ROCHESTER, Minn. — Artificial intelligence (AI) derived from a patient’s electrocardiogram (ECG) could be an innovative solution to enhance heart disease risk assessment. Atherosclerotic cardiovascular disease (a disease in which arteries become narrowed or blocked by the buildup of fatty plaques) is the world’s leading cause of death and is often caused by coronary artery disease. New research data published in eClinicalMedicine shows that ECG-AI can identify signs of coronary artery disease such as calcifications and blockages, as well as evidence of previous heart attacks, several years earlier than current risk formulas. It suggests that risks can be flagged.
Many people may have coronary artery disease and not realize it. Unfortunately, the first sign of this disease can be sudden death or a serious heart attack. Clinician tools such as pooled cohort equations can help determine a patient’s 10-year risk of heart attack or stroke. Although these tools guide shared decisions about the timing of treatment, these tools have limitations.
ECG is a widely available test that measures the electrical activity of the heart, and AI can be trained to identify and detect hidden patterns of disease from these electrical signals.
The ECG-AI that predicts coronary artery disease was developed at Mayo Clinic and Anumana using a retrospective analysis of electronic health data of more than 7 million patients across the United States. Three separate his ECG-AI models were trained to detect, respectively, coronary artery calcium, coronary artery blockage and parts of the heart’s left ventricle not working well – signs of a previous heart attack.
“Using a combination of three independent ECG-AI models, we predicted which patients were at high risk for hidden coronary artery disease and therefore at high risk of having a heart attack. “In particular, AI helped us calculate these risks over a short period of three years,” said Mayo Clinic cardiologist and lead author of the paper. said Dr. Francisco López Jiménez. “Using the pooled cohort equation alone estimates the 10-year risk of developing cardiovascular disease. Adding ECG-AI to identify hidden risks sooner could save more lives. “This model could also help identify people who are unaware they have cardiovascular disease and may benefit from life-saving treatment.”
The study was funded by nference and Anumana, a Mayo Clinic spinoff company. Mayo Clinic licensed the technology to Anumana. If this technology becomes commercially viable, Mayo Clinic and the authors of this study could benefit financially in the future.
###
About Mayo Clinic
Mayo Clinic is a nonprofit organization dedicated to innovation in clinical practice, education and research, providing compassion, expertise and answers to all who need healing. For more Mayo Clinic news, visit Mayo Clinic News Network.
Media contact:
