Heterogeneous computing model for post‐injury walking pattern restoration and postural stability rehabilitation exercise recognition

Bijalwan, Vishwanath; Semwal, Vijay Bhaskar; Singh, Ghanapriya; Crespo, Rubén González

doi:10.1111/exsy.12706

Cited by 38 publications

(15 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Bijalwan, Semwal, Singh, and González‐Crespo (2021), the pattern of 10 different rehabilitation activities was captured using RGB‐Depth (RGB‐D) camera and classified using heterogeneous deep learning models. Different deep learning models Convolutional Neural Network (CNN) and CNN‐LSTM (CNN‐Long Short Term Memory) were used for the classification of these rehabilitation exercises.…”

Section: Literature Surveymentioning

confidence: 99%

Human activity recognition by wireless body area networks through multi‐objective feature selection with deep learning

Boga

Kumar

2022

Expert Systems

View full text Add to dashboard Cite

Wireless body area network (WBAN) is a novel technology with the incorporation of numerous types of devices, which is also employed in health monitoring applications.Human activity recognition (HAR) receives more interest in recent times along with wearable sensors. HAR system provides information about a person's identity, personality, and psychological state. In the present scenario, it becomes important to model the active learning paradigms with the help of wearable sensors for analysing human activities. Although various deep learning models and existing algorithms secure better outcomes through the sensor data analysis regarding HAR, the decision-making evaluation seems to be a complex one. The main intent of this paper is to implement the WBAN-based HAR system using the improved deep learning model. By connecting the wearable sensors, the signals are gathered regarding human activity from diverse benchmark sources. After collecting the required signals, the pre-processing of the input signals is done using artefact removal and median filtering. Further, the feature extraction is performed, which intends to extract a set of features by utilizing short-time Fourier transform (STFT) and statistical features. For reducing the feature-length, a multi-objective-based optimal feature selection is adopted. Human activities such as 'walking, walking upstairs, walking downstairs, sitting, standing, lying, and jogging' are recognized with help of selected optimal features. The optimized probabilistic neural network (PNN) and convolutional neural network (CNN) are combined and named as adaptive probabilistic-based CNN (AP-CNN). The effective performance of optimal feature selection and recognition is accomplished by incorporating the developed backward updating position-based sea lion optimization algorithm (BU-SLnO). Finally, the performance of the BU-SLnO-AP-CNN-based suggested model is analysed, which shows 1.001%, 1.237%, 0.811%, and 0.859% advanced than SLnO-AP-CNN, feed-forward-AP-CNN (FF-AP-CNN), grey wolf optimization-AP-CNN (GWO-AP-CNN), particle swarm optimization-AP-CNN (PSO-AP-CNN) when observing the dataset 1. The experimental outcomes from comparison with various classification techniques demonstrate the efficiency of the developed technique.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Human activity recognition by wireless body area networks through multi‐objective feature selection with deep learning

Boga

Kumar

2022

Expert Systems

View full text Add to dashboard Cite

show abstract

“…GRU involves two gates: reset gate rt and update gate zt. The state of rt and zt is updated by (10) and (11) individually:…”

