A Survey on Detection and Prediction Methods for Sleep Apnea

Abstract

Sleep disorders are common health issues that can affect the multiple aspects of life. Sleep apnea (SA) is the most common sleep disorder, and it is described as a reduction or cessation of airflow to the lungs during sleep. This disorder is usually diagnosed and tested using polysomnography (PSG) in a special laboratory. However, this method is costly, inconvenient, time consuming, often causes anxiety for the patient, and the equipment cannot be moved from the lab. There are several methods suggested to address these shortcomings, including testing and analysis at the patient‘s home and the sleep laboratory, by using sensors to detect physiological signals that can be automatically analysed based on specific algorithms. The purpose of this study was to explore the previous works related to SA in such a way that highlights the methods of detection or diagnoses that use different sensors. The researcher aimed to adopt algorithms and make a comparison between those works to explain the accuracy, sensitivity, and specificity of SA detection and prediction. This review was conducted to provide information for those researchers who want to implement algorithms for detection and predication of sleep apnea event (SAE). Limitations and challenges are also discussed.