An EEG majority vote based BCI classification system for discrimination of hand motor attempts in stroke patients

Stroke patients have symptoms of cerebral functional disturbance that could aggressively impair patient's physical mobility, such as hand impairments. Although rehabilitation training from external devices is beneficial for hand movement recovery, for initiating motor function restoration purposes, there are still valuable research merits for identifying the side of hands in motion. In this preliminary study, we used an electroencephalogram (EEG) dataset from 8 stroke patients, with each subject conducting 40 EEG trials of left motor attempts and 40 EEG trials of right motor attempts. Then, we proposed a majority vote based EEG classification system for identifying the side in motion. In specific, we extracted 1-50 Hz power spectral features as input for a series of well-known classification models. The predicted labels from these classification models were compared and a majority vote based method was applied, which determined the finalised predicted label. Our experiment results showed that our proposed EEG classification system achieved 99:83 ± 0:42% accuracy, 99:98 ± 0:13% precision, 99:66 ± 0:84% recall, and 99:83 ± 0:43% f-score, which outperformed the performance of single well-known classification models. Our findings suggest that the superior performance of our proposed majority vote based EEG classification system has the potential for stroke patients' hand rehabilitation.