Classification of Depression Through Resting-State Electroencephalogram as a Novel Practice in Psychiatry: Review

Background Machine learning applications in health care have increased considerably in the recent past, and this review focuses on an important application in psychiatry related to the detection of depression. Since the advent of computational psychiatry, research based on functional magnetic resonance imaging has yielded remarkable results, but these tools tend to be too expensive for everyday clinical use. Objective This review focuses on an affordable data-driven approach based on electroencephalographic recordings. Web-based applications via public or private cloud-based platforms would be a logical next step. We aim to compare several different approaches to the detection of depression from electroencephalographic recordings using various features and machine learning models. Methods To detect depression, we reviewed published detection studies based on resting-state electroencephalogram with final machine learning, and to predict therapy outcomes, we reviewed a set of interventional studies using some form of stimulation in their methodology. Results We reviewed 14 detection studies and 12 interventional studies published between 2008 and 2019. As direct comparison was not possible due to the large diversity of theoretical approaches and methods used, we compared them based on the steps in analysis and accuracies yielded. In addition, we compared possible drawbacks in terms of sample size, feature extraction, feature selection, classification, internal and external validation, and possible unwarranted optimism and reproducibility. In addition, we suggested desirable practices to avoid misinterpretation of results and optimism. Conclusions This review shows the need for larger data sets and more systematic procedures to improve the use of the solution for clinical diagnostics. Therefore, regulation of the pipeline and standard requirements for methodology used should become mandatory to increase the reliability and accuracy of the complete methodology for it to be translated to modern psychiatry.

[1]  Yongtian He,et al.  Deep learning for electroencephalogram (EEG) classification tasks: a review , 2019, Journal of neural engineering.

[2]  Hiie Hinrikus,et al.  Methods for classifying depression in single channel EEG using linear and nonlinear signal analysis , 2018, Comput. Methods Programs Biomed..

[3]  James F. Cavanagh,et al.  Larger Error Signals in Major Depression are Associated with Better Avoidance Learning , 2011, Front. Psychology.

[4]  Hiie Hinrikus,et al.  Spectral Asymmetry and Higuchi's Fractal Dimension Measures of Depression Electroencephalogram , 2013, Comput. Math. Methods Medicine.

[5]  Oliver Faust,et al.  DEPRESSION DIAGNOSIS SUPPORT SYSTEM BASED ON EEG SIGNAL ENTROPIES , 2014 .

[6]  Alej,et al.  Complexity and Irregularity in the Brain Oscillations of Depressive Patients: A Systematic Review , 2017 .

[7]  Mitsuo Kawato,et al.  Computational neuroscience approach to biomarkers and treatments for mental disorders , 2017, Psychiatry and clinical neurosciences.

[8]  Aviv Segev,et al.  A Pilot Study of Possible Easy-to-Use Electrophysiological Index for Early Detection of Antidepressive Treatment Non-Response , 2017, Front. Psychiatry.

[9]  HosseinifardBehshad,et al.  Classifying depression patients and normal subjects using machine learning techniques and nonlinear features from EEG signal , 2013 .

[10]  Xin Zhao,et al.  Neural complexity in patients with poststroke depression: A resting EEG study. , 2015, Journal of affective disorders.

[11]  J. Concato,et al.  A simulation study of the number of events per variable in logistic regression analysis. , 1996, Journal of clinical epidemiology.

[12]  Bin Hu,et al.  A Pervasive Approach to EEG-Based Depression Detection , 2018, Complex..

[13]  Karl J. Friston,et al.  Computational psychiatry , 2012, Trends in Cognitive Sciences.

[14]  Diego Urgelés Puértolas,et al.  Data source analysis in mood disorder research , 2018 .

[15]  Alexander Cerquera,et al.  Non-linear EEG analyses predict non-response to rTMS treatment in major depressive disorder , 2014, Clinical Neurophysiology.

[16]  Yi-Hung Liu,et al.  Major Depression Detection from EEG Signals Using Kernel Eigen-Filter-Bank Common Spatial Patterns , 2017, Sensors.

