Feature Extraction and Reduction of Wavelet Transform Coefficients for EMG Pattern Classification

Improved Surface Electromyogram-Based Hand-Wrist Force Estimation Using Deep Neural Networks and Cross-Joint Transfer Learning

Deep neural networks (DNNs) and transfer learning (TL) have been used to improve surface electromyogram (sEMG)-based force estimation. However, prior studies focused mostly on applying TL within one joint, which limits dataset size and diversity. Herein, we investigated cross-joint TL between two upper-limb joints with four DNN architectures using sliding windows. We used two feedforward and two recurrent DNN models with feature engineering and feature learning, respectively. We found that the dependencies between sEMG and force are short-term (<400 ms) and that sliding windows are sufficient to capture them, suggesting that more complicated recurrent structures may not be necessary. Also, using DNN architectures reduced the required sliding window length. A model pre-trained on elbow data was fine-tuned on hand-wrist data, improving force estimation accuracy and reducing the required training data amount. A convolutional neural network with a 391 ms sliding window fine-tuned using 20 s of training data had an error of 6.03 ± 0.49% maximum voluntary torque, which is statistically lower than both our multilayer perceptron model with TL and a linear regression model using 40 s of training data. The success of TL between two distinct joints could help enrich the data available for future deep learning-related studies.

Evaluation of Hand-Crafted Feature Extraction for Fault Diagnosis in Rotating Machinery: A Survey

This article presents a comprehensive collection of formulas and calculations for hand-crafted feature extraction of condition monitoring signals. The documented features include 123 for the time domain and 46 for the frequency domain. Furthermore, a machine learning-based methodology is presented to evaluate the performance of features in fault classification tasks using seven data sets of different rotating machines. The evaluation methodology involves using seven ranking methods to select the best ten hand-crafted features per method for each database, to be subsequently evaluated by three types of classifiers. This process is applied exhaustively by evaluation groups, combining our databases with an external benchmark. A summary table of the performance results of the classifiers is also presented, including the percentage of classification and the number of features required to achieve that value. Through graphic resources, it has been possible to show the prevalence of certain features over others, how they are associated with the database, and the order of importance assigned by the ranking methods. In the same way, finding which features have the highest appearance percentages for each database in all experiments has been possible. The results suggest that hand-crafted feature extraction is an effective technique with low computational cost and high interpretability for fault identification and diagnosis.

Post-stroke hand gesture recognition via one-shot transfer learning using prototypical networks

BACKGROUND

In-home rehabilitation systems are a promising, potential alternative to conventional therapy for stroke survivors. Unfortunately, physiological differences between participants and sensor displacement in wearable sensors pose a significant challenge to classifier performance, particularly for people with stroke who may encounter difficulties repeatedly performing trials. This makes it challenging to create reliable in-home rehabilitation systems that can accurately classify gestures.

METHODS

Twenty individuals who suffered a stroke performed seven different gestures (mass flexion, mass extension, wrist volar flexion, wrist dorsiflexion, forearm pronation, forearm supination, and rest) related to activities of daily living. They performed these gestures while wearing EMG sensors on the forearm, as well as FMG sensors and an IMU on the wrist. We developed a model based on prototypical networks for one-shot transfer learning, K-Best feature selection, and increased window size to improve model accuracy. Our model was evaluated against conventional transfer learning with neural networks, as well as subject-dependent and subject-independent classifiers: neural networks, LGBM, LDA, and SVM.

RESULTS

Our proposed model achieved 82.2% hand-gesture classification accuracy, which was better (P<0.05) than one-shot transfer learning with neural networks (63.17%), neural networks (59.72%), LGBM (65.09%), LDA (63.35%), and SVM (54.5%). In addition, our model performed similarly to subject-dependent classifiers, slightly lower than SVM (83.84%) but higher than neural networks (81.62%), LGBM (80.79%), and LDA (74.89%). Using K-Best features improved the accuracy in 3 of the 6 classifiers used for evaluation, while not affecting the accuracy in the other classifiers. Increasing the window size improved the accuracy of all the classifiers by an average of 4.28%.

CONCLUSION

Our proposed model showed significant improvements in hand-gesture recognition accuracy in individuals who have had a stroke as compared with conventional transfer learning, neural networks and traditional machine learning approaches. In addition, K-Best feature selection and increased window size can further improve the accuracy. This approach could help to alleviate the impact of physiological differences and create a subject-independent model for stroke survivors that improves the classification accuracy of wearable sensors.

TRIAL REGISTRATION NUMBER

The study was registered in Chinese Clinical Trial Registry with registration number CHiCTR1800017568 in 2018/08/04.

