Why the study?
Pediatric ECG values vary across age and sex, but existing standards fail to capture multi-dimensional relationships and machine learning-enhanced ECG analysis remains scarce in children.
Can machine learning models accurately classify age and sex using resting 12-lead ECG features in healthy pediatric and young adult populations?
Observational (n=29,408)
Yes
Can machine learning models accurately classify age and sex using resting 12-lead ECG features in healthy pediatric and young adult populations?
Effect estimate: True positive rate (TPR) 60% for age group classification with SVM (95% CI TPR range 46-72% across age groups; Adjusted TPR 94% allowing 1-group misclassification)
Traditional supervised machine learning models, particularly SVM, can accurately capture age- and sex-related ECG signatures in youth, outperforming neural networks and providing a foundation for automated pediatric ECG interpretation.
Electrocardiogram (ECG) values vary significantly across age and sex, particularly during childhood and adolescence. While age- and sex-specific ECG standards exist, they often fail to capture complex multi-dimensional relationships and have not been applied in machine learning (ML) enhanced ECG analysis. Accuracy of automated ECG analysis in clinical practice improved significantly by applying ML models, however there is a paucity of such studies in the pediatric population. Our aim was to develop age- and sex-classification for children using ECG features with various ML models. We analyzed 29,408 curated resting 12-lead ECGs from healthy subjects aged 0–21 years using 177 digitized ECG variables combined with various ML models including regression and classification analyses and semi-supervised neural networks. Primary outcome variables were age and sex. Model performance was evaluated using F1-score, AUROC, and confusion matrices across repeated train-test splits. Support vector machine (SVM) achieved the highest accuracy in modeling both age and sex. Key predictive features included heart rate, PR interval, QRS duration, and T-wave amplitude. Age-group classification achieved an average true positive rate of 60% with SVM, improving to 94% when allowing one-group misclassification. Sex classification reached F1-scores of 0.91 and AUROC of 0.95 in adolescents and young adults, and moderate accuracy in younger children. Traditional supervised ML models can accurately model physiologic ECG changes related to age and sex, outperforming neural networks, particularly in smaller subgroups. These findings support the feasibility of ML models to capture of age- and sex-related ECG signatures to may aid future research and clinical applications in pediatric cardiology.
No takes yet. Share an insight, caveat, or question.
Hong-gen et al. (2026) conducted an observational in Healthy children and young adults aged 0–21 years without known heart condition (n=29,408). Machine learning based classification models on ECG features vs. Other machine learning models and methods was evaluated on Accuracy of ECG-based age and sex classification using machine learning models (True positive rate (TPR) 60% for age group classification with SVM, 95% CI TPR range 46-72% across age groups; Adjusted TPR 94% allowing 1-group misclassification).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: