Brain–Computer Interface Classifier for Wheelchair Commands Using Neural Network With Fuzzy Particle Swarm Optimization

摘要: This paper presents the classification of a three-class mental task-based brain-computer interface (BCI) that uses Hilbert-Huang transform for features extractor and fuzzy particle swarm optimization with cross-mutated-based artificial neural network (FPSOCM-ANN) classifier. The experiments were conducted on five able-bodied subjects patients tetraplegia using electroencephalography signals from six channels, different time-windows data examined to find highest accuracy. For practical purposes, best two channel combinations chosen presented. three relevant tasks used BCI letter composing, arithmetic, Rubik's cube rolling forward, these are associated wheelchair commands: left, right, respectively. An additional eyes closed task was collected testing on-off commands. results show dominant alpha wave during closure average accuracy above 90%. accuracies lower compared subjects; however, this improved by increasing duration time-windows. FPSOCM-ANN provides genetic algorithm-based (GA-ANN) tasks-based classifications achieved 7-s time-window: 84.4% 77.4% (GA-ANN). More comparisons feature extractors classifiers included. two-channel classification, channels O1 C4, followed second at P3 O2, third C3 O2. Mental arithmetic most correctly classified task, forward composing.