Role of Artificial Intelligence in Paediatric Sleep Medicine

Ashok Kumar Angamuthu¹, Venkateswaramurthy Nallasamy^2*, Vidhya Lekshmi Krishnan²and Chitra Thara Sughumaran²

¹Medical Coder CPC,Cotiviti, Coimbatore, Tamil Nadu, India

²Department of Pharmacy Practice, J.K.K.Nattraja College of Pharmacy, Kumarapalayam, Tamil Nadu, India.

Corresponding Author E-mail:nvmurthi@gmail.com

ABSTRACT: Pediatric sleep apnea, including obstructive (OSA) and central sleep apnea (CSA), can severely affect a child's physical health and cognitive function. Traditionally, polysomnography (PSG) has been the gold standard for diagnosis, but its high cost, invasiveness, and difficulty in pediatric use present significant challenges. Advances in Artificial Intelligence (AI) and Machine Learning (ML) are now offering transformative solutions, enhancing diagnostic accuracy, personalizing treatments, and increasing accessibility. AI-powered home sleep apnea tests (HSATs) provide a non-invasive alternative by analyzing key physiological signals such as oxygen saturation (SpO₂), electrocardiograms (ECG), and nasal airflow pressure (NAP) to detect apneic events and assess severity. These AI-driven tools offer greater convenience compared to traditional PSG. Additionally, ML models show promise in predicting adherence to therapies like positive airway pressure (PAP) for OSA, while advanced AI algorithms are improving CSA detection by analyzing complex physiological patterns more effectively. Cutting-edge innovations, including transformer models and edge AI, are enabling real-time sleep staging tailored management, making diagnostic tools more efficient and widely available. By integrating AI-driven solutions, healthcare providers can offer earlier and more accurate diagnoses, leading to timely interventions that improve long-term health outcomes for children. Despite these advancements, further validation through large-scale clinical studies is necessary to establish AI's reliability across diverse pediatric populations. With continued research and refinement, these technologies could become standard tools for detecting and managing pediatric sleep apnea, paving the way for a future where diagnosis is more accessible, cost-effective, and child-friendly.

KEYWORDS: Artificial intelligence (AI); Convolutional neural networks (CNNs); Central sleep apnea (CSA); Machine learning (ML); Obstructive sleep apnea (OSA); Pediatric sleep apnea; Polysomnography (PSG)

[ HTML Full Text]