Comparing resting state and task-based EEG using machine learning to predict vulnerability to depression

Abstract

Major depressive disorder affects a large portion of the population and levies a huge societal burden. It has serious consequences like decreased productivity and reduced quality of life, hence there is considerable interest in understanding and predicting it for example using neural measures. As most of these studies have either explored resting state EEG (rs-EEG) data or task-based EEG data but not both, we seek to compare their respective efficacy. We work with data from non-clinicallydepressed individuals who score higher and lower on the depression scale and hence are more and less vulnerable to depression, respectively. Forty participants volunteered for the study. Questionnaires and EEG data were collected from participants. We found that in rs-EEG, people who are more vulnerable to depression had on average increased activity in the right temporal channel, and decreased activity in the left fronto-central and right occipital channels for raw data (rs-EEG). Intask-based EEG data, an increased activity in the central part of the brain for individuals with low vulnerability and an increased activity in right temporal, occipital and parietal regions in individuals more vulnerable to depression were found. In an attempt to predict vulnerability (high/low) to depression, we found that a Long Short Term Memory model gave the maximum accuracy of 91.42 in delta wave for task-based data whereas 1D-Convolution neural network gave the maximum accuracy of 98.06 corresponding to raw rs-EEG data. Hence if one has to look at the primary question of which data will be good for predicting vulnerability to depression, rs-EEG seems to be better than task-based EEG data. However, if mechanisms driving depressionlike rumination or stickiness are to be understood, task-based data may be more effective. Higuchi fractal dimension, phase lag index, correlation and coherence features were also found to be the most important features for predicting vulnerability todepression using rs-EEG.

Analyzing Fractal Dimension in Electroconvulsive Therapy: Unraveling Complexity in Structural and Functional Neuroimaging

“…HFD analysis has proven invaluable in unraveling the complexities and altered connectivity patterns within the brains of individuals affected by various neurological disorders including migraine and neurodegenerative diseases (Djuričić et al, 2023; Garehdaghi and Sarbaz, 2023; Porcaro et al, 2020; Porcaro et al, 2022; Varley et al, 2020). Surprisingly, while HFD has gained traction in exploring depressive disorders, primarily using ECG (George et al, 2023) and EEG data (Kaushik et al, 2023; Kawe et al, 2019; Lord and Allen, 2023), it has remained largely unexplored in the realm of neuropsychiatric disorders and functional MRI data. Transferring HFD to functional neuroimaging presents an intriguing opportunity to harness HFD’s potential for gaining fresh insights into the landscape of these complex conditions.…”