Internet of Things Based Tired Detection using Deep Learning Techniques

Abstract

Sleep is essential for human survival since it helps to restore and maintain our bodies' immune systems and other essential processes. One-third of a person's life is devoted to sleeping, although few are aware of the many positive aspects of this activity. Two distinct types of sleep, REM and NREM, have been identified. A good night's rest is achieved when REM and NREM sleep alternate in a regular pattern. Disruptions to this cycle, whether they originate physiologically or psychologically, have been linked to a variety of health problems. Polysomnography (PSG) equipment is often used in sleep labs inside hospitals to perform sleep studies. A polysomnogram is an in-depth medical technique that records a patient's vital signs while they sleep and necessitates a hospital stay. Clinically, sleep apnea is defined as a breathing disease in which there are periodic pauses in breathing lasting 10 seconds or more that occur more than five times during the night. Sleep apnea may be classified as either Obstructive, Central, or Mixed. The prevalent sleep problem known as obstructive sleep apnea (OSA) is caused by the relaxation of muscles in the upper airway during sleep. The purpose of this study is to provide a technique for screening for Obstructive Sleep Apnea by analysing Heart Rate Variability of Electrocardiogram (ECG) data while the subject is asleep. The goals of this study are to create computational approaches for identifying OSA based on characteristics extracted from Heart Rate Variability (HRV) signals derived from sleep electrocardiograms (ECGs). Physio Net's Apnea-ECG recordings serve as the source for the ECG data.