Quantification of Depression Disorder Using EEG Signal

摘要

Depression is one of the most common psychological disorders, which has been an ever-growing concern in science world. In mental health questionnaires, including Beck's questionnaire, the patient is assigned a numerical indicator. Researches have shown that level of depression in people is associated with structural changes in the brain, therefore, by analyzing brain signals, it is possible to detect depression level. This paper presents a method that estimates the beck's index of each subject by extracting specific features from the patient's EEG signal. In order to quantify depression, an algorithm has been designed and implemented that uses membership values obtained from the fuzzy classifier and the support vector machine(SVM). In this approach, desirable results have been obtained indicating that the proposed algorithm has a good ability to determine the numerical index for depression. The results were obtained with a percent relative difference(PRD) of 5% and a Pearson correlation of 0.92. The results of the experiments show that the estimated numerical value of the designed system is of high correlation and low amount of PRD in comparison with the original beck number, related to each person.