Classification of EMG signals with CNN features and voting ensemble classifier

Electromyography (EMG) signals are primarily used to control prosthetic hands. Classifying hand gestures efficiently with EMG signals presents numerous challenges. In addition to overcoming these challenges, a successful combination of feature extraction and classification approaches will improve classification accuracy. In the current work, convolutional neural network (CNN) features are used to reduce the redundancy problems associated with time and frequency domain features to improve classification accuracy. The features from the EMG signal are extracted using a CNN and are fed to the 'k' nearest neighbor (KNN) classifier with a different number of neighbors ( 1 N N , 3 N N , 5 N N , and   7 N N ) . It results in an ensemble of classifiers that are combined using a hard voting-based classifier. Based on the benchmark Ninapro DB4 database and CapgMyo database, the proposed framework obtained 91.3 % classification accuracy on CapgMyo and 89.5 % on Ninapro DB4.

Supervised Myoelectrical Hand Gesture Recognition in Post-Acute Stroke Patients with Upper Limb Paresis on Affected and Non-Affected Sides

In clinical practice, acute post-stroke paresis of the extremities fundamentally complicates timely rehabilitation of motor functions; however, recently, residual and distorted musculoskeletal signals have been used to initiate feedback-driven solutions for establishing motor rehabilitation. Here, we investigate the possibilities of basic hand gesture recognition in acute stroke patients with hand paresis using a novel, acute stroke, four-component multidomain feature set (ASF-4) with feature vector weight additions (ASF-14NP, ASF-24P) and supervised learning algorithms trained only by surface electromyography (sEMG). A total of 19 (65.9 ± 12.4 years old; 12 men, seven women) acute stroke survivors (12.4 ± 6.3 days since onset) with hand paresis (Brunnstrom stage 4 ± 1/4 ± 1, SIAS 3 ± 1/3 ± 2, FMA-UE 40 ± 20) performed 10 repetitive hand movements reflecting basic activities of daily living (ADLs): rest, fist, pinch, wrist flexion, wrist extension, finger spread, and thumb up. Signals were recorded using an eight-channel, portable sEMG device with electrode placement on the forearms and thenar areas of both limbs (four sensors on each extremity). Using data preprocessing, semi-automatic segmentation, and a set of extracted feature vectors, support vector machine (SVM), linear discriminant analysis (LDA), and k-nearest neighbors (k-NN) classifiers for statistical comparison and validity (paired t-tests, p-value < 0.05), we were able to discriminate myoelectrical patterns for each gesture on both paretic and non-paretic sides. Despite any post-stroke conditions, the evaluated total accuracy rate by the 10-fold cross-validation using SVM among four-, five-, six-, and seven-gesture models were 96.62%, 94.20%, 94.45%, and 95.57% for non-paretic and 90.37%, 88.48%, 88.60%, and 89.75% for paretic limbs, respectively. LDA had competitive results using PCA whereas k-NN was a less efficient classifier in gesture prediction. Thus, we demonstrate partial efficacy of the combination of sEMG and supervised learning for upper-limb rehabilitation procedures for early acute stroke motor recovery and various treatment applications.

Real-Time Classification of Pain Level Using Zygomaticus and Corrugator EMG Features

The real-time recognition of pain level is required to perform an accurate pain assessment of patients in the intensive care unit, infants, and other subjects who may not be able to communicate verbally or even express the sensation of pain. Facial expression is a key pain-related behavior that may unlock the answer to an objective pain measurement tool. In this work, a machine learning-based pain level classification system using data collected from facial electromyograms (EMG) is presented. The dataset was acquired from part of the BioVid Heat Pain database to evaluate facial expression from an EMG corrugator and EMG zygomaticus and an EMG signal processing and data analysis flow is adapted for continuous pain estimation. The extracted pain-associated facial electromyography (fEMG) features classification is performed by K-nearest neighbor (KNN) by choosing the value of k which depends on the nonlinear models. The presentation of the accuracy estimation is performed, and considerable growth in classification accuracy is noticed when the subject matter from the features is omitted from the analysis. The ML algorithm for the classification of the amount of pain experienced by patients could deliver valuable evidence for health care providers and aid treatment assessment. The proposed classification algorithm has achieved a 99.4% accuracy for classifying the pain tolerance level from the baseline (P0 versus P4) without the influence of a subject bias. Moreover, the result on the classification accuracy clearly shows the relevance of the proposed approach.

Research Hotspots of the Rehabilitation Medicine Use of sEMG in Recent 12 Years: A Bibliometric Analysis

Objective

Surface electromyography (sEMG) has been widely applied to rehabilitation medicine. However, the bibliometric analysis of the rehabilitation medicine use of sEMG is vastly unknown. Therefore, this research aimed to investigate the current trends of the rehabilitation medicine use of sEMG in the recent 12 years by using CiteSpace (5.8).

