Heart disease remains the leading cause of death around the world. Every year millions of people die from conditions such as heart attacks, strokes, and heart failure.
According to global health estimates, cardiovascular diseases were responsible for nearly 19.8 million deaths in 2022 alone. Because of this enormous health burden, doctors and researchers are constantly searching for better ways to detect heart problems early and prevent them before serious damage occurs.
Traditionally, doctors estimate a person’s risk of heart disease by looking at several common health indicators. These often include age, blood pressure, cholesterol levels, smoking habits, weight, and family medical history.
While these factors can provide useful information, they do not always capture the earliest biological changes happening inside the body. Many people develop heart disease slowly over many years without obvious warning signs. By the time symptoms appear, the disease may already be advanced.
To improve early detection, a research team from the Department of Pharmacology and Pharmacy at the LKS Faculty of Medicine of the University of Hong Kong has developed a new artificial intelligence system called CardiOmicScore.
This innovative tool uses information from a simple blood test to estimate a person’s future risk of several major cardiovascular diseases. The findings of the research were published in the scientific journal Nature Communications.
The CardiOmicScore system uses artificial intelligence to analyze large amounts of biological data from blood samples. Instead of focusing only on traditional risk factors, the system looks at many molecular signals that reflect what is happening inside the body at a given moment.
These signals include proteins and small molecules called metabolites that circulate in the bloodstream.
Proteins and metabolites are important because they respond to many influences such as diet, exercise, stress, inflammation, and overall health. In this way, they can act like real-time indicators of how the body is functioning.
While a person’s genes remain mostly unchanged throughout life, proteins and metabolites can change as a person’s lifestyle and environment change. This makes them especially useful for monitoring current health status.
In recent years, scientists have used genetic information to estimate the risk of certain diseases. These calculations are known as polygenic risk scores. However, genetic risk is fixed at birth and cannot change as a person’s behavior or environment changes. As a result, genetic scores may not fully capture the impact of lifestyle choices such as diet, exercise, or smoking.
To address this limitation, the Hong Kong research team combined several types of biological information in what scientists call a “multiomics” approach. This means they analyzed different layers of biological data together, including genetic information, proteins, and metabolites. Artificial intelligence was then used to identify patterns in these complex datasets.
The researchers built their model using data from the UK Biobank, one of the largest health research databases in the world.
The study examined blood samples from thousands of participants and analyzed 2,920 different circulating proteins along with 168 metabolites found in the blood. These molecules provided detailed signals about immune activity, metabolism, and blood vessel health.
Using deep learning techniques, the researchers trained the AI system to recognize patterns linked to future cardiovascular disease.
The resulting CardiOmicScore model was able to estimate the risk of six major heart and blood vessel conditions. These included coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease, and venous thromboembolism.
The results of the study were impressive. The new system was able to predict the risk of these diseases more accurately than many existing methods. When the molecular data were combined with simple clinical information such as age and gender, the prediction accuracy improved even further.
One of the most striking findings was that the system could identify elevated disease risk many years before symptoms appeared. In some cases, the model could detect warning signs up to 15 years before a cardiovascular disease was diagnosed. This early detection could give doctors and patients much more time to take preventive action.
Professor Zhang Qingpeng, an associate professor involved in the research, explained that genes determine the starting point of a person’s health risk, but proteins and metabolites provide a picture of the body’s current condition.
By decoding these signals using artificial intelligence, the system may help doctors identify health problems earlier and guide people toward healthier lifestyles before disease develops.
In the future, the researchers believe that a small blood sample could be enough to generate a detailed cardiovascular risk profile. This profile could help doctors recommend personalized prevention strategies, such as improving diet, increasing physical activity, controlling blood pressure, or taking preventive medication.
The study also reflects a broader shift in modern medicine toward what is often called precision medicine. Instead of using a one-size-fits-all approach, precision medicine aims to tailor prevention and treatment strategies to each individual based on their unique biological information.
Looking closely at the findings, the CardiOmicScore system appears to offer several important advantages. First, it captures real-time biological signals that can change with lifestyle and environment.
This means the system may be more responsive to changes in health behaviors compared with genetic risk tools. Second, the system can evaluate the risk of multiple cardiovascular diseases at the same time, providing a more comprehensive picture of a person’s heart health.
However, it is also important to interpret these findings carefully. While the prediction model performed well in the research setting, further studies will be needed before the system can be widely used in routine medical practice.
Researchers must confirm that the tool works reliably in different populations and healthcare systems. They must also study how doctors and patients can best use the information in real-world clinical decisions.
Despite these remaining questions, the study highlights the growing potential of artificial intelligence in preventive medicine.
If tools like CardiOmicScore become widely available, they could help shift healthcare from reacting to disease after it appears to preventing disease before it begins. This shift could improve health outcomes for millions of people and reduce the enormous global burden of heart disease.
Overall, the research demonstrates how combining artificial intelligence with advanced biological measurements can open new possibilities for early disease prediction. By identifying risk many years before symptoms develop, tools like CardiOmicScore may help people take action earlier and protect their heart health for the long term.
If you care about heart disease, please read studies that herbal supplements could harm your heart rhythm, and how eating eggs can help reduce heart disease risk.
For more health information, please see recent studies that apple juice could benefit your heart health, and results showing yogurt may help lower the death risks in heart disease.
Copyright © 2026 Knowridge Science Report. All rights reserved.
