Classification of Electromyographic Hand Gesture Signals Using Modified Fuzzy C-Means Clustering and Two-Step Machine Learning Approach

Comparative Study of sEMG Feature Evaluation Methods Based on the Hand Gesture Classification Performance

Effective feature extraction and selection are crucial for the accurate classification and prediction of hand gestures based on electromyographic signals. In this paper, we systematically compare six filter and wrapper feature evaluation methods and investigate their respective impacts on the accuracy of gesture recognition. The investigation is based on several benchmark datasets and one real hand gesture dataset, including 15 hand force exercises collected from 14 healthy subjects using eight commercial sEMG sensors. A total of 37 time- and frequency-domain features were extracted from each sEMG channel. The benchmark dataset revealed that the minimum Redundancy Maximum Relevance (mRMR) feature evaluation method had the poorest performance, resulting in a decrease in classification accuracy. However, the RFE method demonstrated the potential to enhance classification accuracy across most of the datasets. It selected a feature subset comprising 65 features, which led to an accuracy of 97.14%. The Mutual Information (MI) method selected 200 features to reach an accuracy of 97.38%. The Feature Importance (FI) method reached a higher accuracy of 97.62% but selected 140 features. Further investigations have shown that selecting 65 and 75 features with the RFE methods led to an identical accuracy of 97.14%. A thorough examination of the selected features revealed the potential for three additional features from three specific sensors to enhance the classification accuracy to 97.38%. These results highlight the significance of employing an appropriate feature selection method to significantly reduce the number of necessary features while maintaining classification accuracy. They also underscore the necessity for further analysis and refinement to achieve optimal solutions.

Deep Learning for Electromyographic Lower-Limb Motion Signal Classification Using Residual Learning

Electromyographic (EMG) signals have gained popularity for controlling prostheses and exoskeletons, particularly in the field of upper limbs for stroke patients. However, there is a lack of research in the lower limb area, and standardized open-source datasets of lower limb EMG signals, especially recording data of Asian race features, are scarce. Additionally, deep learning algorithms are rarely used for human motion intention recognition based on EMG, especially in the lower limb area. In response to these gaps, we present an open-source benchmark dataset of lower limb EMG with Asian race characteristics and large data volume, the JJ dataset, which includes approximately 13,350 clean EMG segments of 10 gait phases from 15 people. This is the first dataset of its kind to include the nine main muscles of human gait when walking. We used the processed time-domain signal as input and adjusted ResNet-18 as the classification tool. Our research explores and compares multiple key issues in this area, including the comparison of sliding time window method and other preprocessing methods, comparison of time-domain and frequency-domain signal processing effects, cross-subject motion recognition accuracy, and the possibility of using thigh and calf muscles in amputees. Our experiments demonstrate that the adjusted ResNet can achieve significant classification accuracy, with an average accuracy rate of 95.34% for human gait phases. Our research provides a valuable resource for future studies in this area and demonstrates the potential for ResNet as a robust and effective method for lower limb human motion intention pattern recognition.

Developmental Prediction of Poststroke Patients in Activities of Daily Living by Using Tree-Structured Parzen Estimator-Optimized Stacking Ensemble Approaches

Poststroke injuries limit the daily activities of patients and cause considerable inconvenience. Therefore, predicting the activities of daily living (ADL) results of patients with stroke before hospital discharge can assist clinical workers in formulating more personalized and effective strategies for therapeutic intervention, and prepare hospital discharge plans that suit the patients needs. This study used the leave-one-out cross-validation procedure to evaluate the performance of the machine learning models. In addition, testing methods were used to identify the optimal weak learners, which were then combined to form a stacking model. Subsequently, a hyperparameter optimization algorithm was used to optimize the model hyperparameters. Finally, optimization algorithms were used to analyze each feature, and features of high importance were identified by limiting the number of features to be included in the machine learning models. After various features were fed into the learning models to predict the Barthel index (BI) at discharge, the results indicated that random forest (RF), adaptive boosting (AdaBoost), and multilayer perceptron (MLP) produced suitable results. The most critical prediction factor of this study was the BI at admission. Machine learning models can be used to assist clinical workers in predicting the ADL of patients with stroke at hospital discharge.

Novel near E-Field Topography Sensor for Human-Machine Interfacing in Robotic Applications

This work investigates a new sensing technology for use in robotic human-machine interface (HMI) applications. The proposed method uses near E-field sensing to measure small changes in the limb surface topography due to muscle actuation over time. The sensors introduced in this work provide a non-contact, low-computational-cost, and low-noise method for sensing muscle activity. By evaluating the key sensor characteristics, such as accuracy, hysteresis, and resolution, the performance of this sensor is validated. Then, to understand the potential performance in intention detection, the unmodified digital output of the sensor is analysed against movements of the hand and fingers. This is done to demonstrate the worst-case scenario and to show that the sensor provides highly targeted and relevant data on muscle activation before any further processing. Finally, a convolutional neural network is used to perform joint angle prediction over nine degrees of freedom, achieving high-level regression performance with an RMSE value of less than six degrees for thumb and wrist movements and 11 degrees for finger movements. This work demonstrates the promising performance of this novel approach to sensing for use in human-machine interfaces.

