EMG Signal Classification Using Reflection Coefficients and Extreme Value Machine

Azhiri, Reza Bagherian; Esmaeili, Mohammad; Jafarzadeh, Mohsen; Nourani, Mehrdad

doi:10.1109/biocas49922.2021.9644978

Cited by 9 publications

(2 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The most common measures are mean absolute values (MAV), root mean square (RMS), zero-crossing (ZC), waveform length (WL), and Willison amplitude (WA), among others [21,30]. Frequency domain features include measures from the fast Fourier transform or power spectrum of the signal, such as mean frequency, peak frequency, frequency ratio, and total power, among others [21,[31][32][33][34]. On the other hand, time-frequency features include short-time Fourier transform and wavelet transform [32,[35][36][37].…”

Section: Introductionmentioning

confidence: 99%

Progressive Rehabilitation Based on EMG Gesture Classification and an MPC-Driven Exoskeleton

Bonilla,

Bravo,

Bonilla

et al. 2023

Bioengineering

View full text Add to dashboard Cite

Stroke is a leading cause of disability and death worldwide, with a prevalence of 200 millions of cases worldwide. Motor disability is presented in 80% of patients. In this context, physical rehabilitation plays a fundamental role for gradually recovery of mobility. In this work, we designed a robotic hand exoskeleton to support rehabilitation of patients after a stroke episode. The system acquires electromyographic (EMG) signals in the forearm, and automatically estimates the movement intention for five gestures. Subsequently, we developed a predictive adaptive control of the exoskeleton to compensate for three different levels of muscle fatigue during the rehabilitation therapy exercises. The proposed system could be used to assist the rehabilitation therapy of the patients by providing a repetitive, intense, and adaptive assistance.

show abstract

Section: Introductionmentioning

confidence: 99%

Progressive Rehabilitation Based on EMG Gesture Classification and an MPC-Driven Exoskeleton

Bonilla,

Bravo,

Bonilla

et al. 2023

Bioengineering

View full text Add to dashboard Cite

show abstract

“…For classification, the authors utilized support vector machine (SVM) as a classifier. Azhiri et al [2] extracted the same features of [7], but their classifier was extreme value machine (EVM), and improved the accuracy. Esa et al [5] used Hudgins features, root-mean-square (RMS) and finally combination of all these features, and an SVM classifier.…”

Section: Introductionmentioning

confidence: 99%

EMG-Based Feature Extraction and Classification for Prosthetic Hand Control

Azhiri¹,

Esmaeili²,

Nourani³

EPiC Series in Computing

Self Cite

View full text Add to dashboard Cite

In recent years, real-time control of prosthetic hands has gained a great deal of attention. In particular, real-time analysis of Electromyography (EMG) signals has several challenges to achieve an acceptable accuracy and execution delay. In this paper, we address some of these challenges by improving the accuracy in a shorter signal length. We first introduce a set of new feature extraction functions applying on each level of wavelet decomposition. Then, we propose a postprocessing ap- proach to process the neural network outputs. The experimental results illustrate that the proposed method enhances the accuracy of real-time classification of EMG signals up to 95.5 percent for 800 msec signal length. The proposed postprocessing method achieves higher consistency compared with conventional majority voting and Bayesian fusion methods.

show abstract