Methods

Literature relating to rehabilitation medicine use of sEMG from 2010 to 2021 was retrieved from the Web of Science. CiteSpace analyzed country, institution, cited journals, authors, cited references and keywords. An analysis of counts and centrality was used to reveal publication outputs, countries, institutions, core journals, active authors, foundation references, hot topics and frontiers.

Results

A total of 1949 publications were retrieved from 2010 to 2021. The total number of publications continually increased over the past 12 years, and the most active countries, institutions, journals and authors in rehabilitation medicine use of sEMG were identified. The most productive country and institution in this field were America (484) and the University of Sao Paulo (36). Andersen LL (28) was the most prolific author, and Dario Farina ranked first among the cited authors. Besides, there were three main frontiers in keywords for sEMG research, including “activation”, “exercise”, and “strength”.

Conclusion

The findings from this bibliometric study provide the current status and trends in clinical research of rehabilitation medicine use of sEMG over the past ten years, which may help researchers identify hot topics and explore new directions for future research in this field.

Myoelectric Pattern Recognition Performance Enhancement Using Nonlinear Features

The multichannel electrode array used for electromyogram (EMG) pattern recognition provides good performance, but it has a high cost, is computationally expensive, and is inconvenient to wear. Therefore, researchers try to use as few channels as possible while maintaining improved pattern recognition performance. However, minimizing the number of channels affects the performance due to the least separable margin among the movements possessing weak signal strengths. To meet these challenges, two time-domain features based on nonlinear scaling, the log of the mean absolute value (LMAV) and the nonlinear scaled value (NSV), are proposed. In this study, we validate the proposed features on two datasets, the existing four feature extraction methods, variable window size, and various signal-to-noise ratios (SNR). In addition, we also propose a feature extraction method where the LMAV and NSV are grouped with the existing 11 time-domain features. The proposed feature extraction method enhances accuracy, sensitivity, specificity, precision, and F1 score by 1.00%, 5.01%, 0.55%, 4.71%, and 5.06% for dataset 1, and 1.18%, 5.90%, 0.66%, 5.63%, and 6.04% for dataset 2, respectively. Therefore, the experimental results strongly suggest the proposed feature extraction method, for taking a step forward with regard to improved myoelectric pattern recognition performance.

A Novel Methodology for the Synchronous Collection and Multimodal Visualization of Continuous Neurocardiovascular and Neuromuscular Physiological Data in Adults with Long COVID

Background: Reports suggest that adults with post-COVID-19 syndrome or long COVID may be affected by orthostatic intolerance syndromes, with autonomic nervous system dysfunction as a possible causal factor of neurocardiovascular instability (NCVI). Long COVID can also manifest as prolonged fatigue, which may be linked to neuromuscular function impairment (NMFI). The current clinical assessment for NCVI monitors neurocardiovascular performance upon the application of orthostatic stressors such as an active (i.e., self-induced) stand or a passive (tilt table) standing test. Lower limb muscle contractions may be important in orthostatic recovery via the skeletal muscle pump. In this study, adults with long COVID were assessed with a protocol that, in addition to the standard NCVI tests, incorporated simultaneous lower limb muscle monitoring for NMFI assessment. Methods: To conduct such an investigation, a wide range of continuous non-invasive biomedical sensing technologies were employed, including digital artery photoplethysmography for the extraction of cardiovascular signals, near-infrared spectroscopy for the extraction of regional tissue oxygenation in brain and muscle, and electromyography for assessment of timed muscle contractions in the lower limbs. Results: With the proposed methodology described and exemplified in this paper, we were able to collect relevant physiological data for the assessment of neurocardiovascular and neuromuscular functioning. We were also able to integrate signals from a variety of instruments in a synchronized fashion and visualize the interactions between different physiological signals during the combined NCVI/NMFI assessment. Multiple counts of evidence were collected, which can capture the dynamics between skeletal muscle contractions and neurocardiovascular responses. Conclusions: The proposed methodology can offer an overview of the functioning of the neurocardiovascular and neuromuscular systems in a combined NCVI/NMFI setup and is capable of conducting comparative studies with signals from multiple participants at any given time in the assessment. This could help clinicians and researchers generate and test hypotheses based on the multimodal inspection of raw data in long COVID and other cohorts.

