Researchers have demonstrated that by using AI to analyse standard 12-lead electrocardiograph (ECG) data taken from almost half a million people, they were able to create an algorithm to predict the biological age of the heart. This algorithm could be used to identify those most at risk of cardiovascular events and mortality.
While everybody’s heart has an absolute chronological age – which is as old as that person is – hearts also have a theoretical ‘biological’ age that is based on how the heart functions. So, someone who is 50 but has poor heart health could have a biological heart age of 60, while someone of 50 with optimal heart health could have a biological heart age of 40.
“Our research showed that when the biological age of the heart exceeded its chronological age by seven years, the risk of all-cause mortality and major adverse cardiovascular events increased sharply,” said associate professor Yong-Soo Baek, Inha University Hospital, in South Korea.
“Conversely, if the algorithm estimated the biological heart as seven years younger than the chronological age, that reduced the risk of death and major adverse cardiovascular events.”
The integration of AI into clinical diagnostics presents novel opportunities for enhancing predictive accuracy in cardiology.
“Using AI to develop algorithms in this way introduces a potential paradigm shift in cardiovascular risk assessment,” said Baek.
Their study evaluated the prognostic capabilities of a deep learning based algorithm that calculates biological ECG heart age (AI ECG-heart age) from 12-lead ECGs, comparing its predictive power against traditional chronological age (CA) for mortality and cardiovascular outcomes.
A deep neural network was developed and trained on a substantial dataset of 425,051 12-lead ECGs collected over fifteen years, with subsequent validation and testing on an independent cohort of 97,058 ECGs. Comparative analyses were conducted among age and sex-matched patients differentiated by ejection fraction (EF).
In statistical models, an AI ECG-heart age exceeding the heart’s chronical age by seven years was associated with an increased the risk of all-cause mortality by 62 per cent and of MACE by 92 per cent. In contrast, an AI ECG heart age that was seven years younger than its chronological age reduced the risk of all-cause mortality by 14 per cent and MACE by 27 per cent.
Additionally, subjects with reduced ejection fraction consistently exhibited increased AI ECG heart ages along with prolonged QRS durations (the time taken for the heart’s electrical signal to travel through the ventricles, causing contraction) and corrected QT intervals (the total time needed for the heart’s electrical system to complete one cycle of contraction and relaxation).
The authors explain that the significance of the observed correlation between reduced ejection fraction and increased AI ECG heart ages, alongside prolonged QRS durations and corrected QT intervals, suggests that AI ECG heart age effectively reflects various cardiac depolarisation and repolarisation processes.
These indicators of electrical remodelling within the heart may signify underlying cardiac health conditions and their association with ejection fraction (EF).
However, Baek said: “It is crucial to obtain a statistically sufficient sample size in future studies to substantiate these findings further. This approach will enhance the robustness and applicability of AI ECG in clinical assessments of cardiac function and health.
“Biological heart age estimated by artificial intelligence from 12-lead electrocardiograms is strongly associated with increased mortality and cardiovascular events, underscoring its utility in enhancing early detection and preventive strategies in cardiovascular healthcare. This study confirms the transformative potential of AI in refining clinical assessments and improving patient outcomes.”