Attia, Zachi I.

Abstract Aims We sought to investigate whether artificial intelligence (AI) and specifically deep neural networks (NNs) for electrocardiogram (ECG) signal analysis can be explained using human-selected features. We also sought to quantify such explainability and test if the AI model learns features that are similar to a human expert. Methods and results We used a set of 100 000 ECGs that were annotated by human explainable features. We applied both linear and non-linear models to predict published ECG AI models output for the detection of patients’ age and sex. We further used canonical correlation analysis to quantify the amount of shared information between the NN features and human-selected features. We reconstructed single human-selected ECG features from the unexplained NN features using a simple linear model. We noticed a strong correlation between the simple models and the AI output (R2 of 0.49–0.57 for the linear models and R2 of 0.69–0.70 for the non-linear models). We found that the correlation of the human explainable features with either 13 of the strongest age AI features or 15 of the strongest sex AI features was above 0.85 (for comparison, the first 14 principal components explain 90% of the human feature variance). We linearly reconstructed single human-selected ECG features from the AI features with R2 up to 0.86. Conclusion This work shows that NNs for ECG signals extract features in a similar manner to human experts and that they also generate additional novel features that help achieve superior performance.

Introduction: An artificial intelligence (AI) algorithm applied to electrocardiography (ECG) during normal sinus rhythm (NSR) has been shown to predict concomitant atrial fibrillation (AF). We sought to characterize the value of AI-ECG as a predictor of future AF and assess its performance compared to other clinical prediction scores in a population-based sample. Methods: We calculated the AI-ECG probability during NSR in patients who enrolled in the population-based Mayo Clinic Study of Aging with at least one ECG in NSR within two years prior and no history of AF at the time of the baseline study visit. The cumulative incidence of AF was estimated for strata defined by AI-ECG probability and CHARGE-AF score. Cox proportional hazards were fit to assess the independent prognostic value and interaction of AI-ECG probability and CHARGE-AF score. Concordance (c) statistics were calculated for AI-ECG probability, CHARGE-AF score and combined AI-ECG and CHARGE-AF score. Results: A total of 1,936 patients with a median age 75.8 (quartile range [QR] 70.4, 81.8) years, median CHARGE-AF score 14.0 (QR 13.2, 14.7) and median CHADS2VASC score 3 (QR 2, 4) were included in the analysis. The cumulative incidence of AF increased in a stepwise fashion across quartiles of AI-ECG probability and CHARGE-AF score (Figure 1). When compared in the same model, both AI-ECG probability (hazard ratio [HR] 1.76, 95% confidence interval [CI] 1.51-2.04) and CHARGE-AF score (HR 1.90, 95% CI 1.58-2.28) independently predicted AF without significant interaction (p=0.54). C statistics were 0.69 (95% CI 0.66-0.72) for AI-ECG probability, 0.69 (95% CI 0.66-0.71) for CHARGE-AF and 0.72 (95% CI 0.69-0.75) for combined AI-ECG and CHARGE-AF score. Conclusions: In the Mayo Clinic Study of Aging, both the AI-ECG probability and CHARGE-AF score independently predicted time to AF. The AI-ECG may offer a means to assess risk with a single test without requiring manual or automated clinical data abstraction.

Abstract Background We have demonstrated that a neural network is able to predict a person’s age from the electrocardiogram (ECG) [artificial intelligence (AI) ECG age]. However, some discrepancies were observed between ECG-derived and chronological ages. We assessed whether the difference between AI ECG and chronological age (Age-Gap) represents biological ageing and predicts long-term outcomes. Methods and results We previously developed a convolutional neural network to predict chronological age from ECGs. In this study, we used the network to analyse standard digital 12-lead ECGs in a cohort of 25 144 subjects ≥30 years who had primary care outpatient visits from 1997 to 2003. Subjects with coronary artery disease, stroke, and atrial fibrillation were excluded. We tested whether Age-Gap was correlated with total and cardiovascular mortality. Of 25 144 subjects tested (54% females, 95% Caucasian) followed for 12.4 ± 5.3 years, the mean chronological age was 53.7 ± 11.6 years and ECG-derived age was 54.6 ± 11 years (R2 = 0.79, P < 0.0001). The mean Age-Gap was small at 0.88 ± 7.4 years. Compared to those whose ECG-derived age was within 1 standard deviation (SD) of their chronological age, patients with Age-Gap ≥1 SD had higher all-cause and cardiovascular disease (CVD) mortality. Conversely, subjects whose Age-Gap was ≤1 SD had lower all-cause and CVD mortality. Results were unchanged after adjusting for CVD risk factors and other survival influencing factors. Conclusion The difference between AI ECG and chronological age is an independent predictor of all-cause and cardiovascular mortality. Discrepancies between these possibly reflect disease independent biological ageing.

