A joint research team has developed an artificial intelligence (AI)-based cardiovascular disease prediction model for Koreans.

The team, led by Professor Kang Si-hyuk of Seoul National University Bundang Hospital and Professor Cho Sang-young of Gyeongsang National University Changwon Hospital, used data from about 220,000 adults aged 40 to 80 who participated in the National Health Insurance Service's health checkup from 2009 to 2010.

The predictive model predicts the risk of cardiovascular disease through the subject's data, such as age, sex, systolic blood pressure, cholesterol level, smoking status, and diabetes history.

Researchers confirmed that 7,819, or 3.51 percent of the total 220,000 people, had the atherosclerotic cardiovascular disease during a five-year follow-up.

After analyzing the accuracy of their cardiovascular disease risk prediction model, the team confirmed that existing models usually showed a prediction accuracy of 70-80 percent. Specifically, the prediction accuracy through the pooled cohort equation of the U.S. AHA prediction model, which was the main target of comparison, was 73.8 percent.

In comparison, the model developed by the team showed a slightly higher accuracy of 75.1 percent, confirming its superior predictive performance compared to the existing model.

"The advantage of machine learning is increased accuracy and reduced error even with the same parameters," the team said. "We expect the new model will be advantageous for individualized, customized treatments for patients as it can accurately calculate the individual's risk."

Professor Kang said, "AI machine learning will continue to be used not only for health but also for enriching life."

Suppose researchers broaden the application of machine learning in the medical field. In that case, it will be possible to accurately predict the risk of diseases and provide more effective treatments while reducing human effort, he added.

Professor Cho also noted that the study's core was to select the risk group through the cardiovascular disease prediction model and present an effective preventive treatment policy.

"As the study confirmed that the predictive power of our model developed is superior compared to the previously used model, we will continue researching the development and use of high-accuracy evaluation tool," he said.