Advances in Machine Learning Prediction Models for the Screening of Obstructive Sleep Apnea in Adults

This is the authors' abstract. We don't add key points for this paper.

Obstructive sleep apnea (OSA) is a global health problem. Patients with OSA may experience the upper airway collapsing during sleep, resulting in decreased oxygen saturation and sleep disruption, which is characterized by hypoxemia and sleep fragmentation, thereby reducing sleep quality and harming quality of life. In addition, OSA is associated with the occurrence of a variety of systemic diseases, which brings a huge burden to public health. Therefore, timely diagnosis of OSA is crucial. Polysomnography (PSG) is the most accurate method for diagnosing OSA at present, which can be used to determine the severity of sleep apnea and to monitor therapeutic efficacy. However, the PSG is difficult to be popularized because of its cumbersome operation, patients' non-compliance, and expensive medical expenses. Therefore, it is imperative to find a convenient and fast OSA diagnosis method. In recent years, the development of machine learning prediction models and their application in the medical field have provided a new method for OSA severity diagnosis, making it possible to identify OSA severities efficiently and accurately. The purpose of this paper is to review relevant research on machine learning prediction models for OSA severity diagnosis and to provide sleep specialists with recommendations for more effective early identification and diagnosis of OSA. In addition, the challenges faced by machine learning at the level of diagnostic applications are summarized and future trends are envisioned.