Prediction of Treatment Outcome to Transcranial Direct Current Stimulation in Major Depression Based on Deep Learning of EEG Data

Major Depressive Disorder (MDD) is a leading cause of disability worldwide. Current first line treatments are antidepressant medication and psychotherapy. However, they have limited effectiveness and there are no biomarkers that can predict treatment response at the individual level. Transcranial direct current stimulation (tDCS) is non-invasive brain stimulation method that is a potential novel treatment for MDD. The present study sought to investigate neural biomarkers for predicting response to tDCS at the individual level using portable EEG. The clinical trial was a double-blinded, placebo-controlled, randomized, superiority trial of home-based tDCS. Participants were randomized to a 10-week course of either active or sham tDCS sessions. Resting state, eyes closed EEG data were acquired at baseline, prior to starting tDCS, and at week 10. EEG data acquisition was conducted using a portable, 4-electrode EEG device (Muse). The baseline EEG data from 21 participants were used to train and test the deep learning models of 1D convolutional neural networks (1DCNNs), Long Short-Term Memory (LSTM), Gated recurrent units (GRU) and the hybrid models combining 1DCNN and LSTM/GRU. A prediction rule was proposed and applied to the classifier outputs of each participant and the treatment outcomes were predicted. Different combinations of power spectral density vectors extracted from the EEG frequency bands of four electrodes were selected to improve the treatment outcome prediction. Using 1DCNN model the work achieved a treatment outcome prediction accuracy 85.7%, with a specificity of 71.4% for predicting treatment remission and sensitivity of 92.8% for predicting residual depressive symptoms, which was based on the combined theta and alpha EEG band power spectral density from the TP10 electrode.

I. Introduction

Major depressive disorder (MDD) represents a significantly prevalent and debilitating mental health disorder, characterized by persistent feelings of a low mood or inability to experience pleasure that is associated with a diminished interest in daily activities and changes in neurovegetative symptoms [1]. It stands as one of the leading contributors to global disability [2], with a lifetime prevalence estimated at 17% [3], thereby constituting a substantial economic burden [4]. Treatment of MDD remains challenging, related to the heterogeneity of the disorder and limited effectiveness of current treatment options [5]. Furthermore, treatments require several weeks duration to evaluate efficacy [6]. In recent years, research has focused on identifying neurological biomarkers of treatment response from electroencephalogram (EEG) data. EEG, being portable, offering high temporal resolution, and is cost-effective, emerging as a potential tool for such investigations. It enables the observation of neurological changes in the brain and has shown promising results in detecting treatment outcomes in MDD [7]. Pretreatment differences in theta band resting activity in the rostral anterior cingulate cortex were observed between responders and non-responders to antidepressant treatment [8]. Increased anterior cingulate cortex activity is reported as a reliable biomarker for antidepressant treatment response [9]. EEG signals from 21 electrodes during eyes closed resting state prior to treatment in 52 MDD participants showed that improvements in depressive severity were negatively related to delta and theta wave activity and positively related to beta activity at frontal recording sites [10]. Moreover, increased frontocentral theta EEG power, a slower anterior individual alpha peak frequency, a larger P300 amplitude, and decreased pre-frontal delta and beta cordance were predictors of non-response to repetitive transcranial magnetic stimulation (rTMS) [11].