[17]  Aamir Saeed Malik,et al.  Electroencephalogram (EEG)-based computer-aided technique to diagnose major depressive disorder (MDD) , 2017, Biomed. Signal Process. Control..

[18]  J. Jonides,et al.  Depression, rumination and the default network. , 2011, Social cognitive and affective neuroscience.

[19]  Marcia K. Johnson,et al.  Cross-trial prediction of treatment outcome in depression: a machine learning approach. , 2016, The lancet. Psychiatry.

[20]  R Jouvent,et al.  Decrease of complexity in EEG as a symptom of depression. , 1994, Neuroreport.

[21]  Charles E McCulloch,et al.  Relaxing the rule of ten events per variable in logistic and Cox regression. , 2007, American journal of epidemiology.

[22]  Subha D. Puthankattil,et al.  Analysis of EEG Signals Using Wavelet Entropy and Approximate Entropy: A Case Study on Depression Patients , 2014 .

[23]  Claire M. Gillan,et al.  What big data can do for treatment in psychiatry , 2017, Current Opinion in Behavioral Sciences.

[24]  E. Başar,et al.  Brain’s alpha activity is highly reduced in euthymic bipolar disorder patients , 2012, Cognitive Neurodynamics.

[25]  N. Daw,et al.  Taking Psychiatry Research Online , 2016, Neuron.

[26]  Aamir Saeed Malik,et al.  A machine learning framework involving EEG-based functional connectivity to diagnose major depressive disorder (MDD) , 2018, Medical & Biological Engineering & Computing.

[27]  J. Geddes,et al.  Comparative efficacy and acceptability of 12 new-generation antidepressants: a multiple-treatments meta-analysis , 2009, The Lancet.

[28]  C. Beckmann,et al.  Beyond Lumping and Splitting: A Review of Computational Approaches for Stratifying Psychiatric Disorders , 2016, Biological psychiatry. Cognitive neuroscience and neuroimaging.

[29]  R. Tibshirani,et al.  Improvements on Cross-Validation: The 632+ Bootstrap Method , 1997 .

[30]  Damiaan Denys,et al.  Relation Between Structural and Functional Connectivity in Major Depressive Disorder , 2013, Biological Psychiatry.

[31]  Leanne M Williams,et al.  A Cognitive–Emotional Biomarker for Predicting Remission with Antidepressant Medications: A Report from the iSPOT-D Trial , 2015, Neuropsychopharmacology.

[32]  Dragoljub Pokrajac,et al.  The successful discrimination of depression from EEG could be attributed to proper feature extraction and not to a particular classification method , 2020, Cognitive Neurodynamics.

[33]  Donel M. Martin,et al.  Predicting tDCS treatment outcomes of patients with major depressive disorder using automated EEG classification. , 2017, Journal of affective disorders.

[34]  Peter A. Flach,et al.  Machine Learning - The Art and Science of Algorithms that Make Sense of Data , 2012 .

[35]  D. Kupfer,et al.  Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. , 2006, The American journal of psychiatry.

[36]  Tien-Wen Lee,et al.  The implication of functional connectivity strength in predicting treatment response of major depressive disorder: A resting EEG study , 2011, Psychiatry Research: Neuroimaging.

[37]  Ioannis Kalatzis,et al.  Design and implementation of an SVM-based computer classification system for discriminating depressive patients from healthy controls using the P600 component of ERP signals , 2004, Comput. Methods Programs Biomed..

[38]  Aleksandra Vuckovic,et al.  Modeling the relationship between Higuchi’s fractal dimension and Fourier spectra of physiological signals , 2012, Medical & Biological Engineering & Computing.

[39]  J. Reilly,et al.  A pilot study to determine whether machine learning methodologies using pre-treatment electroencephalography can predict the symptomatic response to clozapine therapy , 2010, Clinical Neurophysiology.

[40]  Daniël Lakens,et al.  Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs , 2013, Front. Psychol..

