SerpApi.Org
A new blood test powered by artificial intelligence could flag a person’s risk of developing serious heart and circulatory disease more than a decade before any symptoms appear, according to researchers from The University of Hong Kong who developed the tool.
“Genes determine where we start—they define our baseline health risk; however, proteins and metabolites reflect our current physical health,” Professor Zhang Qingpeng, the study’s lead author, said in a statement. “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.”
Newsweek has contacted the research team by email for further comment.
A Single Test, Six Diseases
The tool, called CardiOmicScore, uses a single blood sample to estimate a person’s future risk of six major cardiovascular conditions: coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease and venous thromboembolism.
In people found to be at elevated risk, the model detected warning signs as much as 15 years before disease onset.
Cardiovascular disease is still the leading cause of death worldwide, responsible for an estimated 19.8 million deaths in 2022. Doctors typically judge a patient’s risk using factors such as age, blood pressure and smoking history.
These measures are useful, but they can miss subtle biological changes happening inside the body long before a diagnosis is made, meaning some patients are not flagged as high-risk until prevention options have narrowed.
Why It Differs From Genetic Testing
While genetic testing offers another way to gauge risk, with polygenic risk scores combining the effects of many inherited gene variants into a single number, a person’s DNA is fixed from birth, and these scores cannot capture how risk shifts with diet, exercise, aging, illness or other lifestyle and environmental factors.
CardiOmicScore was built to fill that gap by measuring what is happening in the body right now, rather than what was inherited.
“This research offers an exciting glimpse into the future of cardiovascular prevention,” Rick Snyder, a board-certified advanced interventional cardiologist and president of HeartPlace Cardiology, told Newsweek. “By using AI to interpret thousands of proteins and metabolites circulating in the blood, CardiOmicScore may eventually help physicians recognize biological warning signs years before a patient develops symptoms.”
How the Tool Was Built
The Hong Kong-based team used deep learning to combine multiple layers of biological data, an approach known as multiomics. It draws on genomics, the study of genetic material; proteomics, the study of proteins; and metabolomics, the study of small molecules called metabolites that result from digestion, energy production and the body’s response to disease.
Using population data from the U.K. Biobank, the researchers built their model from 2,920 circulating proteins and 168 metabolites found in blood samples. Together, these molecules can reveal early shifts in immune activity, metabolism and blood vessel health before any symptoms show up.
Outperforming Genetic Scores
CardiOmicScore outperformed conventional polygenic risk scores at predicting cardiovascular disease, and its accuracy improved further once the researchers factored in basic clinical details such as age and gender.
The findings point to a broader change in how doctors may one day assess heart risk. Where genetic tests offer a fixed, lifelong estimate, tools like CardiOmicScore can track biological signals that shift over time, potentially giving patients and doctors a much longer window to intervene before disease takes hold.
“Still, promising technology must move through the appropriate clinical stage gates,” Snyder said. “This model was developed primarily from U.K. Biobank data, has not yet been independently validated and should not replace established risk assessments or physician judgment.
“We need to know that any new tool works accurately across diverse patient populations, leads to clear and actionable care decisions, improves outcomes and can be delivered affordably, but the future of outpatient cardiovascular care will be increasingly predictive, personalized and preventive.”
Reference
Luo, Y., Zhang, N., et al. (2026). AI-based multiomics profiling reveals complementary omics contributions to personalized prediction of cardiovascular disease. Nature Communications. 10.1038/s41467-026-68956-6
Contact Newsweek editors on this story: Kara Dolman and Emma Lee-Sang
Related Articles