Deep Learning-Based Approaches for Decoding Motor Intent From Peripheral Nerve Signals

Previous literature shows that deep learning is an effective tool to decode the motor intent from neural signals obtained from different parts of the nervous system. However, deep neural networks are often computationally complex and not feasible to work in real-time. Here we investigate different approaches' advantages and disadvantages to enhance the deep learning-based motor decoding paradigm's efficiency and inform its future implementation in real-time. Our data are recorded from the amputee's residual peripheral nerves. While the primary analysis is offline, the nerve data is cut using a sliding window to create a “pseudo-online” dataset that resembles the conditions in a real-time paradigm. First, a comprehensive collection of feature extraction techniques is applied to reduce the input data dimensionality, which later helps substantially lower the motor decoder's complexity, making it feasible for translation to a real-time paradigm. Next, we investigate two different strategies for deploying deep learning models: a one-step (1S) approach when big input data are available and a two-step (2S) when input data are limited. This research predicts five individual finger movements and four combinations of the fingers. The 1S approach using a recurrent neural network (RNN) to concurrently predict all fingers' trajectories generally gives better prediction results than all the machine learning algorithms that do the same task. This result reaffirms that deep learning is more advantageous than classic machine learning methods for handling a large dataset. However, when training on a smaller input data set in the 2S approach, which includes a classification stage to identify active fingers before predicting their trajectories, machine learning techniques offer a simpler implementation while ensuring comparably good decoding outcomes to the deep learning ones. In the classification step, either machine learning or deep learning models achieve the accuracy and F1 score of 0.99. Thanks to the classification step, in the regression step, both types of models result in a comparable mean squared error (MSE) and variance accounted for (VAF) scores as those of the 1S approach. Our study outlines the trade-offs to inform the future implementation of real-time, low-latency, and high accuracy deep learning-based motor decoder for clinical applications.

Detection of k-complexes in EEG signals using a multi-domain feature extraction coupled with a least square support vector machine classifier

Sleep scoring is one of the primary tasks for the classification of sleep stages using electroencephalogram (EEG) signals. It is one of the most important diagnostic methods in sleep research and must be carried out with a high degree of accuracy because any errors in the scoring in the patient's sleep EEG recordings can cause serious problems. The aim of this research is to develop a new automatic method for detecting the most important characteristics in sleep stage 2 such as k-complexes based on multi-domain features. In this study, each EEG signal is divided into a set of segments using a sliding window technique. Based on extensive experiments during the training phase, the size of the sliding window is set to 0.5 s (s). Then a set of statistical, fractal, frequency and non-linear features are extracted from each epoch based on the time domain, Katz's algorithm, power spectrum density (PSD) and tunable Q-factor wavelet transform (TQWT). As a result, a vector of twenty-two features is obtained to represent each EEG segment. In order to detect k-complexes, the extracted features were analysed for their ability to detect the k-complex waveforms. Based on the analysis of the features, twelve out of twenty-two features are selected and forwarded to a least square support vector machine (LS-SVM) classifier to identify k-complexes in EEG signals. A set of various classification techniques of K-means and extreme learning machine classifiers are used to compare the obtained results and to evaluate the performance of the proposed method.The experimental results showed that the proposed method, based on multi-domain features, achieved better recognition results than other methods and classifiers. An average accuracy, sensitivity and specificity of 97.7 %, 97 %, and 94.2 % were obtained, respectively, with the CZ-A1 channel according to the R&K standard. The experimental results with high classification performance demonstrated that the technique can help doctors optimize the diagnosis and treatment of sleep disorders.

EMG based classification for pick and place task

The hand amputee is deprived of number of activities of daily living. To help the hand amputee, it is important to learn the pattern of muscles activity. There are several elements of tasks, which involve forearm along with the wrist and hand. The one very important task is pick and place activity performed by the hand. A pick and place action is a compilation of different finger motions for the grasping of objects at different force levels. This action may be better understood by learning the electromyography signals of forearm muscles. Electromyography is the technique to acquire electrical muscle activity that is used for the pattern recognition technique of assistive devices. Regarding this, the different classification characterizations of EMG signals involved in the pick and place action, subjected to variable grip span and weights were considered in this study. A low-level force measuring gripper, capable to bear the changes in weights and object spans was designed and developed to simulate the task. The grip span varied from 6 cm to 9 cm and the maximum weight used in this study was 750 gms. The pattern recognition classification methodology was performed for the differentiation of phases of the pick and place activity, grip force, and the angular deviation of metacarpal phalangeal (MCP) joint. The classifiers used in this study were decision tree (DT), support vector machines (SVM) and k-nearest neighbor (k-NN) based on the feature sets of the EMG signals. After analyses, it was found that k-NN performed best to classify different phases of the activity and relative deviation of MCP joint with an average classification accuracy of 82% and 91% respectively. However; the SVM performed best in classification of force with a particular feature set. The findings of the study would be helpful in designing the assistive devices for hand amputee.

