EEG Signal for Diagnosing Diseases using Machine Learning

Abstract

Alzheimer's disease is a chronic neurodegenerative disease that usually starts slowly and worsens over time. Alzheimer’s is an irreversible, progressive brain disorder that slowly destroys memory and thinking skills, and eventually the ability to carry out the simplest tasks. In most people, symptom first appears in their mid-60s. Studies have reported that electroencephalogram (EEG) signals of Alzheimer’s disease patients usually have less synchronization when compare to healthy people. Changes in EEG signals start at early stage but, clinically, these changes are not easily detected. Early detection of Alzheimer’s can have treatments with more positive outcomes. The aim of this paper is to classify Alzheimer’s disease patients using EEG signal processing in order to support medical doctors in the right diagnosis formulation. The proposed system consists of mainly five steps: Signal Acquisition, Pre-processing, Feature Extraction, Feature Selection, and Classification. The Signal Acquisition makes use of EEG dataset, Band-pass-filtering is used in the pre-processing stage to get artifact free signal. The necessary features are then extracted from EEG signals using Wavelet Transform and they are subjected to Principal Component Analysis (PCA) for feature selection. The classification is done using Block Based Neural Network (BBNN). Based on the changes in EEG the structure and internal configuration of BBNN are modified.