Sleep-Stage Identification Using Recurrent Neural Network for ECG Wearable-Sensor System

Sleep is one of the most important factors for humans. The condition of the human body, e.g., the electrocardiogram (ECG) signal, can be monitored regularly by means of a portable sensor device. Sleep-stage identification based on ECG and HRV using the neural network algorithm would provide a regular overview of a patient's current health compared to a doctor's diagnosis that can only be made during a hospital visit. This work aimed to develop an accurate model for classifying sleep stages by features of heart rate variability (HRV) extracted from an ECG signal using a recurrent neural network (RNN). An ARNN network is proposed as a solution to the sleep-stage identification problem. This model achieves accuracy of 77.00± 5% across the entire database.