Locomotion Mode Recognition for Walking on Three Terrains Based on sEMG of Lower Limb and Back Muscles

Gait phase detection on different terrains is an essential procedure for amputees with a lower limb assistive device to restore walking ability. In the present study, the intent recognition of gait events on three terrains based on sEMG was presented. The class separability and robustness of time, frequency, and time-frequency domain features of sEMG signals from five leg and back muscles were quantitatively evaluated by statistical analysis to select the best features set. Then, ensemble learning method that combines the outputs of multiple classifiers into a single fusion-produced output was implemented. The results obtained from data collected from four human participants revealed that the light gradient boosting machine (LightGBM) algorithm has an average accuracy of 93.1%, a macro-F1 score of 0.929, and a calculation time of prediction of 15 ms in discriminating 12 different gait phases on three terrains. This was better than traditional voting-based multiple classifier fusion methods. LightGBM is a perfect choice for gait phase detection on different terrains in daily life.

Fundamental research on surface electromyography analysis using discrete wavelet transform-an analysis of the central nervous system factors affecting muscle strength

[Purpose] We aimed to investigate the central nervous system factors that affect muscle strength based on the differences in load and time using the discrete wavelet transform, which is capable of a time-frequency-potential analysis. [Participants and Methods] Surface electromyography (EMG) of the right upper bicep muscle in 16 healthy adult males were measured at 10% MVC (maximum voluntary isometric contraction), 30%, 50%, 70%, and 80% to 100% MVC. We used a discrete wavelet transform for the electromyographic analysis and calculated the median instantaneous frequency spectrum (MDF) and frequency band component content rate (FCR) at 1-ms intervals as well as their spectrum integrated values (I-EMG). [Results] MDF and FCR tended to be high throughout the measurements. Specifically, the high-frequency band component content rate was high at the time of low muscle strength; fast-twitch muscle fibers may be involved during these muscle contractions. We found significant changes in the I-EMG as the muscle strength increased from 10% MVC to 100% MVC. [Conclusion] Analyzing the surface electromyograph using discrete wavelet transform enabled us to assess the central nervous system factors that increase in the EMG amplitude integrated values and change in the median instantaneous frequency spectrum and in the frequency band component content rate.

Improving Human Activity Recognition Performance by Data Fusion and Feature Engineering

Human activity recognition (HAR) is essential in many health-related fields. A variety of technologies based on different sensors have been developed for HAR. Among them, fusion from heterogeneous wearable sensors has been developed as it is portable, non-interventional and accurate for HAR. To be applied in real-time use with limited resources, the activity recognition system must be compact and reliable. This requirement can be achieved by feature selection (FS). By eliminating irrelevant and redundant features, the system burden is reduced with good classification performance (CP). This manuscript proposes a two-stage genetic algorithm-based feature selection algorithm with a fixed activation number (GFSFAN), which is implemented on the datasets with a variety of time, frequency and time-frequency domain features extracted from the collected raw time series of nine activities of daily living (ADL). Six classifiers are used to evaluate the effects of selected feature subsets from different FS algorithms on HAR performance. The results indicate that GFSFAN can achieve good CP with a small size. A sensor-to-segment coordinate calibration algorithm and lower-limb joint angle estimation algorithm are introduced. Experiments on the effect of the calibration and the introduction of joint angle on HAR shows that both of them can improve the CP.

Comparison study of classification methods of intramuscular electromyography data for non-human primate model of traumatic spinal cord injury

Traumatic spinal cord injury is a serious neurological disorder. Patients experience a plethora of symptoms that can be attributed to the nerve fiber tracts that are compromised. This includes limb weakness, sensory impairment, and truncal instability, as well as a variety of autonomic abnormalities. This article will discuss how machine learning classification can be used to characterize the initial impairment and subsequent recovery of electromyography signals in an non-human primate model of traumatic spinal cord injury. The ultimate objective is to identify potential treatments for traumatic spinal cord injury. This work focuses specifically on finding a suitable classifier that differentiates between two distinct experimental stages (pre-and post-lesion) using electromyography signals. Eight time-domain features were extracted from the collected electromyography data. To overcome the imbalanced dataset issue, synthetic minority oversampling technique was applied. Different ML classification techniques were applied including multilayer perceptron, support vector machine, K-nearest neighbors, and radial basis function network; then their performances were compared. A confusion matrix and five other statistical metrics (sensitivity, specificity, precision, accuracy, and F-measure) were used to evaluate the performance of the generated classifiers. The results showed that the best classifier for the left- and right-side data is the multilayer perceptron with a total F-measure of 79.5% and 86.0% for the left and right sides, respectively. This work will help to build a reliable classifier that can differentiate between these two phases by utilizing some extracted time-domain electromyography features.

