An Ensemble Method for EEG-based Texture Discrimination during Open Eyes Active Touch

• DOI: 10.48084/etasr.6455
• Publication Date: 2024-02-12T06:15:26Z
• Authors: Andreas Miltiadous; Vasileios Aspiotis; Dimitrios Peschos; Katerina D. Tzimo...; Al Husein Sami Ab...; Nikolaos Giannakeas; Alexandros Tzallas
• Keywords: Neurophysiology; Robustness

Touch sensation is a key modality that allows humans to understand and interact with their environment. More often than not, touch depends on vision accumulate validate the received information. The ability distinguish between materials surfaces through active consists of complex neurophysiological operations. To unveil functionality these operations, neuroimaging research tools are employed, electroencephalography being most used. In this paper, we attempt brain states when touching different natural textures (smooth, rough, liquid). Recordings were obtained commercially available EEG wearable device. Time frequency-based features extracted, transformed PCA decomposition, an ensemble classifier combining Random Forest, Support Vector Machine, Neural Network was utilized. High accuracy scores 79.64% for four-class problem 89.34% three-class (Null-Rough-Water) accordingly achieved. Thus, methodology's robustness indicates its classify under haptic stimuli.