Section: Deep Regression Neural Networkmentioning

confidence: 99%

Synergy-Based Neural Interface for Human Gait Tracking With Deep Learning

Xiong

Zhang

Zhao

et al. 2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Neural information decomposed from electromyography (EMG) signals provides a new path of EMG-based human-machine interface. Instead of the motor unit decomposition-based method, this work presents a novel neural interface for human gait tracking based on muscle synergy, the high-level neural control information to collaborate muscle groups for performing movements. Three classical synergy extraction approaches include Principle Component Analysis (PCA), Factor Analysis (FA), and Nonnegative Matrix Factorization (NMF), are employed for muscle synergy extraction. A deep regression neural network based on the bidirectional gated recurrent unit (BGRU) is used to extract temporal information from the synergy matrix to estimate joint angles of the lower limb. Eight subjects participated in the experiment while walking at four types of speed: 0.5km/h, 1.0km/h, 2.0km/h, and 3.0km/h. Two machine learning methods based on linear regression (LR) and multilayer perceptron (MLP) are set as the contrast group. The result shows that the synergy-based approach's performance outperforms two contrast methods with scores of 0.83~0.88. PCA reaches the highest performance of 0.871±0.029, corresponding to RMSE of 3.836⁰, 6.278⁰, 2.197⁰ for hip, knee, and ankle, respectively. The effect of walking speed, synergy number, and joint location will be analyzed. The performance shows that muscle synergy has a good correlation will joint angles which can be unearthed by deep learning. The proposed method explores a new way for gait analysis and contributes to building a novel neural interface with muscle synergy and deep learning.

show abstract

“…In these, the researchers used some Machine Learning algorithm to classify the signals. However, it is still possible to highlight other works that could use ANN for classification of biopotential signals with applications in robotic prostheses as in (Ahirwar et al, 2022; Bijalwan et al, 2021; Challa et al, 2021; Patil et al, 2019; Semwal et al, 2021; Semwal et al, 2022). For this work, the following contributions were made: Conception of a low‐cost prototype that performs movement in a robotic arm prosthesis. System integration with acquisition subsystems, information processing, artificial neural network training, cloud storage and application with graphical user interface. Final prototype with the ability to correctly classify the largest number of gestures performed by different subjects under analysis. …”

Section: Introductionmentioning

confidence: 99%

Development of intelligent and integrated technology for pattern recognition in EMG signals for robotic prosthesis command

Barbosa

Neto

et al. 2022

Expert Systems

View full text Add to dashboard Cite

Prostheses play an important role in the rehabilitation of people who have suffered some type of amputation. However, due to its high‐cost and high complexity in performing movements of everyday tasks, users of these prostheses may encounter many difficulties. Therefore, this work proposes the development of a future artificial intelligence technology based on a low‐cost functional prosthesis prototype (manufactured in a 3D printer). In the present work, we describe an intelligent system that uses an artificial neural network to recognize patterns in muscle biopotential signals in order to control a prosthesis prototype in real time. Such a system is divided into three parts: the first that performs a human–machine integration through a graphical user interface; the second that performs the signal acquisition; the third that performs the training and generalization steps of the artificial neural network. The developed interface runs on a web application that has a database hosted in the cloud and in it the system user can: Acquisition of electromyography signals; Training phase of the artificial neural network; Sends the matrix of weights of the trained network to the microcontroller; Activates in the microcontroller, the state of action of the commands from the identified gestures. To compose the results of the present work, a search was initially carried out for the ideal parameters of the artificial neural network through signals obtained from 20 volunteers. In this step, it was possible to identify the topology that best classifies the signals of each gesture, as well as the investigation of the number of neurons in the hidden layer that causes a low generalization power due to overfitting. At the end of the project, it was possible to validate the use of the system with 15 new volunteers, and it was observed that in most cases, the performance of the commands in the prosthesis prototype were performed correctly. In addition, a project cost analysis was carried out, and it was possible to verify that the prototype developed is viable and has an affordable cost in relation to the Brazilian cost of living standards. In this way, the objective of the present work is in the development of a low cost artificial intelligence technology. Such a system is equipped with an algorithm based on neural networks that can deal with different muscle biopotential signals, in order to command a robotic prosthesis.

show abstract

Heterogeneous computing model for post‐injury walking pattern restoration and postural stability rehabilitation exercise recognition

Cited by 38 publications

References 36 publications

Human activity recognition by wireless body area networks through multi‐objective feature selection with deep learning

Human activity recognition by wireless body area networks through multi‐objective feature selection with deep learning

Synergy-Based Neural Interface for Human Gait Tracking With Deep Learning

Development of intelligent and integrated technology for pattern recognition in EMG signals for robotic prosthesis command

Contact Info

Product

Resources

About