Artificial Intelligence-Enhanced Implantable Loop Recorders in Pediatric Patients: Effects on Device Performance and Clinical Workflow in a Single-Center Experience.

INTRODUCTION: Artificial intelligence (AI)-based algorithms have been developed to reduce alert burden in patients with implantable loop recorders (ILRs), but their performance in pediatric populations has not been previously evaluated. METHODS: We retrospectively analyzed pediatric patients implanted with an AI-enhanced ILR (Linq2, Medtronic) incorporating AI-based filtering (AccuRhythm algorithms). The impact of AI on nonactionable alerts (NAAs) and estimated clinical workflow was assessed using previously published time-and-motion models. RESULTS: Forty-five patients (10.8 ± 4.7 years) were included. AI-based filtering significantly reduced NAAs (43.7%), driven predominantly by a marked reduction in pause-related alerts (75%), corresponding to an estimated saving of 14 clinic hours over 3 months (58 projected hours annually). CONCLUSIONS: In this exploratory pediatric experience, AI-enhanced ILR monitoring substantially reduced the burden of NAAs, particularly for pause alerts. These findings highlight the importance of population-specific AI algorithms and suggest that AI-based filtering may improve the efficiency and sustainability of monitoring in pediatric electrophysiology.