Heart age estimated using explainable advanced electrocardiography

Abstract

Electrocardiographic (ECG) Heart Age conveying cardiovascular risk has been estimated by both Bayesian and artificial intelligence approaches. We hypothesised that explainable measures from the 10-s 12-lead ECG could successfully predict Bayesian 5-min ECG Heart Age. Advanced analysis was performed on ECGs from healthy subjects and patients with cardiovascular risk or proven heart disease. Regression models were used to predict patients' Bayesian 5-min ECG Heart Ages from their standard, resting 10-s 12-lead ECGs. The difference between 5-min and 10-s ECG Heart Ages were analyzed, as were the differences between 10-s ECG Heart Age and the chronological age (the Heart Age Gap). In total, 2,771 subjects were included (n = 1682 healthy volunteers, n = 305 with cardiovascular risk factors, n = 784 with cardiovascular disease). Overall, 10-s Heart Age showed strong agreement with the 5-min Heart Age (R² = 0.94, p < 0.001, mean ± SD bias 0.0 ± 5.1 years). The Heart Age Gap was 0.0 ± 5.7 years in healthy individuals, 7.4 ± 7.3 years in subjects with cardiovascular risk factors (p < 0.001), and 14.3 ± 9.2 years in patients with cardiovascular disease (p < 0.001). Heart Age can be accurately estimated from a 10-s 12-lead ECG in a transparent and explainable fashion based on known ECG measures, without deep neural network-type artificial intelligence techniques. The Heart Age Gap increases markedly with cardiovascular risk and disease.

Figure 1

Left panel: Scatter plot showing the relation between the 10-s ECG Heart Age and the 5-min ECG Heart Age in all participants. The R² value was 0.94 (p < 0.001). Right panel: Bland–Altman plot showing the difference between the 10-s and 5-min ECG Heart Age in relation to the mean of both ECG Heart Ages. The agreement between methods is strong, with minimal deviation from the identity line (dashed) or bias (0.0 ± 5.2 years).

Figure 2

(A) The difference (the Heart Age Gap) between the 10-s ECG Heart Age and chronological age in healthy subjects (left, dark green), subjects at cardiovascular (CV) risk (middle, light blue), and patients with CV disease (right, yellow). On average, there is no difference between Heart Age and chronological age in healthy subjects. Heart Age Gap is higher in subjects at CV risk, and highest for those with overt CV disease. (B) Scatter plots showing the relationship between the 10-s ECG Heart Age and chronological age in healthy subjects (left, dark green), subjects at CV risk (middle, light blue), and patients with CV disease (right, yellow). The dashed diagonal line is the identity line, i.e. indicating no difference between Heart Age and chronological age.

Figure 3

Example of the transparency and explainability of the Heart Age from two subjects with equal chronological age but different Heart Ages, illustrated by a commensurately aged female heart, but a disproportionately aged male heart. The ECG measures for each of the two patients are shown in the table in the middle of the figure, presented in the order of the relative strength (strongest first, based on t ratio [not shown]) of contribution to the Heart Age. Notably, P-wave duration is markedly different between these two patients, and heart rate is higher for the male than the female, helping drive the Heart Age higher in the male. The R-wave amplitude in lead Y is also much larger in the female, and the difference between the spatial ventricular gradient and the spatial mean QRS in the female is also larger, likely due to preserved T-wave amplitudes, contributing to her relatively younger Heart Age. Furthermore, possibly due to ischemic myocardial injuries, T-wave complexity is increased in the male, suggesting that increased myocardial repolarisation heterogeneity also contributes to driving Heart Age higher in the male, in spite of his shorter QT interval.