Predicting Clinical Response to Transcranial Magnetic Stimulation in Major Depression using Time

Elias Ebrahimzadeh 1,2,\*, Mostafa Asgarinejad 3, Sara Saliminia 4, Servanaz Ashuri 4 and Masoud Saraji 5,6 1 CIPCE, Faculty of Electrical and Computer Engineering, Technical Campus, University of Tehran, Tehran, Iran 2 Faculty of Cognitive Sciences, Fundamental Sciences Research Institute (IPM), Tehran, Iran 3 Institute of Cognitive Science Studies, Tehran, Iran 4 Department of Biomedical Engineering, Faculty of Electrical Engineering, Tehran Payam Noor University, Tehran, Iran 5 Behavioral and Neuroscience Doctoral Program, Rutgers University, Newark, NJ, 07102, USA 6 Image Analysis Laboratory, Department of Radiology, Henry Ford Hospital, Detroit, MI, USA Repetitive cranial magnetic stimulation (rTMS) is defined as a non-invasive method of brain stimulation for both diagnostic and therapeutic purposes....rTMS can effectively stimulate brain neurons and increase brain plasticity, which is particularly useful in the fields of psychiatry and neuroscience....Biomarkers that predict clinical outcomes to increase the prediction of depression....Electroencephalogram (EEG) is a non-invasive neurophysiological test that shows promise as a sensitive biomarker of therapeutic effects.... The aim of our study was to investigate a novel nonlinear index of resting-state EEG activity as a predictor of clinical outcome and to compare its predictive capacity with traditional frequency-based indices. (DLPFC) were treated... EEG signals were first analyzed using the ICA algorithm and the extracted components were processed by time-frequency analysis... Then we compare the participants' depression severity before, after and 2 months after the end of the last treatment session using the proposed rTMS treatment method... Absolute powers (APs), band powers (BPs) and theta band entropies and Betas (BAs), extracted from EEG, can be used as features to classify changes in patients and normal cases after application. Clinically, the EEG signal is classified into depression and normal.... The results showed that 78.37%, 74.32% and 82.43% for artificial neural network (ANN), nearest neighbor (KNN) and support vector machine (SVM) classifiers, respectively, which indicate the superiority of the presented method in similar electrophysiological studies.. with major depression.... Our data shows that the time-frequency index has a better predictive performance compared to band indices. Traditional frequency provides.... Our findings warrant further research on EEG-based biomarkers in depression.