A New Approach for Detecting Sleep Arousal from Single-Channel EMG Signal

Sleep arousal is a brief waking that occurs while a person is asleep and usually occurs without conscious awareness. The muscle relaxation at the back of the throat causes the airway to narrow or close during breathing, leading to decreased oxygen levels in the blood. The brain responds to this by briefly waking the person to reopen the airway, which is known as sleep arousal. The study aims to develop an algorithm using single-channel electromyography (EMG) signals for accurately detecting arousal occurrences. The algorithm was designed in such a way that took single-channel EMG signal as the input and after rigorous automatic operation, the output was produced as the number of arousal and normal epochs. The algorithm was trained and tested using 30 and 20 whole-night EMG records. The sensitivity, specificity, and accuracy of 90%, 98%, and 98% respectively were achieved from the testing dataset. The performance was much better than the other existing literature. This automatic approach of sleep arousal detection can significantly contribute to the process of sleep apnea diagnosis.