EEG and ECG Signal Based Depression Detection Using Machine Learning

Abstract

The COVID-19 pandemic is the century's most memorable event. Stress was a common problem for many during the pandemic. Prolonged stress can cause serious mental and even physical health problems. Manually detecting depression takes a great deal of experience, patience, and time. The current approach uses EEG and ECG readings to identify and analyze depression. The extraction and choosing methods for classification and degrading approaches, as well as combination methodology, are included in the system layout calculations and strategies. After being retrieved, the EEG and ECG features are forwarded for categorization. From ECG signals, the ST segment, P wave, and QRS wave are extracted as characteristics. Hjorth activity (HA), standard deviation, entropy, and band power alpha are the most important characteristics that are analyzed from EEG signals. The Long Short-Term Memory (LSTM) autoencoder and RNN deep learning model approach were used for depression analysis for ECG signals and Support Vector Machine (SVM) and Convolutional Neural Network (CNN) classification methods are used for EEG signals.