Although everyone’s heart has an absolute ideal age (as old as that person), the heart also has a theoretical “biology” age1, which is based on the function of the heart. Therefore, people who are 50 years old but have poor heart health may suffer from a biological heart age of 60, while the best heart health 50 years old may have a biological heart age of 40.
Researchers who have demonstrated that by using artificial intelligence (AI) to analyze standard 12-lead electrocardiogram (ECG) from nearly half a million cases, they were able to create an algorithm to predict the biological age of the heart by using artificial intelligence (AI). The algorithm can be used to identify those who are most at risk of cardiovascular events and mortality.
“Our study shows that when the biological age of the heart exceeds its age seven, the risk of all-cause mortality and major adverse cardiovascular events increases dramatically,” explained Associate Professor Yong-soo Baek of Iha University Hospital in South Korea. “In contrast, if the algorithm estimates that the biological heart is seven years younger than age, it reduces the risk of death and major adverse cardiovascular events.”
The integration of artificial intelligence (AI) in clinical diagnosis provides new opportunities to enhance the accuracy of cardiology predictions. “Using AI to develop algorithms in this way will introduce a potential paradigm shift in cardiovascular risk assessment,” said Associate Professor Baek.
Their study evaluated the prognostic capabilities of a deep learning-based algorithm that computed the biological ECG cardiac era (AI ECG-HEART age) from 12 lead electrocardiograms, comparing its predictive capabilities to the mortality and cardiovascular outcomes of traditional chronological school age (CA).
A deep neural network was developed and trained on a large data set of 425,051 12-lead ECGs collected over 15 years, and the independent queue of 97,058 ECGs was subsequently validated and tested. A comparative analysis was performed on age- and gender-matched patients and differentiated by ejection fraction (EF).
In the statistical model, AI ECG heart age exceeding the age of the heart is over seven years, which is associated with a 62% increase in risk of all-cause mortality and a 92% increase in risk of MACE. By comparison, AI ECG heart age, seven years younger than age, reduced the risk of all-cause mortality by 14% and MACE by 27%.
In addition, subjects with reduced ejection fractions always exhibited increased AI ECG heart age and prolonged QRS duration (the time spent by the heart’s election signal passing through the ventricle, causing contraction) and corrected QT interval (the total time required for the heart’s electrical system to complete a cycle and relaxation).
The authors explain that the observed correlation between reducing ejection fraction and increasing AI ECG heart age and the importance of prolonging QRS duration and correcting QT intervals suggest that AI ECG heart age effectively reflects various heart depolarization and replication processes. These indicators of electroremodeling in the heart may indicate the underlying heart health status of heart health and its association with ejection fraction (EF). However, Associate Professor Baek explained, “It is crucial to obtain statistically sufficient sample size in future studies to further confirm these findings. This approach will enhance the robustness and applicability of AI ECG in clinical evaluation of cardiac function and health.”
He concluded: “Biological cardiac age estimated by AI through 12 lead electrocardiograms is closely related to increased mortality and cardiovascular events, which emphasizes its practicality in enhancing early detection and prevention strategies in cardiovascular disease. This study confirms the transformational potential of AI in refining clinical evaluation and improving patients.
If you find this report useful, consider supporting our work with a small donation. Your contribution allows us to continue to bring you accurate, thought-provoking scientific and medical news that you can trust. Independent reporting requires time, effort, and resources, and your support makes it possible for us to continue exploring stories that are important to you. Thank you so much!
