A Novel Hybrid Model for Drawing Trace Reconstruction from Multichannel Surface Electromyographic Activity

Abstract

Recently, several researchers have considered the problem of reconstruction of handwriting and other meaningful arm and hand movements from surface electromyography (sEMG). Although much progress has been made, several practical limitations may still affect the clinical applicability of sEMG-based techniques. In this paper, a novel three-step hybrid model of coordinate state transition, sEMG feature extraction and gene expression programming (GEP) prediction is proposed for reconstructing drawing traces of 12 basic one-stroke shapes from multichannel surface electromyography. Using a specially designed coordinate data acquisition system, we recorded the coordinate data of drawing traces collected in accordance with the time series while 7-channel EMG signals were recorded. As a widely-used time domain feature, Root Mean Square (RMS) was extracted with the analysis window. The preliminary reconstruction models can be established by GEP. Then, the original drawing traces can be approximated by a constructed prediction model. Applying the three-step hybrid model, we were able to convert seven channels of EMG activity recorded from the arm muscles into smooth reconstructions of drawing traces. The hybrid model can yield a mean accuracy of 74% in within-group design (one set of prediction models for all shapes) and 86% in between-group design (one separate set of prediction models for each shape), averaged for the reconstructed x and y coordinates. It can be concluded that it is feasible for the proposed three-step hybrid model to improve the reconstruction ability of drawing traces from sEMG.

sEMG-Based Drawing Trace Reconstruction: A Novel Hybrid Algorithm Fusing Gene Expression Programming into Kalman Filter

“…The remaining seven channels were set to record sEMG signals over four forearm muscles: the Extensor Carpi Radialis Brevis, Extensor Carpi Ulnaris, Extensor Digitorum, Flexor Carpi Radialis and three upper arm muscles: Biceps Brachii, Deltoid Muscle and Triceps Brachii. For a detailed illustration of the locations of the electrodes, see [ 24 ].…”

“…The sEMG signals acquisition scheme, experimental instruments, sensors and electrodes are the same as those used in Reference [ 24 ]. The raw sEMG signals were sampled at 1000 Hz and band-pass filtered at 10–500 Hz with a notch filter implemented to remove the 50 Hz line interference.…”

“…During the training phase, the coordinates were collected during each trial every 50 ms and converted to the differences between the present state and the previous one. Simultaneously, based on the adjacent windowing techniques [ 26 , 27 ], sEMG signals were segmented for feature extraction every 50 ms. Then, we applied Root Mean Square (RMS) transformation to sEMG data to extract features every 50 ms. For a detailed illustration of data pre-processing, see [ 24 ].…”

“…Based on our previous study [ 24 ], we proposed a novel hybrid algorithm (KF-GEP). It deeply merges the Kalman Filter with the GEP for the sEMG-based drawing and handwriting trace reconstruction.…”

“…In our previous study [ 24 ], we proposed a three-step hybrid model based on Gene Expression Programming (GEP) for the reconstruction of hand-drawn shapes from multichannel sEMG signals. Although the results showing that this GEP model slightly outperforms the KF, we can still consider the temporal and spatial context properties of drawing through KF to improve the reconstruction performance of GEP model.…”