Hand motion detection from EMG signals by using ANN based classifier for human computer interaction

Hand gesture acquisition refers to the process whereby a computer is able to detect and record human hand movements. Vision-based hand gesture detection and a data glove-based approach can do this. In the realm of vision-based methods, we find both static hand and real-time hand gesture detection approaches. Five flex sensors are sewn onto the glove for the data glove method. Compared to the data glove-based method, the vision-based method is more consistent and robust. Designing a user-friendly vision-based interface, however, is a formidable challenge. The collected hand motions will need to be manually categorised by the user once again. So, it's no surprise that deciphering data is a laborious process that eats up plenty of time. This analysis should ideally be done mechanically. Relating the electrical nature of the human nervous system to computers presents a significant challenge in the field of Human Computer Interaction (HCI). Electromyographic (EMG) impulses, which change depending on the sort of hand motion utilised, are used to establish a brain link with computers. Because of this, an EMG-based system for recognising hand gestures can be created. The goal of this research is to create signal processing methods and Artificial Neural Networks (ANN) that can automatically recognise hand motions. The recorded Electromyographic (EMG) signals for a wide variety of hand motions are analysed using signal processing techniques, from which a number of features are extracted. Characteristics are extracted using Wavelet Transform (WT) based techniques including Discrete Wavelet Transform (DWT) and Wavelet Packet Transform (WPT). Better performance was achieved with the recommended methods, as shown by the simulation results.

Electromyography (EMG) signals can be used for clinical/biomedical applications, Evolvable Hardware Chip (EHW) development and modern Human Computer Interaction. EMG signals acquired from muscles require advanced methods for detection, decomposition, processing, and classification. For properly analyze EMG signal there need good quality of decomposition so that it can reveal the total characteristics of EMG signals. Because EMG signal is Non-Stationary signal so it needs such a method that can decompose non-stationary signal thus wavelet decomposition is a good choice for this type. There are different types of wavelet available. Henceforth, it is necessary that proper attempt should be taken to choice the best one. Here analyses of EMG Signals were made by Various Wavelet Decomposition method with different types of wavelet and it illustrates the comparative study on best possible energy localization in the time-scale plane in order to show the performance. Thus we can choice the right one. The EMG signals used for this analysis - were found both from locally collected as well as from www.emglab.net[5] which provides EMG signal related raw data and other facilities.

The goal of the research presented in this chapter is to improve the classification process of electromyography (EMG) signals that are contaminated with noise. If the existence of noise in EMG signals is not accounted for, it can degrade the performance of the classification task. Therefore, it is necessary to utilize an efficient filtering process to improve the classification of EMG signals. Guided by the need to filter the noise out of EMG signals, this chapter proposes to employ a Gaussian smoothing filter (GSF) that is simple in its implementation with an efficient filtering performance. The GSF, which is a Gaussian function, offers equal support in both frequency and time domains, allowing it to yield a performance compromise in removing the noise while preserving high frequency components of EMG signals. It is additionally shown that the use of GSF not only enhances the classification accuracy, but it also reduces the computational time needed in the training and testing of the classification process. To evaluate the performance of the GSF in EMG signals classification problem, two experiments are considered. The first experiment consists of classification of multiple hand gestures using EMG signals and the second experiment considers classifying phases of hand motion for a grasping task. An array of standard classification techniques are considered in both experiments and the use of GSF in filtering out noise is shown to enhance classification accuracy with remarkably reduced computational time for the considered classification techniques. This illustrates the feasibility of GSF in filtering EMG signals for classification tasks. To gain further insights into the GSF, its performance is compared with that of a median filter (MF), one of the well-known filtering techniques. By using the overall classification accuracy as an index of comparison, the GSF is shown to result in a superior classification accuracy, demonstrating its efficacy for EMG signals filtering process. Thus, employing GSF proves to provide enhancement in the classification accuracy and required computational efforts.

Today, human-computer interfaces are increasingly more often used and become necessary for human daily activities. Among some remarkable applications, we find: Wireless-computer controlling through hand movement, wheelchair directing/guiding with finger motions, and rehabilitation. Such applications are possible from the analysis of electromyographic (EMG) signals. Despite some research works have addressed this issue, the movement classification through EMG signals is still an open challenging issue to the scientific community -especially, because the controller performance depends not only on classifier but other aspects, namely: used features, movements to be classified, the considered feature-selection methods, and collected data. In this work, we propose an exploratory work on the characterization and classification techniques to identifying movements through EMG signals. We compare the performance of three classifiers (KNN, Parzen-density-based classifier and ANN) using spectral (Wavelets) and time-domain-based (statistical and morphological descriptors) features. Also, a methodology for movement selection is proposed. Results are comparable with those reported in literature, reaching classification errors of 5.18% (KNN), 14.7407% (ANN) and 5.17% (Parzen-density-based classifier).

