Background Wearable devices such as biosensors and smartwatches have demonstrated clinical utility in arrhythmia detection in adult populations, but their integration into pediatric congenital heart disease (CHD) care remains limited. Despite advances in CHD management, challenges persist in the timely identification of arrhythmias in children. The use of wearables in pediatrics is constrained by research gaps, adherence issues, and complex data interpretation. Traditional tools like Holter monitors and event recorders face limitations in detecting intermittent arrhythmias and ensuring compliance. These devices cannot leverage artificial intelligence (AI) and machine learning (ML) to enhance ECG interpretation. This study analysis evaluates the potential of AI-augmented wearable biosensors to detect arrhythmias and early signs of decompensation more accurately and promptly than conventional methods, aiming to optimize outcomes in this vulnerable population.

Methods A comprehensive literature review was conducted using data from observational studies, randomized controlled trials, and pediatric cohorts. The primary outcome was the performance of wearable devices capable of ECG capture, photoplethysmography, or physiological pattern recognition integrated with AI or ML algorithms. Pediatric-focused studies were prioritized, though adult findings were considered when applicable to children. A secondary outcome assessed the incidence of postoperative arrhythmias in common CHDs and strategies for long-term rhythm surveillance.

Results Commercial wearables, especially smartwatches, have shown promise in detecting atrial fibrillation (AF). In one study, smartwatch-detected AF achieved 96.2% sensitivity and 93.1% accuracy, with strong concordance with cardiologist interpretation. The Apple Heart Study (420,000 adults) reported that 0.5% received an irregular pulse notification; 34% of these were confirmed to have AF on follow-up ECG. Although Apple Watch ECG features are approved for users ≥22 years old, one pediatric-focused study found that 71% of new arrhythmias were detected using these devices, 29% of which were missed by traditional monitors. ML-enhanced AF detection further improved sensitivity to 100% and specificity to 96%. Tandon et al. demonstrated that AI can efficiently process complex datasets and detect early deviations from individual health baselines.

Conclusion While limited, emerging data suggest AI-integrated wearables could transform arrhythmia detection in pediatric CHD. These technologies promise earlier identification, timely intervention, and better long-term outcomes. There is an urgent need for further research and funding to develop pediatric-specific, AI-driven wearable monitoring solutions.