Intelligent Classification of Multi-Gesture EMG Signals Based on LSTM

Li, Zhongzhi; Zuo, Jiankai; Han, Zhixuan; Han, Xu; Sun, Chengquan; Wang, Zhijian

doi:10.1109/aiea51086.2020.00020

Cited by 8 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To obtain time-frequency data, multidimensional features are extracted from the one-dimensional EMG signal data by applying short-time Fourier transform (STFT) [87,88], which is often utilized to determine the frequency component of a signal, as shown in equation ( 1)…”

Section: Comparison Of Pre-processing Methodsmentioning

confidence: 99%

“…In addition, the weight loss problem may occur as the layer gets deeper which might cause inaccurate classification results. Lastly, long short-term memory (LSTM) is the improved version of RNN, in which each memory and result value can be controlled, so it is utilized for time series prediction just as RNN [86][87][88][89]. Although it can cope with extended layers with more flexibility than RNN does, it also shows problematic features such as low computation speed, potential memory over-writing, and incapability to process an infinite length of data.…”

Section: Comparison Of Different Deep-learning Algorithmsmentioning

confidence: 99%

See 1 more Smart Citation

Recent advances in wearable exoskeletons for human strength augmentation

Lee¹,

Kwon²,

Yeo³

2022

Flex. Print. Electron.

View full text Add to dashboard Cite

The decline in muscular strength and control due to age or stroke-related side-effect has afflicted many individuals with neuromotor disorders because it affects essential motor functions to perform everyday activities and restrains their functional independence. In this regard, a myriad of wearable exoskeletons and functional components have been developed to deliver mechanical force for assisting the movements of different human body parts. However, many of the reported wearable exoskeletons suffer from several critical drawbacks that limit functional usage and practicality despite the significant technological advance of general wearable exoskeletons. This review offers a comprehensive summary of the recent advances of wearable exoskeletons and their constituting functional components. In addition, we discuss the essential challenges that need to be tackled to enhance the functional practicality of the next-generation wearable exoskeletons in assisting the strength and control of individuals with neuromotor disorders.

show abstract

Section: Comparison Of Pre-processing Methodsmentioning

confidence: 99%

Section: Comparison Of Different Deep-learning Algorithmsmentioning

confidence: 99%

Recent advances in wearable exoskeletons for human strength augmentation

Lee¹,

Kwon²,

Yeo³

2022

Flex. Print. Electron.

View full text Add to dashboard Cite

show abstract

“…Temporal order is very important in processing data derived from continuous human motion [36]. For data learning and prediction considering this temporal order, LSTM has shown good results in several studies [37].…”

Section: ) Lstmmentioning

confidence: 99%

Continuous Intention Prediction of Lifting Motions Using EMG-Based CNN-LSTM

Gwon,

Woo,

Sahithi

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Industrial exoskeletons is a field of ongoing research for improving human safety and conveniences. However, the adoption of industrial exoskeleton robots still remains challenging. One problem that needs to be solved is the control delay that inevitably occurs due to data transmission and processing issues. Recently, there has been active research employing deep learning to address control delay by leveraging diverse information extracted from human motion. One crucial source of information is electromyography (EMG) signals, known for their quicker activation compared to actual motion. This study specifically focused on predicting changing motion intentions within the squat, a representative lifting motion in industrial contexts. In an experimental setup involving 24 participants and utilizing 7 EMG electrodes, we categorized motion intentions during the squat into four types. We developed a CNN-LSTM model capable of predicting motion intentions 300 milliseconds ahead using EMG signals. The model's prediction performance was assessed by comparing them with existing models. The findings propose a methodology for utilizing EMG signals in predicting changing motion intentions for lower extremity movements, facilitating feedforward control of industrial exoskeleton robots. This research is anticipated to contribute to the advancement of acceptance in the realm of exoskeleton robots.

show abstract

“…In these approaches, EMG data are represented as images, and a Convolutional Neural Network (CNN) is used to determine the type of gesture. Although EMG signals are time-series data, to our knowledge, only recently appropriate DL models (e.g., Recurrent Neural Network-RNN, such as Long Short-Term Memory-LSTM) have been utilized [6][7][8]. In this work, taking into account the outcomes of [9], we investigate the application of temporal convolutional networks (TCN) [10] to the problem of sEMGbased gesture recognition.…”

Section: Introductionmentioning

confidence: 99%

Real-Time Analysis of Hand Gesture Recognition with Temporal Convolutional Networks

Tsinganos

Jansen

Cornelis

et al. 2022

Sensors

View full text Add to dashboard Cite

In recent years, the successful application of Deep Learning methods to classification problems has had a huge impact in many domains. (1) Background: In biomedical engineering, the problem of gesture recognition based on electromyography is often addressed as an image classification problem using Convolutional Neural Networks. Recently, a specific class of these models called Temporal Convolutional Networks (TCNs) has been successfully applied to this task. (2) Methods: In this paper, we approach electromyography-based hand gesture recognition as a sequence classification problem using TCNs. Specifically, we investigate the real-time behavior of our previous TCN model by performing a simulation experiment on a recorded sEMG dataset. (3) Results: The proposed network trained with data augmentation yields a small improvement in accuracy compared to our existing model. However, the classification accuracy is decreased in the real-time evaluation, showing that the proposed TCN architecture is not suitable for such applications. (4) Conclusions: The real-time analysis helps in understanding the limitations of the model and exploring new ways to improve its performance.

show abstract

Intelligent Classification of Multi-Gesture EMG Signals Based on LSTM

Cited by 8 publications

References 15 publications

Recent advances in wearable exoskeletons for human strength augmentation

Recent advances in wearable exoskeletons for human strength augmentation

Continuous Intention Prediction of Lifting Motions Using EMG-Based CNN-LSTM

Real-Time Analysis of Hand Gesture Recognition with Temporal Convolutional Networks

Contact Info

Product

Resources

About