Electromyography (EMG) signals hold critical significance in biomedical research, capturing muscle electrical activity during rest and contraction in the upper limb. Their versatility in applications, particularly in human-assisting robotic tools, drives ongoing exploration and research. This paper presents an original study focusing on leveraging machine learning techniques to classify EMG datasets and efficiently control a robotic arm based on predicted gestures. Data acquisition involves strategically placing an EMG muscle sensor on the forearm to ensure precise measurement of signals associated with hand gestures and movements. Diverse classifiers including random forest, support vector machine (SVM), K-nearest neighbors (KNN), Gaussian naïve Bayes, gated recurrent unit (GRU), long short-term memory (LSTM), artificial neural network (ANN), recurrent neural network (RNN), convolutional neural network (CNN), and vision transformer (ViT) are employed. Performance results are meticulously analyzed and presented in tabular format, showcasing the ViT classifier as the most successful, achieving an impressive 97.7% accuracy in robotic arm control. Notably, ANN, RNN, and CNN also exhibit high accuracy exceeding 90%. Furthermore, this work is comprehensively compared with existing literature, laying the groundwork for future advancements in human-robot interaction and cutting-edge assistive technologies that markedly enhance the quality of life for individuals with motor impairments or disabilities. The findings carry significant implications for designing and implementing intuitive, responsive robotic systems based on EMG signals.

People with disabilities correspond to about 25% of the Brazilian population. A great part of these people have physical impairments that difficult the use computer peripherals. This article presents the development of a system for detection of hand movements through the acquisition and classification of electromyographic (EMG) signals using machine learning techniques. The purpose of the proposed system is to be used by people with disabilities to control an adapted text editor. The signals are capture by surface EMG electrodes and used to the detect 4 different hand movements. In addition, a database with 3200 EMG signals generated by the hand movements was created, made by one user diagnosed with cerebral palsy and another user without diagnosed motor disabilities. Several tests were carried out, showing the good accuracy of the proposed system, with a success classification rate of 96% to 98%.

<p>The shapes and firing rates of motor unit action potentials (MUAPs) in an electromyographic (EMG) signal provide an important source of information for the diagnosis of neuromuscular disorders. In order to extract this information from EMG signals recorded at force levels up to 20% of maximum voluntary contraction (MVC) it is required: (i) To identify the MUAPs composing the EMG signal, (ii) To classify MUAPs with similar shape and (iii) To decompose the superimposed MUAP waveforms into their constituent MUAPs. For the classification of MUAPs two different pattern recognition techniques are presented (i) An artificial neural network (ANN) technique based on unsupervised learning using the self-organizing feature maps (SOFM) algorithm and learning vector quantization (LVQ) and (ii) A statistical pattern recognition technique based on the euclidian distance. The success rate on real data for the ANN technique is about 96% and for the statistical one about 94%. For the decomposition of the superimposed waveforms the following technique is used: (i) Cross-correlation of each of the unique MUAP waveforms, obtained by the classification process with the superimposed waveforms in order to find the best matching point and (ii) A combination of euclidian distance and area measures in order to classify the components of the decomposed waveform. The success rate for the decomposition procedure is about 90%.</p>

Over the past several years, electromyography (EMG) signals have been used as a natural interface to interact with computers and machines. Recently, deep learning algorithms such as Convolutional Neural Networks (CNNs) have gained interest for decoding the hand movement intention from EMG signals. However, deep networks require a large dataset to train appropriately. Creating such a database for a single subject could be very time-consuming. In this study, we addressed this issue from two perspectives: (i) we proposed a subject-transfer framework to use the knowledge learned from other subjects to compensate for a target subject's limited data; (ii) we proposed a task-transfer framework in which the knowledge learned from a set of basic hand movements is used to classify more complex movements, which include a combination of mentioned basic movements. We introduced two CNN-based architectures for hand movement intention detection and a subject-transfer learning approach. Classifiers are tested on the Nearlab dataset, a sEMG hand/wrist movement dataset including 8 movements and 11 subjects, along with their combination, and on open-source hand sEMG dataset "NinaPro DataBase 2 (DB2)." For the Nearlab database, the subject-transfer learning approach improved the average classification accuracy of the proposed deep classifier from 92.60 to 93.30% when classifier was utilizing 10 other subjects' data via our proposed framework. For Ninapro DB2 exercise B (17 hand movement classes), this improvement was from 81.43 to 82.87%. Moreover, three stages of analysis in task-transfer approach proved that it is possible to classify combination hand movements using the knowledge learned from a set of basic hand movements with zero, few samples and few seconds of data from the target movement classes. First stage takes advantage of shared muscle synergies to classify combined movements, while second and third stages take advantage of novel algorithms using few-shot learning and fine-tuning to use samples from target domain to further train the classifier trained on the source database. The use of information learned from basic hand movements improved classification accuracy of combined hand movements by 10%.

