes.
These proteins and metabolites act as real-time indicators of health, reflecting changes in immune function, metabolism, and vascular health. As a result, the system provides a dynamic assessment of disease risk that may offer greater predictive value than genetic risk scores alone, which remain largely unchanged throughout life.
According to the research team, CardiOmicScore significantly outperformed existing polygenic risk models by converting complex multi-omics information into personalised risk estimates. When combined with standard clinical data such as age and sex, the tool substantially improved prediction accuracy across multiple cardiovascular diseases.
One of the study’s most notable findings is its ability to identify elevated cardiovascular risk as much as 15 years before symptoms emerge or a diagnosis is made. Researchers believe this could provide a critical window for lifestyle modifications, preventive treatment strategies, and closer medical monitoring.
The study highlights a growing shift toward precision medicine, where health assessments are based not only on genetic predisposition but also on real-time biological activity within the body. In the future, a single blood sample may be sufficient to generate a comprehensive cardiovascular risk profile, enabling earlier and more personalised prevention strategies.
Researchers say their ultimate goal is to move healthcare beyond treating disease after it develops and toward predicting and preventing illness before it occurs, potentially improving outcomes for millions of people worldwide.