Get started
Work

On The Classification of Hand Movements with Electromyogram Signals Obtained From Arm Muscles for Controlling Hand Prosthesis

Year: 2017
Type: article
Abstract: Electromyography (EMG) signals are outcomes of skeletalmuscle activities. In this study EMG signal is read non-invasively from theskin surface by placing electrodes on the skin of specified muscle (surface EMG- SEMG). The aim of the study is to generate control signals from SEMGsmeasured from four hand muscles; Extensor carpi radialis, Palmaris longus,Pronator quadratus and Flexor digitorum superficialis to navigate a prosthetichand. The SEMGs for five hand movements; finger flexion, wrist flexion, wristextension, pronation, supination have been acquired. The features have beencomputed from the windowed EMG of a 0.512 second interval. From each muscle (channel), root mean squarevalue, mean frequency and peak frequency are employed as features. The meanfrequency is computed from the discrete Fourier transform, by counting numberof zero crossings and using minimum norm subspace frequency estimationtechnique. The peak frequency is also obtained by employing the discreteFourier transform. These features and their pairwise combinations have beenclassified with support vector machine. The classifications have been done fortwo scenarios: 1. For each subject the right (left) hand movement is classifiedfrom the right (left) arm EMG data. 2. The left (right) hand movement of a subject is classified from the right(left) arm EMG data of the same subject. The right hand and left hand data recorded from two males andtwo females. The average right-handsuccess of the classification was 82.0%, while the left-hand categorization was83.5%. Interestingly, the left-hand versus right-hand and the right-hand versusleft-hand classification success was obtained 65.7%.
Language: English
Open Access status: green
OpenAlex ID: W2948800049
FWCI: 0.113
Cites: 8
Cited by: 3
Related to: 10