Current Trends and Confounding Factors in Myoelectric Control: Limb Position and Contraction Intensity

This manuscript presents a hybrid study of a comprehensive review and a systematic(research) analysis. Myoelectric control is the cornerstone ofmany assistive technologies used in clinicalpractice, such as prosthetics and orthoses, and human-computer interaction, such as virtual reality control.Although the classification accuracy of such devices exceeds 90% in a controlled laboratory setting,myoelectric devices still face challenges in robustness to variability of daily living conditions.The intrinsic physiological mechanisms limiting practical implementations of myoelectric deviceswere explored: the limb position effect and the contraction intensity effect. The degradationof electromyography (EMG) pattern recognition in the presence of these factors was demonstratedon six datasets, where classification performance was 13% and 20% lower than the controlledsetting for the limb position and contraction intensity effect, respectively. The experimental designsof limb position and contraction intensity literature were surveyed. Current state-of-the-art trainingstrategies and robust algorithms for both effects were compiled and presented. Recommendationsfor future limb position effect studies include: the collection protocol providing exemplars of at least 6positions (four limb positions and three forearm orientations), three-dimensional space experimentaldesigns, transfer learning approaches, and multi-modal sensor configurations. Recommendationsfor future contraction intensity effect studies include: the collection of dynamic contractions, nonlinearcomplexity features, and proportional control.

Wearable-Sensor-based Detection and Prediction of Freezing of Gait in Parkinson's Disease: A Review

Freezing of gait (FOG) is a serious gait disturbance, common in mid- and late-stage Parkinson's disease, that affects mobility and increases fall risk. Wearable sensors have been used to detect and predict FOG with the ultimate aim of preventing freezes or reducing their effect using gait monitoring and assistive devices. This review presents and assesses the state of the art of FOG detection and prediction using wearable sensors, with the intention of providing guidance on current knowledge, and identifying knowledge gaps that need to be filled and challenges to be considered in future studies. This review searched the Scopus, PubMed, and Web of Science databases to identify studies that used wearable sensors to detect or predict FOG episodes in Parkinson's disease. Following screening, 74 publications were included, comprising 68 publications detecting FOG, seven predicting FOG, and one in both categories. Details were extracted regarding participants, walking task, sensor type and body location, detection or prediction approach, feature extraction and selection, classification method, and detection and prediction performance. The results showed that increasingly complex machine-learning algorithms combined with diverse feature sets improved FOG detection. The lack of large FOG datasets and highly person-specific FOG manifestation were common challenges. Transfer learning and semi-supervised learning were promising for FOG detection and prediction since they provided person-specific tuning while preserving model generalization.

Selection of Features and Classifiers for EMG-EEG-Based Upper Limb Assistive Devices-A Review

Bio-signals are distinctive factors in the design of human-machine interface, essentially useful for prosthesis, orthosis, and exoskeletons. Despite the progress in the analysis of pattern recognition based devices; the acceptance of these devices is still questionable. One reason is the lack of information to identify the possible combinations of features and classifiers. Besides; there is also a need for optimal selection of various sensors for sensations such as touch, force, texture, along with EMGs/EEGs. This article reviews the two bio-signal techniques, named as electromyography and electroencephalography. The details of the features and the classifiers used in the data processing for upper limb assist devices are summarised here. Various features and their sets are surveyed and different classifiers for feature sets are discussed on the basis of the classification rate. The review was carried out on the basis of the last 10-12 years of published research in this area. This article also outlines the influence of modality of EMGs and EEGs with other sensors on classifications. Also, other bio-signals used in upper limb devices and future aspects are considered.

Support Vector Machine-Based EMG Signal Classification Techniques: A Review

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.

putEMG-A Surface Electromyography Hand Gesture Recognition Dataset

In this paper, we present a putEMG dataset intended for the evaluation of hand gesture recognition methods based on sEMG signal. The dataset was acquired for 44 able-bodied subjects and include 8 gestures (3 full hand gestures, 4 pinches and idle). It consists of uninterrupted recordings of 24 sEMG channels from the subject's forearm, RGB video stream and depth camera images used for hand motion tracking. Moreover, exemplary processing scripts are also published. The putEMG dataset is available under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0). The dataset was validated regarding sEMG amplitudes and gesture recognition performance. The classification was performed using state-of-the-art classifiers and feature sets. An accuracy of 90% was achieved for SVM classifier utilising RMS feature and for LDA classifier using Hudgin's and Du's feature sets. Analysis of performance for particular gestures showed that LDA/Du combination has significantly higher accuracy for full hand gestures, while SVM/RMS performs better for pinch gestures. The presented dataset can be used as a benchmark for various classification methods, the evaluation of electrode localisation concepts, or the development of classification methods invariant to user-specific features or electrode displacement.

