Artificial Intelligence-Enabled ECG to Detect Congenitally Corrected Transposition of the Great Arteries

Abstract

L-loop congenitally corrected transposition of the great arteries (ccTGA) is a rare congenital heart defect that may remain undiagnosed for decades and lead to significant morbidities, making it of interest for early detection. In this study, we address this gap by developing and internally testing an artificial intelligence-enabled electrocardiogram (AI-ECG) model to diagnose ccTGA from standard 12-lead ECGs. The dataset included the first ECG from 61,482 patients (0.7% with ccTGA), which was partitioned into training (70%) and testing (30%) cohorts. The convolutional neural network model achieved an area under the receiver-operating characteristic curve of 0.95 [95% CI 0.94-0.96] and an area under the precision-recall curve of 0.16 [95% CI 0.12-0.21]. The model performed well across different age groups, with slightly lower performance in patients < 1 month old. Key features identified by the model included widened QRS complexes, negative QRS complexes in leads V1-V2, and the lack of Q waves in lateral precordial leads. This study highlights the potential of AI-ECG to detect subtle patterns in rare congenital heart defects, providing a scalable method for early diagnosis and improving access to care. Future studies may include external validation in diverse clinical settings and multi-modal models to enhance performance and clinical utility.

Keywords: Artificial intelligence; Congenital heart disease; Congenitally corrected transposition of the great arteries; Electrocardiograms,.