ABSTRACT:
Pediatric obstructive sleep apnea (OSA) affects 3% of children and can cause serious complications if untreated. Polysomnography is costly and limited, prompting interest in artificial intelligence (AI) and machine learning (ML) for better screening and personalized care. This study maps global trends on AI/ML for diagnosing and managing pediatric OSA, covering publication growth, key contributors, collaborations, research themes, and future directions. Relevant publications were retrieved from Web of Science, Scopus, and PubMed. Bibliometric analyses including co-authorship, co-citation, and keyword co-occurrence were performed using VOSviewer, Bibliometrix R, and CiteSpace. Visualizations were used to identify influential trends and research gaps. Publications have increased markedly since 2018, driven by advances in deep learning, wearable devices, and remote monitoring. The United States and China lead in publication volume. Major contributing institutions include Universidad de Valladolid, Cincinnati Children’s Hospital, University of Missouri, and Harvard Medical School. Influential authors such as D. Gozal, D. Álvarez, and R. Hornero have shaped the field, with frequent publications in the International Journal of Pediatric Otorhinolaryngology, Pediatric Pulmonology, Sleep Medicine, and Computers in Biology and Medicine. Thematic mapping reveals strong links among keywords like machine learning, deep learning, neural networks, and signal processing, highlighting a focus on AI-driven diagnosis, portable monitoring, risk prediction, and personalized therapy. Research is also expanding toward risk prediction models and personalized therapy. However, gaps remain in large-scale validation, diverse population datasets, and practical integration into clinical workflows. Studies targeting multi-ethnic cohorts and very young children are limited, highlighting key opportunities for future research. Emerging hotspots include real-time monitoring, portable diagnostic devices, and integration of AI with wearable technologies. Research on AI and ML in pediatric OSA is expanding rapidly, showing strong potential to improve early diagnosis and individualized management. Strengthening collaboration and translating these innovations into practice will be vital for advancing pediatric sleep care.