[41]  Ahmad Khodayari-Rostamabad,et al.  A machine learning approach using EEG data to predict response to SSRI treatment for major depressive disorder , 2013, Clinical Neurophysiology.

[42]  Olaf Sporns,et al.  Disturbed resting state EEG synchronization in bipolar disorder: A graph-theoretic analysis☆ , 2013, NeuroImage: Clinical.

[43]  Ron Kohavi,et al.  A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection , 1995, IJCAI.

[44]  H. Adeli,et al.  Fractality analysis of frontal brain in major depressive disorder. , 2012, International journal of psychophysiology : official journal of the International Organization of Psychophysiology.

[45]  Miodrag Stokić,et al.  Nonlinear analysis of EEG complexity in episode and remission phase of recurrent depression , 2018, International journal of methods in psychiatric research.

[46]  Bijan Raahemi,et al.  Data mining EEG signals in depression for their diagnostic value , 2015, BMC Medical Informatics and Decision Making.

[47]  A Rogier T Donders,et al.  Penalized maximum likelihood estimation to directly adjust diagnostic and prognostic prediction models for overoptimism: a clinical example. , 2004, Journal of clinical epidemiology.

[48]  Aamir Saeed Malik,et al.  A wavelet-based technique to predict treatment outcome for Major Depressive Disorder , 2017, PloS one.

[49]  Shreya Bhat,et al.  Automated Classification of Depression Electroencephalographic Signals Using Discrete Cosine Transform and Nonlinear Dynamics , 2015 .

[50]  N. Jaworska,et al.  Pre-treatment EEG signal variability is associated with treatment success in depression , 2017, NeuroImage: Clinical.

[51]  M. Kenward,et al.  An Introduction to the Bootstrap , 2007 .

[52]  Josselin Houenou,et al.  A meta-analysis of whole-brain diffusion tensor imaging studies in bipolar disorder , 2011, Progress in Neuro-Psychopharmacology and Biological Psychiatry.

[53]  Kenji Doya,et al.  Identification of depression subtypes and relevant brain regions using a data-driven approach , 2018, Scientific Reports.

[54]  R. Whelan,et al.  When Optimism Hurts: Inflated Predictions in Psychiatric Neuroimaging , 2014, Biological Psychiatry.

[55]  M. Ahmadlou,et al.  Brain activity of women is more fractal than men , 2013, Neuroscience Letters.

[56]  Mohamed-Hamza Ibrahim,et al.  Leveraging Machine Learning Approaches for Predicting Antidepressant Treatment Response Using Electroencephalography (EEG) and Clinical Data , 2019, Front. Psychiatry.

[57]  C. Peng,et al.  What is physiologic complexity and how does it change with aging and disease? , 2002, Neurobiology of Aging.

[58]  Ramesh Rajan,et al.  Combining complexity measures of EEG data: multiplying measures reveal previously hidden information , 2015, F1000Research.

[59]  Leanne Togher,et al.  Measuring the social interactions of people with traumatic brain injury and their communication partners: The adapted Kagan scales , 2010 .

[60]  Wlodzimierz Klonowski,et al.  From conformons to human brains: an informal overview of nonlinear dynamics and its applications in biomedicine , 2007, Nonlinear biomedical physics.

[61]  Subha D. Puthankattil,et al.  CLASSIFICATION OF EEG SIGNALS IN NORMAL AND DEPRESSION CONDITIONS BY ANN USING RWE AND SIGNAL ENTROPY , 2012 .

[62]  H. Adeli,et al.  of Depressive Women and Men Spatiotemporal Analysis of Relative Convergence of EEGs Reveals Differences Between Brain Dynamics , 2013 .

[63]  Joel E. W. Koh,et al.  A Novel Depression Diagnosis Index Using Nonlinear Features in EEG Signals , 2015, European Neurology.

[64]  Andrew Y. Ng,et al.  Preventing "Overfitting" of Cross-Validation Data , 1997, ICML.

[65]  Milena Cukic,et al.  The successful discrimination of depression from EEG could be attributed to proper feature extraction and not to a particular classification method , 2018, Cognitive Neurodynamics.