Virtual Sensor of Surface Electromyography in a New Extensive Fault-Tolerant Classification System

Abstract

A few prosthetic control systems in the scientific literature obtain pattern recognition algorithms adapted to changes that occur in the myoelectric signal over time and, frequently, such systems are not natural and intuitive. These are some of the several challenges for myoelectric prostheses for everyday use. The concept of the virtual sensor, which has as its fundamental objective to estimate unavailable measures based on other available measures, is being used in other fields of research. The virtual sensor technique applied to surface electromyography can help to minimize these problems, typically related to the degradation of the myoelectric signal that usually leads to a decrease in the classification accuracy of the movements characterized by computational intelligent systems. This paper presents a virtual sensor in a new extensive fault-tolerant classification system to maintain the classification accuracy after the occurrence of the following contaminants: ECG interference, electrode displacement, movement artifacts, power line interference, and saturation. The Time-Varying Autoregressive Moving Average (TVARMA) and Time-Varying Kalman filter (TVK) models are compared to define the most robust model for the virtual sensor. Results of movement classification were presented comparing the usual classification techniques with the method of the degraded signal replacement and classifier retraining. The experimental results were evaluated for these five noise types in 16 surface electromyography (sEMG) channel degradation case studies. The proposed system without using classifier retraining techniques recovered of mean classification accuracy was of 4% to 38% for electrode displacement, movement artifacts, and saturation noise. The best mean classification considering all signal contaminants and channel combinations evaluated was the classification using the retraining method, replacing the degraded channel by the virtual sensor TVARMA model. This method recovered the classification accuracy after the degradations, reaching an average of 5.7% below the classification of the clean signal, that is the signal without the contaminants or the original signal. Moreover, the proposed intelligent technique minimizes the impact of the motion classification caused by signal contamination related to degrading events over time. There are improvements in the virtual sensor model and in the algorithm optimization that need further development to provide an increase the clinical application of myoelectric prostheses but already presents robust results to enable research with virtual sensors on biological signs with stochastic behavior.

Keywords: biomedical signal modelling; cross-correlation; fault-tolerant sensor; self-recovery; signal disturbance; virtual sensor.

Figure 1

The experimental procedures (a) Flowchart of the method performed; (b) Description of SVM classification setting.

Figure 2

The positions of the electrodes.

Figure 3

The hand-arm segment movements: (a) rest position; (b) hand movements; (c) Rotational movements; and (d) Wrist movements. The sequence of movements from 1 to 18: rest position; thumb up; flexion of ring and little finger, thumb flexed over middle and little finger; flexion of ring and little finger; thumb opposing base of little finger; abduction of the fingers; fingers flexed together; pointing index; fingers closed together; wrist supination and pronation (rotation axis through the middle finger); wrist supination and pronation (rotation axis through the little finger); wrist flexion and extension; wrist radial and ulnar deviation and wrist extension with closed hand.

Figure 4

The comparison of each contaminant insertion in the clean sEMG signal of the subject 1 sample in four repetitions of movement 7. The clean sEMG signal sample in (a) is artificially contaminated by Motion artefacts in the first column in (b), by Amplifier Saturation in the first column in (c), by Electrode displacements in the first column in (d), by Power line interference in the first column in (e) and by ECG interference in the first column in (f). The second column in (b), (c), (d) and (e) are the sEMG signal samples with the acquisition of real noise.

Figure 5

The operating logic of the SFTD with the virtual sensor.

Figure 6

Classification setting comparison for each contaminant insertion type and the clean sEMG signal classification.

Figure 7

Degradation channel cases comparison for each classification setting.