AI Tool Developed to Predict Cardiovascular Events and Heart-Related Risks

New Delhi: A team of researchers from Inha University Hospital in South Korea has developed a cutting-edge artificial intelligence (AI) algorithm capable of predicting the risk of major cardiovascular events and heart-related mortality using data from electrocardiograms (ECGs).

The innovative algorithm was trained on a vast dataset comprising 12-lead ECG recordings from 425,051 individuals collected over a span of 15 years. It was later validated using an independent set of 97,058 ECGs. By analysing this data, the AI model estimates the “biological age” of the heart—an indicator of cardiac function—rather than simply relying on a person’s chronological age.

According to the researchers, discrepancies between a person’s biological heart age and actual age can provide valuable insights into cardiovascular health. For instance, if a 50-year-old individual's heart functions like that of a 60-year-old, their biological heart age is considered older, potentially flagging higher risk. Conversely, a biological heart age younger than the actual age suggests better cardiovascular health.

Lead investigator Dr. Yong-Soo Baek highlighted the implications of their findings: “Our research showed that when the biological heart age was seven years older than the chronological age, the risk of all-cause mortality rose by 62%, and the risk of major adverse cardiovascular events (MACE) increased by 92%.” MACE includes conditions such as heart attacks, strokes, cardiovascular deaths, and revascularization procedures like angioplasty or bypass surgery.

In contrast, if the AI-estimated heart age was seven years younger than the individual’s actual age, the risk of all-cause mortality dropped by 14%, and MACE risk was reduced by 27%.

The study, presented at the European Heart Rhythm Association (EHRA) 2025 Congress in Vienna, underscores the potential of AI in transforming cardiovascular risk assessment. “This approach offers a paradigm shift in how we evaluate heart health, allowing for more precise, individualized risk prediction,” Dr. Baek noted.

The integration of AI into routine clinical diagnostics, as demonstrated by this study, opens new avenues for enhancing preventive cardiology and improving patient outcomes.