Background/Objectives: There have been attempts to detect depression using medical-grade electroencephalograph (EEG) data based on a machine learning approach. EEG has garnered interest as a method for assessing brainwaves by attaching electrodes to the scalp to obtain electrical activity in the brain. Recently, machine learning has been applied to the EEG data to detect depression, with encouraging results. Specifically, studies using medical-grade EEG data have shown that depression can be accurately detected. However, there is a need to expand the range of applications by achieving a score with machine learning using simpler consumer-grade brain wave sensors. At present, a sufficient score has not been achieved.; Methods: To improve the score of depression detection, we quantified various EEG indices to train models such as power spectrum, asymmetry, complexity, and functional connectivity. In addition, feature selection was performed to ensure that the model learns only promising EEG indices for depression detection. The feature selection methods were Light Gradient Boosting Machine (LightGBM) feature importance, mutual information, ReliefF and ElasticNet coefficients. The selected EEG indices were learned by the LightGBM model, which is reported to be as accurate as the latest deep learning models. In cross-validation, the independence of test and training data was ensured to avoid excessively calculated score; Results: The results showed that the Macro F1 score was 91.59%, suggesting that a consumer-grade EEG can detect depression. In addition, analysis of the EEG indices selected by feature selection indicated that the Macro F1 score was about 80% for single EEG indices such as differential entropy in the frequency band β and functional connectivity in the left frontal region in the frequency band 1–128 Hz; Conclusions: Although the data were obtained from a consumer-grade EEG, the results suggest that these EEG indices are promising for detection depression.
Alexithymia is regarded as one of the risk factors for several prevalent mental disorders, and there is a growing need for convenient and objective methods to assess alexithymia. Therefore, this study proposes a method for constructing models to assess alexithymia using machine learning and electroencephalogram (EEG) signals. The explanatory variables for the models were functional connectivity calculated from resting-state EEG data, reflecting the default mode network (DMN). The functional connectivity was computed for each frequency band in brain regions estimated by source localization. The objective variable was defined as either low or high alexithymia severity. Explainable artificial intelligence (XAI) was used to analyze which features the models relied on for their assessments. The results indicated that the classification model suggested effective assessment depending on the threshold used to define low and high alexithymia. The maximum receiver operating characteristic area under the curve (ROC-AUC) score was 0.70. Furthermore, analysis of the classification model indicated that functional connectivity in the theta and gamma frequency bands, and specifically in the Left Hippocampus, was effective for alexithymia assessment. This study demonstrates the potential applicability of EEG signals and machine learning in alexithymia assessment.