Pervasive EEG diagnosis of depression using Deep Belief Network with three-electrodes EEG collector

According to the World Health Organization, it is predicted that in 2020, depression will become the second largest illness threatening the health of mankind. In order to alleviate the worldwide damage caused by depression, a portable and accurate diagnosing technique is the most essential. This research uses three-electrode pervasive EEG collector to collect EEG data from Fp1, Fp2, and Fpz as locations of scalp electrodes, since these locations are closely related to emotions, and uncovered by hair. Special designed experiment has been conducted and totally 178 subjects' EEG data have been collected. Then the research uses KNN (k-Nearest Neighbor), SVM (Support Vector Machine), ANN (Artificial Neuro Network) and DBN (Deep Belief Network) to analyze the data. The results show DBN performed better than traditional methods using shallow algorithms. Moreover, the results suggested the absolute power of beta wave is a valid characteristic, which could be used for detection of depression. The accuracy reached 78.24% using the combination of DBN and the absolute power of beta wave. This research proves the feasibility of smaller-size pervasive system for depression diagnosis.

[1]  Demetrios S. Alexopoulos,et al.  PSYCHOMETRIC PROPERTIES OF EYSENCK PERSONALITY QUESTIONNAIRE-REVISED (EPQ-R) SHORT SCALE IN GREECE , 2004 .

[2]  Therese Dowswell,et al.  Interventions (other than pharmacological, psychosocial or psychological) for treating antenatal depression. , 2013, The Cochrane database of systematic reviews.

[3]  Daniel J Buysse,et al.  Quantification of subjective sleep quality in healthy elderly men and women using the Pittsburgh Sleep Quality Index (PSQI) , 1991, Sleep.

[4]  U. Rajendra Acharya,et al.  Entropies for detection of epilepsy in EEG , 2005, Comput. Methods Programs Biomed..

[5]  R. Spitzer,et al.  Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. , 1999, JAMA.

[6]  Yufei Huang,et al.  A Deep Learning method for classification of images RSVP events with EEG data , 2013, 2013 IEEE Global Conference on Signal and Information Processing.

[7]  C. Dennis,et al.  Interventions (other than pharmacological, psychosocial or psychological) for treating antenatal depression. , 2008, The Cochrane database of systematic reviews.

[8]  Matthew Sunderland,et al.  Psychometric Comparison of the Generalized Anxiety Disorder Scale-7 and the Penn State Worry Questionnaire for Measuring Response during Treatment of Generalised Anxiety Disorder , 2011, Cognitive behaviour therapy.

[9]  Amy Loutfi,et al.  Sleep Stage Classification Using Unsupervised Feature Learning , 2012, Adv. Artif. Neural Syst..

[10]  D. Tucker.  Lateral brain function, emotion, and conceptualization. , 1981, Psychological bulletin.

[11]  Aidong Zhang,et al.  A Novel Semi-Supervised Deep Learning Framework for Affective State Recognition on EEG Signals , 2014, 2014 IEEE International Conference on Bioinformatics and Bioengineering.

[12]  I. Levav,et al.  Israeli PERI life event scale: ratings of events by a community sample. , 1981, Israel journal of medical sciences.

[13]  Razvan C. Bunescu,et al.  A Shortest Path Dependency Kernel for Relation Extraction , 2005, HLT.

[14]  Juri D. Kropotov,et al.  Independent component approach to the analysis of EEG recordings at early stages of depressive disorders , 2010, Clinical Neurophysiology.

[15]  J. Bhattacharya,et al.  Complexity analysis of spontaneous EEG. , 2000, Acta neurobiologiae experimentalis.

[16]  Juri D Kropotov,et al.  EEG Power Spectra at Early Stages of Depressive Disorders , 2009, Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society.

[17]  Gro Harlem Brundtland,et al.  Mental Health: New Understanding, New Hope , 2001 .

[18]  Pasin Israsena,et al.  EEG-Based Emotion Recognition Using Deep Learning Network with Principal Component Based Covariate Shift Adaptation , 2014, TheScientificWorldJournal.

[19]  J. L. Mata,et al.  51. The international affective digitized sounds (IADS): Their use in the field of affective neuroscience , 2008, Clinical Neurophysiology.

[20]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[21]  Christoph Lehmann,et al.  Application and comparison of classification algorithms for recognition of Alzheimer's disease in electrical brain activity (EEG) , 2007, Journal of Neuroscience Methods.

[22]  D. Sheehan,et al.  The Mini-International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. , 1998, The Journal of clinical psychiatry.

[23]  H Berger,et al.  On the electroencephalogram of man. Sixth report. , 1969, Electroencephalography and clinical neurophysiology.

[24]  Zhangang Han,et al.  Feature extraction of EEG signals from epilepsy patients based on Gabor Transform and EMD Decomposition , 2010, 2010 Sixth International Conference on Natural Computation.