Detection and Analysis of Brain Disorders Using Artificial Neural Networks

The rapid improvement in technology enables electroencephalogram (EEG) techniques to detect a diverse range of brain disorders easily. Knowledge of various signal processing techniques for the analysis of EEG signals is extremely essential. Raw EEG signals comprise noise and artifacts that modify the appearance of the EEG, rendering clinical interpretation incorrect and difficult due to inaccuracy and distortion; thus, denoising of the signal to refine signal quality is an obligation. This study highlights three conditions of the brain: stroke, brain death, and a normal condition. The major objective is to detect the most abnormal conditions of the brain, specifically an EEG reflecting a critical stage. This paper introduces a unique method for the analysis of EEG signals from the three conditions using wavelet transform, Fast Fourier Transform (FFT), and signal classification by Artificial Neural Network (ANN) techniques to obtain highly accurate resultant signals.