Sensor Reduction on EMG-based Hand Gesture Classification

This work concerns a system based on EMG sensors, signal conditioning circuitry, classification algorithm based on Artificial Neural Network, and virtual avatar representation, useful to identify hand movements within a set of five. This is to potentially make any trans-radial upper-limb amputee able to drive a virtual or real limb prosthetic hand. We focused on differences resulting with the adoption of a different number of sensors and therefore, by means of a very simple heuristic method, we compared different subsets of features, excluding the less significant sensors. We found optimal subsets of one, two, three, four and five sensors, demonstrating a decrease of the performance of only 0.8% when using five sensors, while with three sensors the accuracy can be as high as 81.7%. As shown in Fig. 1. In such a frame, an automatic system, able to analyze the hand gestures and classify their effectiveness, can be strategically adopted. In recent years, different systems was proposed to use surface EMG (sEMG) signal acquired on human forearms as input data to control a real prosthesis [1] or virtual device [2], either for interactive or clinical/rehabilitative [3] purposes. Most of the EMG-controlled device users are radial upper-limb amputees, i.e. amputation occurred below elbow. For these people, the replacement of missing arm functionalities could be a significant improvement to the quality of life. Moreover research showed that the visual-sensorial feedback provided by following the prosthetic or virtual hand movements can be useful to alleviate the phantom limb pain [4], an invalidating condition that affects between 50% and 80% of amputees [5]. In our system, after acquisition, raw EMG data were segmented using the overlapped windowing technique [6]: the windows length was fixed to 256ms, with 64ms of overlap between two successive. For every sensor we considered the following features: • Mean (M): represents the mean value of the EMG amplitude. • Root Mean Square (RMS): represents the mean power of the signal. • Willison Amplitude (WA): represents the number of counts for each change in the EMG signal amplitude that exceeds a predefined threshold, set to avoid background noise-induced counts. It is related to the level of muscle contraction. • Slope Sign Change (SSC): represents the number of times the slope of the EMG signal changes sign. • Simple Square Integral (SSI): represents, similarly to Energy in continuous-time signal, the area under the curve of the squared signal. • Variance (V): represents …