This paper gives an overview of the different research works related to electromyographic signals (EMG) classification based on Support Vector Machines (SVM). The article summarizes the techniques used to make the classification in each reference. Furthermore, it includes the obtained accuracy, the number of signals or channels used, the way the authors made the feature vector, and the type of kernels used. Hence, this article also includes a compilation about the bands used to filter signals, the number of signals recommended, the most commonly used sampling frequencies, and certain features that can create the characteristics of the vector. This research gathers articles related to different kinds of SVM-based classification and other tools for signal processing in the field.

Zero crossings are a practical and efficient feature to approximate the frequency of a sampled series of data. Some research describes in different ways how to compute the zero crossings feature starting from its definition, and in some of them, a threshold is included as part of it. This research compiles a comprehensive list of description methods for zero crossings, both with or without threshold. In addition, an improvement of one method is proposed, mainly to save time resources. Moreover, it increases the precision when the objective is to perform some classification. This feature is often used as a vector of a matrix of features in signal classification. To test the different variations of the zero crossings methods, a classification of electromyographic signals was performed using support vector machines. The results obtained by the proposed method threw near to a 40% improvement in the classification compared to those approaches that do not consider a threshold and more than 7% compared to those with a threshold. The processing time of this work is shortened compared to others that also take into account a threshold.

The number and position of sEMG electrodes have been studied extensively due to the need to improve the accuracy of the classification they carry out of the intention of movement. Nevertheless, increasing the number of channels used for this classification often increases their processing time as well. This research work contributes with a comparison of the classification accuracy based on the different number of sEMG signal channels (one to four) placed in the right lower limb of healthy subjects. The analysis is performed using Mean Absolute Values, Zero Crossings, Waveform Length, and Slope Sign Changes; these characteristics comprise the feature vector. The algorithm used for the classification is the Support Vector Machine after applying a Principal Component Analysis to the features. The results show that it is possible to reach more than 90% of classification accuracy by using 4 or 3 channels. Moreover, the difference obtained with 500 and 1000 samples, with 2, 3 and 4 channels, is not higher than 5%, which means that increasing the number of channels does not guarantee 100% precision in the classification.

Electromyography (EMG) processing is a fundamental part of medical research. It offers the possibility of developing new devices and techniques for the diagnosis, treatment, care, and rehabilitation of patients, in most cases non-invasively. However, EMG signals are random, non-stationary, and non-linear, making their classification difficult. Due to this, it is of vital importance to define which factors are helpful for the classification process. In order to improve this process, it is possible to apply algorithms capable of identifying which features are most important in the categorization process. Algorithms based on metaheuristic methods have demonstrated an ability to search for suitable subsets of features for optimization problems. Therefore, this work proposes a methodology based on genetic algorithms for feature selection to find the parameter space that offers the slightest classification error in 250 ms signal segments. For classification, a support vector machine is used. For this work, two databases were used, the first corresponding to the right upper extremity and the second formed by movements of the right lower extremity. For both databases, a feature space reduction of over 65% was obtained, with a higher average classification efficiency of 91% for the best subset of parameters. In addition, particle swarm optimization (PSO) was applied based on right upper extremity data, obtaining an 88% average error and a 46% reduction for the best subset of parameters. Finally, a sensitivity analysis was applied to the characteristics selected by PSO and genetic algorithms for the database of the right upper extremity, obtaining that the parameters determined by the genetic algorithms show greater sensitivity for the classification process.

This article provides an overview of the implementation of electromyography (EMG) signal classification algorithms in various embedded system architectures. They address the specifications used for implementation in different devices, such as the number of movements and the type of classification method. Architectures analyzed include microcontrollers, DSP, FPGA, SoC, and neuromorphic computers/chips in terms of precision, processing time, energy consumption, and cost. This analysis highlights the capabilities of each technology for real-time wearable applications such as smart prosthetics and gesture control devices, as well as the importance of local inference in artificial intelligence models to minimize execution times and resource consumption. The results show that the choice of device depends on the required system specifications, the robustness of the model, the number of movements to be classified, and the limits of knowledge concerning design and budget. This work provides a reference for selecting technologies for developing embedded biomedical solutions based on EMG.

The best way to represent EMG signals for classification is a topic that has been widely studied due to the need to improve precision when identifying the type of movement being performed. However, by increasing the number of features when forming a matrix that represents the signals, the processing time increases since it not only involves calculating the features that are extracted from the signal but also the time that the classifier takes to answer. The central purpose of this research is to develop and validate a methodology that uses the Fisher Score to select a set of features in the classification of sEMG signals. This selected set is descriptive enough to achieve high levels of accuracy in detecting EMG signal patterns across multiple subjects. The analysis shows that using a variant of MAV, SSC, WAMP, RMS, and the maximum value together with the Shannon entropy and zero crossings of the Wavelet transform has an accuracy greater than 99%. Finally, a group of features is proposed to classify EMG signals that yield an accuracy greater than 98% and do not require more than 15 ms of processing time.