Objective: Depression is a mental disorder that causes emotional changes and even suicide. However, there is still a lack of objective physiological data to support current clinical depression diagnosis. Accurate computer-aided diagnosis systems are becoming more and more crucial and urgent for future depression diagnosis. The purpose of this study is to analyze the electroencephalogram (EEG) regularity of depression using fuzzy measure entropy (FMEn), and thus to explore its role in the computer-aided diagnosis of depression.
Approach: Three-channel EEG signals among 35 subjects (divided into two groups according to the severity of the disease) were recorded in this study. First, the frontal delta, theta, alpha and beta frequency bands were extracted after preprocessing, and the sample entropy (SEn) and the FMEn were calculated. Then, the difference between the two groups and the correlation between the entropy values and the Hamilton Depression Rating Scale scores were analyzed using statistical analysis. Finally, the results of FMEn were compared with those of SEn.
Main results: A better statistically significant difference between the two groups using FMEn was revealed, with p < 0.01 in the theta and alpha bands. In terms of SEn, only SEn_Fp2 in the delta band, SEn_Fp2 in the theta band and SEn_Fp1 in the alpha band performed better, showing significant differences with p = 0.0006, p = 0.002 and p = 0.0114.
Significance: These findings suggest that frontal EEG signal complexity analysis with depression using FMEn might be more sensitive than that using SEn. FMEn could be considered as a promising biomarker for future clinical depression detection.