Abstract Aims Early detection of aortic stenosis (AS) is becoming increasingly important with a better outcome after aortic valve replacement in asymptomatic severe AS patients and a poor outcome in moderate AS. We aimed to develop artificial intelligence-enabled electrocardiogram (AI-ECG) using a convolutional neural network to identify patients with moderate to severe AS. Methods and results Between 1989 and 2019, 258 607 adults [mean age 63 ± 16.3 years; women 122 790 (48%)] with an echocardiography and an ECG performed within 180 days were identified from the Mayo Clinic database. Moderate to severe AS by echocardiography was present in 9723 (3.7%) patients. Artificial intelligence training was performed in 129 788 (50%), validation in 25 893 (10%), and testing in 102 926 (40%) randomly selected subjects. In the test group, the AI-ECG labelled 3833 (3.7%) patients as positive with the area under the curve (AUC) of 0.85. The sensitivity, specificity, and accuracy were 78%, 74%, and 74%, respectively. The sensitivity increased and the specificity decreased as age increased. Women had lower sensitivity but higher specificity compared with men at any age groups. The model performance increased when age and sex were added to the model (AUC 0.87), which further increased to 0.90 in patients without hypertension. Patients with false-positive AI-ECGs had twice the risk for developing moderate or severe AS in 15 years compared with true negative AI-ECGs (hazard ratio 2.18, 95% confidence interval 1.90–2.50). Conclusion An AI-ECG can identify patients with moderate or severe AS and may serve as a powerful screening tool for AS in the community.

Background: Heart rate–corrected QT interval (QTc) prolongation, whether secondary to drugs, genetics including congenital long QT syndrome, and/or systemic diseases including SARS-CoV-2–mediated coronavirus disease 2019 (COVID-19), can predispose to ventricular arrhythmias and sudden cardiac death. Currently, QTc assessment and monitoring relies largely on 12-lead electrocardiography. As such, we sought to train and validate an artificial intelligence (AI)–enabled 12-lead ECG algorithm to determine the QTc, and then prospectively test this algorithm on tracings acquired from a mobile ECG (mECG) device in a population enriched for repolarization abnormalities. Methods: Using >1.6 million 12-lead ECGs from 538 200 patients, a deep neural network (DNN) was derived (patients for training, n = 250 767; patients for testing, n = 107 920) and validated (n = 179 513 patients) to predict the QTc using cardiologist-overread QTc values as the “gold standard”. The ability of this DNN to detect clinically-relevant QTc prolongation (eg, QTc ≥500 ms) was then tested prospectively on 686 patients with genetic heart disease (50% with long QT syndrome) with QTc values obtained from both a 12-lead ECG and a prototype mECG device equivalent to the commercially-available AliveCor KardiaMobile 6L. Results: In the validation sample, strong agreement was observed between human over-read and DNN-predicted QTc values (−1.76±23.14 ms). Similarly, within the prospective, genetic heart disease–enriched dataset, the difference between DNN-predicted QTc values derived from mECG tracings and those annotated from 12-lead ECGs by a QT expert (−0.45±24.73 ms) and a commercial core ECG laboratory [10.52±25.64 ms] was nominal. When applied to mECG tracings, the DNN’s ability to detect a QTc value ≥500 ms yielded an area under the curve, sensitivity, and specificity of 0.97, 80.0%, and 94.4%, respectively. Conclusions: Using smartphone-enabled electrodes, an AI DNN can predict accurately the QTc of a standard 12-lead ECG. QTc estimation from an AI-enabled mECG device may provide a cost-effective means of screening for both acquired and congenital long QT syndrome in a variety of clinical settings where standard 12-lead electrocardiography is not accessible or cost-effective.

Background An artificial intelligence algorithm that detects age using the 12‐lead ECG has been suggested to signal “physiologic age.” This study aimed to investigate the association of peripheral microvascular endothelial function (PMEF) as an index of vascular aging, with accelerated physiologic aging gauged by ECG‐derived artificial intelligence–estimated age. Methods and Results This study included 531 patients who underwent ECG and a noninvasive PMEF assessment using reactive hyperemia peripheral arterial tonometry. Abnormal PMEF was defined as reactive hyperemia peripheral arterial tonometry index ≤2.0. Accelerated or delayed physiologic aging was calculated by the Δ age (ECG‐derived artificial intelligence–estimated age minus chronological age), and the association between Δ age and PMEF as well as its impact on composite major adverse cardiovascular events were investigated. Δ age was higher in patients with abnormal PMEF than in patients with normal PMEF (2.3±7.8 versus 0.5±7.7 years; P =0.01). Reactive hyperemia peripheral arterial tonometry index was negatively associated with Δ age after adjustment for cardiovascular risk factors (standardized β coefficient, –0.08; P =0.048). The highest quartile of Δ age was associated with an increased risk of major adverse cardiovascular events compared with the first quartile of Δ age in patients with abnormal PMEF, even after adjustment for cardiovascular risk factors (hazard ratio, 4.72; 95% CI, 1.24–17.91; P =0.02). Conclusions Vascular aging detected by endothelial function is associated with accelerated physiologic aging, as assessed by the artificial intelligence–ECG Δ age. Patients with endothelial dysfunction and the highest quartile of accelerated physiologic aging have a marked increase in risk for cardiovascular events.