EEG-based mild depression detection using multi-objective particle swarm optimization

This paper describes a mild depression detection method based on the EEG. Firstly, we present a comprehensible function to categorize volunteers by linear discriminant analysis (LDA). Then, a novel multi-objective particle swarm optimization (MOPSO) for depression detection is proposed, minimum the number of misclassification, minimize the internal distance and maximize the external distance are all included in the objectives of our model. Finally, the results of the experiment with 6 volunteers indicate that accuracies achieve 100%, and our method maybe good candidates for usage in portable systems for mild depression detection.