Neurophysiological Characterization of a Non-Human Primate Model of Traumatic Spinal Cord Injury Utilizing Fine-Wire EMG Electrodes

This study aims to characterize traumatic spinal cord injury (TSCI) neurophysiologically using an intramuscular fine-wire electromyography (EMG) electrode pair. EMG data were collected from an agonist-antagonist pair of tail muscles of Macaca fasicularis, pre- and post-lesion, and for a treatment and control group. The EMG signals were decomposed into multi-resolution subsets using wavelet transforms (WT), then the relative power (RP) was calculated for each individual reconstructed EMG sub-band. Linear mixed models were developed to test three hypotheses: (i) asymmetrical volitional activity of left and right side tail muscles (ii) the effect of the experimental TSCI on the frequency content of the EMG signal, (iii) and the effect of an experimental treatment. The results from the electrode pair data suggested that there is asymmetry in the EMG response of the left and right side muscles (p-value < 0.001). This is consistent with the construct of limb dominance. The results also suggest that the lesion resulted in clear changes in the EMG frequency distribution in the post-lesion period with a significant increment in the low-frequency sub-bands (D4, D6, and A6) of the left and right side, also a significant reduction in the high-frequency sub-bands (D1 and D2) of the right side (p-value < 0.001). The preliminary results suggest that using the RP of the EMG data, the fine-wire intramuscular EMG electrode pair are a suitable method of monitoring and measuring treatment effects of experimental treatments for spinal cord injury (SCI).

A Novel Method for Classifying Liver and Brain Tumors Using Convolutional Neural Networks, Discrete Wavelet Transform and Long Short-Term Memory Networks

Rapid classification of tumors that are detected in the medical images is of great importance in the early diagnosis of the disease. In this paper, a new liver and brain tumor classification method is proposed by using the power of convolutional neural network (CNN) in feature extraction, the power of discrete wavelet transform (DWT) in signal processing, and the power of long short-term memory (LSTM) in signal classification. A CNN-DWT-LSTM method is proposed to classify the computed tomography (CT) images of livers with tumors and to classify the magnetic resonance (MR) images of brains with tumors. The proposed method classifies liver tumors images as benign or malignant and then classifies brain tumor images as meningioma, glioma, and pituitary. In the hybrid CNN-DWT-LSTM method, the feature vector of the images is obtained from pre-trained AlexNet CNN architecture. The feature vector is reduced but strengthened by applying the single-level one-dimensional discrete wavelet transform (1-D DWT), and it is classified by training with an LSTM network. Under the scope of the study, images of 56 benign and 56 malignant liver tumors that were obtained from Fırat University Research Hospital were used and a publicly available brain tumor dataset were used. The experimental results show that the proposed method had higher performance than classifiers, such as K-nearest neighbors (KNN) and support vector machine (SVM). By using the CNN-DWT-LSTM hybrid method, an accuracy rate of 99.1% was achieved in the liver tumor classification and accuracy rate of 98.6% was achieved in the brain tumor classification. We used two different datasets to demonstrate the performance of the proposed method. Performance measurements show that the proposed method has a satisfactory accuracy rate at the liver tumor and brain tumor classifying.

UP-Fall Detection Dataset: A Multimodal Approach

Falls, especially in elderly persons, are an important health problem worldwide. Reliable fall detection systems can mitigate negative consequences of falls. Among the important challenges and issues reported in literature is the difficulty of fair comparison between fall detection systems and machine learning techniques for detection. In this paper, we present UP-Fall Detection Dataset. The dataset comprises raw and feature sets retrieved from 17 healthy young individuals without any impairment that performed 11 activities and falls, with three attempts each. The dataset also summarizes more than 850 GB of information from wearable sensors, ambient sensors and vision devices. Two experimental use cases were shown. The aim of our dataset is to help human activity recognition and machine learning research communities to fairly compare their fall detection solutions. It also provides many experimental possibilities for the signal recognition, vision, and machine learning community.

Development of a deep neural network for automated electromyographic pattern classification

Determining the signal quality of surface electromyography (sEMG) recordings is time consuming and requires the judgement of trained observers. An automated procedure to evaluate sEMG quality would streamline data processing and reduce time demands. This paper compares the performance of two supervised and three unsupervised artificial neural networks (ANNs) in the evaluation of sEMG quality. Manually classified sEMG recordings from various lower-limb muscles during motor tasks were used to train (n=28,000), test performance (n=12,000) and evaluate accuracy (n=47,000) of the five ANNs in classifying signals into four categories. Unsupervised ANNs demonstrated a 30-40% increase in classification accuracy (>98%) compared with supervised ANNs. AlexNet demonstrated the highest accuracy (99.55%) with negligible false classifications. The results indicate that sEMG quality evaluation can be automated via an ANN without compromising human-like classification accuracy. This classifier will be publicly available and will be a valuable tool for researchers and clinicians using electromyography.

