EEG-Based Cognitive Response Classification Using Machine Learning for Communication Support in Amyotrophic Lateral Sclerosis

ანოტაცია

Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder characterized by the loss of voluntary motor function, eventually leading to severe communication impairment. In advanced stages, patients may lose the ability to speak, write, or use conventional assistive technologies. Brain–computer interface (BCI) systems provide a promising alternative by enabling communication through direct interpretation of neural activity.

The present study focuses on the acquisition, preprocessing, and classification of electroencephalographic (EEG) signals corresponding to binary cognitive responses (“Yes” and “No”). EEG data were collected using a CleaveLab - BioRadio system in a controlled laboratory environment. The signals were processed and transformed into informative features, including statistical time-domain characteristics.

Machine learning methods were applied to classify cognitive states based on extracted features. The study emphasizes the importance of data-driven approaches in handling non-stationary EEG signals and improving classification performance. The results demonstrate the feasibility of using simple yet effective features combined with machine learning algorithms for communication-oriented BCI applications in ALS.