Predicting Sleep Disorder using Raw Multi-Channel EEG signal

Abstract

Accurate sleep stage scoring and finding relevant feature form multi-channel EEG signal form different subject (healthy and unhealthy) is complex task. In recent years, deep learning, a type of machine learning that involves training ar tificial neural networks on large data sets, has shown promise for improving the accuracy and reliability of sleep stage scoring. This approach involves analyzing the pre-processed raw data and extract important feature and try to find informa tion and based on that we predict if a person has sleep disorder or not. By using deep learning to train our model the extracted data set we reprocessed and find promising result, it is possible to develop more accurate algorithms or models for automatic prediction of sleep disorder and other abnormal activity in brain