Feature Selection Based on Binary Tree Growth Algorithm for the Classification of Myoelectric Signals

Electromyography (EMG) has been widely used in rehabilitation and myoelectric prosthetic applications. However, a recent increment in the number of EMG features has led to a high dimensional feature vector. This in turn will degrade the classification performance and increase the complexity of the recognition system. In this paper, we have proposed two new feature selection methods based on a tree growth algorithm (TGA) for EMG signals classification. In the first approach, two transfer functions are implemented to convert the continuous TGA into a binary version. For the second approach, the swap, crossover, and mutation operators are introduced in a modified binary tree growth algorithm for enhancing the exploitation and exploration behaviors. In this study, short time Fourier transform (STFT) is employed to transform the EMG signals into time-frequency representation. The features are then extracted from the STFT coefficient and form a feature vector. Afterward, the proposed feature selection methods are applied to evaluate the best feature subset from a large available feature set. The experimental results show the superiority of MBTGA not only in terms of feature reduction, but also the classification performance.

An Investigation of Temporally Inspired Time Domain Features for Electromyographic Pattern Recognition

This paper presents a novel set of temporally inspiredtime domain features for electromyographic (EMG) pattern recognition. The proposed methods employ simple time series measures derived from peak detection, and could better reflect EMG activity over time. Multiple EMG datasets consisting of 68 able-bodied and transradial amputee subjects performing a large variety of hand, wrist, fingers, and grasping movements are used to evaluate the performance of the proposed features and to design robust EMG feature sets. The results show that the average classification accuracy of two novel features, the mean prominence of local peaks and valleys, outperform several commonly used time domain features, autoregressive coefficients, histogram, and zero crossing, by 8{\%, 11{\%, and 17{\%, respectively. The proposed features are also shown to provide additional information as part of a robust feature set when compared to the common Hudgins' timedomain feature set, as selected by sequential forward selection and through empirical feature set design.

Navigating features: a topologically informed chart of electromyographic features space

The success of biological signal pattern recognition depends crucially on the selection of relevant features. Across signal and imaging modalities, a large number of features have been proposed, leading to feature redundancy and the need for optimal feature set identification. A further complication is that, due to the inherent biological variability, even the same classification problem on different datasets can display variations in the respective optimal sets, casting doubts on the generalizability of relevant features. Here, we approach this problem by leveraging topological tools to create charts of features spaces. These charts highlight feature sub-groups that encode similar information (and their respective similarities) allowing for a principled and interpretable choice of features for classification and analysis. Using multiple electromyographic (EMG) datasets as a case study, we use this feature chart to identify functional groups among 58 state-of-the-art EMG features, and to show that they generalize across three different forearm EMG datasets obtained from able-bodied subjects during hand and finger contractions. We find that these groups describe meaningful non-redundant information, succinctly recapitulating information about different regions of feature space. We then recommend representative features from each group based on maximum class separability, robustness and minimum complexity.

EMG Processing Based Measures of Fatigue Assessment during Manual Lifting

Manual lifting is one of the common practices used in the industries to transport or move objects to a desired place. Nowadays, even though mechanized equipment is widely available, manual lifting is still considered as an essential way to perform material handling task. Improper lifting strategies may contribute to musculoskeletal disorders (MSDs), where overexertion contributes as the highest factor. To overcome this problem, electromyography (EMG) signal is used to monitor the workers' muscle condition and to find maximum lifting load, lifting height and number of repetitions that the workers are able to handle before experiencing fatigue to avoid overexertion. Past researchers have introduced several EMG processing techniques and different EMG features that represent fatigue indices in time, frequency, and time-frequency domain. The impact of EMG processing based measures in fatigue assessment during manual lifting are reviewed in this paper. It is believed that this paper will greatly benefit researchers who need a bird's eye view of the biosignal processing which are currently available, thus determining the best possible techniques for lifting applications.

A Flexible Approach for Human Activity Recognition Using Artificial Hydrocarbon Networks

Physical activity recognition based on sensors is a growing area of interest given the great advances in wearable sensors. Applications in various domains are taking advantage of the ease of obtaining data to monitor personal activities and behavior in order to deliver proactive and personalized services. Although many activity recognition systems have been developed for more than two decades, there are still open issues to be tackled with new techniques. We address in this paper one of the main challenges of human activity recognition: Flexibility. Our goal in this work is to present artificial hydrocarbon networks as a novel flexible approach in a human activity recognition system. In order to evaluate the performance of artificial hydrocarbon networks based classifier, experimentation was designed for user-independent, and also for user-dependent case scenarios. Our results demonstrate that artificial hydrocarbon networks classifier is flexible enough to be used when building a human activity recognition system with either user-dependent or user-independent approaches.