An Advanced Design for Depression Analysis through EEG Signal

Shamla Mantri, Dr. Pankaj Agrawal, Dr. Dipti Patil, Dr. Vijay Wadhai, Sharwin Bobde

Abstract: This paper describes a method to diagnose a subject with clinical depression. The approach taken utilizes EEG data from which specific channels are selected to obtain voltage fluctuations from Occipital and Parietal lobes of the brain. Studying these parts of the brain is highly informative about the subjects state of mind when he is awake and help us understand how he processes information, on these basis we can deduce if the subject is depressed or not. These channels are used to extract required alpha wave signals. Then Fast Fourier Transform is applied on the alpha wave and its fundamental frequency and corresponding amplitude of the alpha wave are obtained. The obtained information is used to determine if the test subject is depressed or not using a rule based classifier. The classifier classifies a subject as depressed if the fundamental frequency is below 8 Hz and normal if it is 8 Hz or above.

Keywords: EEG, Depression, Alpha Wave, Discrete Fourier Transform (DFT), Fast Fourier Transform (FFT), Rule Based Classifier.