Electromyography (EMG) signals have been successfully employed for driving prosthetic limbs of a single or double degree of freedom. This principle works by using the amplitude of the EMG signals to decide between one or two simpler movements. This method underperforms as compared to the contemporary advances done at the mechanical, electronics, and robotics end, and it lacks intuition to perform everyday life tasks. Recently, research on myoelectric control based on pattern recognition (PR) shows promising results with the aid of machine learning classifiers. Using the approach termed as EMG-PR, EMG signals are divided into analysis windows, and features are extracted for each window. These features are then fed to the machine learning classifiers as input. By offering multiple class movements and intuitive control, this method has the potential to power Prosthesis an amputated subject to perform everyday life movements. In this paper, we investigate the effect of the analysis window and feature selection on classification accuracy of different hand and wrist movements using time-domain features. We show that effective data preprocessing and optimum feature selection helps to improve the classification accuracy of hand movements. We use publicly available hand and wrist gesture dataset of 40 intact and 11 trans-radial amputated subjects for experimentation. Results computed using different classification algorithms show that the proposed preprocessing and features selection outperforms the baseline and achieve up to \(98\%\) classification accuracy for both intact and trans-radial subjects.

Walking gait event detection based on electromyography signals using artificial neural network

Control of transhumeral prosthetic devices can effectively be performed using the predicted joint torques at the elbow. The joint torque values are generally predicted using the Electromyography (EMG) signals taken from upper arm muscles of the amputee. This paper uses a Bagnoli-16 EMG system to extract EMG signals from the biceps and triceps. The EMG signals are complex to handle mainly due to the stochastic nature of the signal. Independent component analysis (ICA) is utilized to isolate the EMG signals from each muscle. In order to measure the actual torque, a novel kinematic model is proposed in this paper. For the joint torque prediction two classifiers have been developed. First an Artificial Neural Network model (ANN) based classifier is trained to predict the joint torques. Using different test data the ANN model is tested against the arm kinematic based joint torque predictions. The test results indicated 5.6% of root mean square error against the actual predicted torque values. In order to improve the classification an Artificial Neuro-Fuzzy inference system (ANFIS) has been developed. Using the same data the ANFIS based classifier produced 3.3% of the root mean square error against the kinematically predicted joint torques.

An automated system for resolving an intramuscular electromyographic (EMG) signal into its constituent motor unit potential trains (MUPTs) is presented. The system is intended mainly for clinical applications where several physiological parameters for each motor unit (MU), such as the motor unit potential (MUP) template and mean firing rate, are required. The system decomposes an EMG signal off-line by filtering the signal, detecting MUPs, and then grouping the detected MUPs using a clustering and a supervised classification algorithm. Both the clustering and supervised classification algorithms use MUP shape and MU firing pattern information to group MUPs into several MUPTs. Clustering is partially based on the K-means clustering algorithm. Supervised classification is implemented using a certainty-based classifier technique that employs a knowledge-based system to merge trains, detect and correct invalid trains, as well as adjust the assignment threshold for each train. The accuracy (93.2%±5.5%), assignment rate (93.9%±2.6%), and error in estimating the number of MUPTs (0.3±0.5) achieved for 10 simulated EMG signals comprised of 3-11 MUPTs are encouraging for using the system for decomposing various EMG signals.

Rapid disability patients increasing over time and need a solution in the future. Hand amputation is one form of disability that common in Indonesian society. A possible solution would be necessary at the moment is the development of prosthetic hand that has the ability as a human hand. The development of neuroscience has now reached the stage of the body's ability to use the signal as an input signal to operate a system. One of the applications of the science development is the use of electromyography (EMG) signals as an input to the control system to operate the prosthetic hand. This study is divided into two stages: a preliminary study and further research. Initial research focus in the process of EMG signal pattern recognition and advanced research focus in the development of a prototype prosthetic hand that is integrated with the controller system. Preliminary research indicates that the results of pattern recognition EMG signal using wavelet transform and Artificial Neural Network (ANN) classification has an accuracy rate of about 77.5 %. Based on these results, it can be concluded that the study results could be used as a signal input to program control of the prosthetic hand that will be developed in phase two.

Analysis of electromyography (EMG) signals of normal physical actions have found to be important in order to detect certain abnormalities of the musculoskeletal system and diagnose abnormalities in patient behavior. This paper presents the results of the development of an Artificial Neural Network (ANN) for classification of EMG signals, according to the type of human behavior. The developed ANN is able to distinguish between 10 normal behaviors: bowing, clapping, handshaking, hugging, jumping, running, sitting, standing, walking, and waving. Feedforward neural network architecture was developed using dataset from UCI Machine Learning Repository database. QPC of each episode in EMG signal were obtained using bispectrum signal analysis. Training of ANN was performed using k-fold cross validation and impact of different number of neurons in hidden layer on system output was evaluated. Finally, the single-layer, feedforward neural network architecture with 17 neurons in hidden layer achieved the best performance and had sensitivity of 86.67% and of specificity 85.00%. The overall accuracy of developed structure is 86.25%.

Object acquisition and selection are two important functions performed in most of the human computer interaction (HCI) systems. Various techniques are devised by the researchers to perform these operations and the selection of a combination of object acquisition and selection techniques for a particular HCI system has become a research issue, especially, when the user of these systems are differently abled persons. This paper presents a review on object acquisition and selection techniques used in HCI systems. It starts with the introduction to HCI systems, gives an overview on object acquisition & selection techniques, feedback modes, performing mouse analogous actions using eye blinks, the applications of HCI systems, and finally discusses challenges and issues related to these techniques.