POLYSOMNOGRAPHY VERSUS PORTABLE DEVICES AND MOBILE APPLICATIONS IN THE DIAGNOSIS OF OBSTRUCTIVE SLEEP APNEA: A NARRATIVE REVIEW OF CURRENT EVIDENCE

Background: Obstructive sleep apnea (OSA) represents a significant public health challenge affecting millions worldwide, yet remains largely underdiagnosed due to limitations in traditional diagnostic approaches. While polysomnography (PSG) remains the gold standard for OSA diagnosis, emerging portable devices and mobile applications offer promising Ralternatives for improving accessibility and reducing healthcare burden. Objective: This narrative review synthesizes current evidence comparing the diagnostic accuracy and clinical utility of PSG with portable monitoring devices and mobile health applications in OSA detection, examining their role in advancing health equity and leveraging technology for improved healthcare delivery. Methods: A comprehensive analysis of recent literature from 2016-2025 was conducted, examining studies that evaluated portable sleep monitoring devices, wearable technologies, smartphone applications, and artificial intelligence-driven solutions against PSG as the reference standard. The review focused on diagnostic accuracy metrics, technological innovations, and clinical implementation considerations. Results: Portable monitoring devices demonstrated varying degrees of diagnostic accuracy, with home sleep apnea tests (HSATs) showing sensitivities ranging from 0.72-0.95 and specificities from 0.76-0.96 compared to PSG. Wearable devices utilizing photoplethysmography, accelerometry, and artificial intelligence algorithms achieved area under the curve (AUC) values between 0.80-0.95. Novel approaches including smartphone-based acoustic monitoring, radar technology, and bed-mounted sensors showed promising results with sensitivities exceeding 0.85 in several studies. Artificial intelligence integration significantly enhanced diagnostic performance across multiple device categories. Discussion: The evolution of portable OSA diagnostic technologies represents a paradigm shift toward accessible, cost-effective screening and monitoring solutions. While PSG maintains superior diagnostic precision, portable devices offer substantial advantages in terms of patient comfort, cost-effectiveness, and scalability for population-based screening. The integration of artificial intelligence and machine learning algorithms has notably improved the diagnostic accuracy of these technologies. Conclusion: Portable devices and mobile applications demonstrate significant potential for revolutionizing OSA diagnosis and management, particularly in addressing healthcare disparities and improving access to sleep medicine services. However, careful consideration of device limitations, patient selection criteria, and clinical context remains essential for optimal implementation in healthcare delivery systems.