classification of selected finger movements with single-channel electromyography by decision tree

Electromyography signals are used in areas such as artificial limbs, rehabilitation, diagnostic medicine, and wearable and control instruments. in this paper, we present a method for classifying the 12 most widely used everyday movements for controlling modern artificial hands and making them smart. the semg signals were recorded from ten volunteers and classified after the noise removal process using butterworth filter 3, classification, and window placement. in this study, 24 features extracted in the time-frequency domain were used. the results show that using the decision tree classifier one channel semg signal was classified with 91.4% accuracy. in future studies, a combination of other biological signals such as electroencephalography could be used to improve detection and reduce the time of segregation.