A novel depression detection method based on pervasive EEG and EEG splitting criterion

Depression is a mental disorder characterized by persistent occurrences of lower mood states in the affected person. According to the study of World Health Organization (WHO), depression will become the second largest cause of illness threatening the life of human beings in 2020, so early detection, early diagnosis and early treatment of depression is very important to save the health and life of human beings. In order to alleviate the damage caused by depression and make early detection, early diagnosis and early treatment of depression, a portable and accurate depression detection and diagnosis method is most necessary. Due to the highly complexity, nonlinearity and non-stationarity of electroencephalogram (EEG) data in nature, we present a novel method for pervasive EEG-based detection and diagnosis of depression with the resting state eye-closed EEG data of Fp1, Fpz and Fp2 locations of scalp electrodes, which are closely related to emotion, collected through three-electrode pervasive EEG collection device in this paper. Experiment has been conducted and totally 170 (81 depressive patients and 89 normal subjects) subjects' pervasive EEG data have been collected in resting state and eye-closed. Then, Support Vector Machine (SVM) is utilized to analyze the pervasive EEG data and the average accuracy reaches 83.07%. After Friedman Test and post-hoc two-tailed Nemenyi Test, we propose a splitting criterion for pervasive EEG. The data analysis experimental results show that the proposed method for detecting and diagnosing depression is effective and convenient, and it also demonstrate that the three-electrode pervasive EEG collection device has broad prospects in depression detection and diagnosis.