Design of Fuzzy Logic Motion Detection Algorithm for the Bracelet Type Sensor Consisting of Conductive Layer-Polymer Composite Film

To improve the motion detection performance of a bracelet-type sensor that uses only two tiny sensor modules developed using carbon-based conductive polymer composite films, a fuzzy-logic algorithm was developed in this study. A polyethylene terephthalate polymer film with a conductive layer composed of carbon paste was used as the integral material utilized for the composite film; a small sensor module composed of mechanical parts mounted on the film was developed to effectively detect the surface resistance variations of the film. A participant wore a bracelet sensor, which consisted of two sensor modules, on their forearm, and the resistance variations of the contact area between the forearm and the sensor modules corresponding to the flexion changes of the surface of the body due to muscle contraction and relaxation were detected. The surface resistance variations of the film were converted to voltage signals, which were used as inputs to the fuzzy logic algorithm to detect four consecutive motions of the forearm. The results demonstrated that the fuzzy-logic algorithm attained an accuracy of 94%. The fuzzy algorithm successfully detected four motions and the resting state of the forearms; moreover, it showed improved performance compared to previous research.

ESTIMATION OF KNEE JOINT TORQUE DURING SIT–STAND MOVEMENT BASED ON sEMG SIGNALS USING NEURAL NETWORKS

The estimation of knee joint torque is important for the development of powered exoskeletons to achieve ideal gait characteristics. In this study, we proposed three different models to predict the required torque for performing sit-to-stand (STS) and back-to-sit (BTS) movements. The surface electromyography (sEMG) signals were extracted from the biceps femoris and rectus femoris muscles during STS and BTS movements. The time-domain features selected as input to the models for torque prediction are integrated EMG (iEMG), root mean square (RMS), and mean absolute value (MAV). Two-way ANOVA analysis identifies the significance of NN models and EMG features of the muscles in predicting the knee joint torque requirement. The artificial neural network models selected for prediction are the feed-forward back-propagation algorithm, ANFIS, and NARX. The theoretical value of knee joint torque calculated using the Lagrange method was compared with the torque output for each model based on root mean square error (RMSE). The desired torque predicted using the NARX model confirms to have the least average error ([Formula: see text][Formula: see text]Nm), which indicates that NARX can estimate knee joint torque more accurately from sEMG than other models.

Wearable Sensor for Forearm Motion Detection Using a Carbon-Based Conductive Layer-Polymer Composite Film

In this study, we developed a fabrication method for a bracelet-type wearable sensor to detect four motions of the forearm by using a carbon-based conductive layer-polymer composite film. The integral material used for the composite film is a polyethylene terephthalate polymer film with a conductive layer composed of a carbon paste. It is capable of detecting the resistance variations corresponding to the flexion changes of the surface of the body due to muscle contraction and relaxation. To effectively detect the surface resistance variations of the film, a small sensor module composed of mechanical parts mounted on the film was designed and fabricated. A subject wore the bracelet sensor, consisting of three such sensor modules, on their forearm. The surface resistance of the film varied corresponding to the flexion change of the contact area between the forearm and the sensor modules. The surface resistance variations of the film were converted to voltage signals and used for motion detection. The results demonstrate that the thin bracelet-type wearable sensor, which is comfortable to wear and easily applicable, successfully detected each motion with high accuracy.

Toward Hand Pattern Recognition in Assistive and Rehabilitation Robotics Using EMG and Kinematics

Wearable hand robots are becoming an attractive means in the facilitating of assistance with daily living and hand rehabilitation exercises for patients after stroke. Pattern recognition is a crucial step toward the development of wearable hand robots. Electromyography (EMG) is a commonly used biological signal for hand pattern recognition. However, the EMG based pattern recognition performance in assistive and rehabilitation robotics post stroke remains unsatisfactory. Moreover, low cost kinematic sensors such as Leap Motion is recently used for pattern recognition in various applications. This study proposes feature fusion and decision fusion method that combines EMG features and kinematic features for hand pattern recognition toward application in upper limb assistive and rehabilitation robotics. Ten normal subjects and five post stroke patients participating in the experiments were tested with eight hand patterns of daily activities while EMG and kinematics were recorded simultaneously. Results showed that average hand pattern recognition accuracy for post stroke patients was 83% for EMG features only, 84.71% for kinematic features only, 96.43% for feature fusion of EMG and kinematics, 91.18% for decision fusion of EMG and kinematics. The feature fusion and decision fusion was robust as three different levels of noise was given to the classifiers resulting in small decrease of classification accuracy. Different channel combination comparisons showed the fusion classifiers would be robust despite failure of specific EMG channels which means that the system has promising potential in the field of assistive and rehabilitation robotics. Future work will be conducted with real-time pattern classification on stroke survivors.

On the Use of Fuzzy and Permutation Entropy in Hand Gesture Characterization from EMG Signals: Parameters Selection and Comparison

The surface electromyography signal (sEMG) is widely used for gesture characterization; its reliability is strongly connected to the features extracted from sEMG recordings. This study aimed to investigate the use of two complexity measures, i.e., fuzzy entropy (FEn) and permutation entropy (PEn) for hand gesture characterization. Fourteen upper limb movements, sorted into three sets, were collected on ten subjects and the performances of FEn and PEn for gesture descriptions were analyzed for different computational parameters. FEn and PEn were able to properly cluster the expected numbers of gestures, but computational parameters were crucial for ensuring clusters’ separability and proper gesture characterization. FEn and PEn were also compared with other eighteen classical time and frequency domain features through the minimum redundancy maximum relevance algorithm and showed the best predictive importance scores in two gesture sets; they also had scores within the subset of the best five features in the remaining one. Further, the classification accuracies of four different feature sets presented remarkable increases when FEn and PEn are included as additional features. Outcomes support the use of FEn and PEn for hand gesture description when computational parameters are properly selected, and they could be useful in supporting the development of robotic arms and prostheses myoelectric control.