Temporal Dilation of Deep LSTM for Agile Decoding of sEMG: Application in Prediction of Upper-Limb Motor Intention in NeuroRobotics

Sun, Tianyun; Hu, Qin; Gulati, Paras; Atashzar, S. Farokh

doi:10.1109/lra.2021.3091698

Cited by 31 publications

(18 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we determine the previous contents required for forgetting gate in the process of graphic language representation and recognition, as well as the data memory to be retained in neurons in the optimized model. e specific expressions are shown in formulas ( 8) and (9).…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Graphic Language Representation in Visual Communication Design Based on Two-Way Long- and Short-Memory Model

Zhang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

With the popularity of neural network research, the application based on neural network model is gradually applied to all aspects of people’s life. Neural network model can not only solve the algebraic problems that traditional machine learning can solve but also recognize and analyze graphics through self-learning. For example, face recognition, web page recognition, product packaging design, and application are inseparable from the dissemination of graphic language. When these processes are realized through computer language, it is necessary to accurately identify these graphic languages. However, traditional machine language learning has poor performance in graphic language learning, which further leads to the application to achieve the purpose of the original visual communication design. Therefore, based on the neural network algorithm, this paper improves a new neural network model—two-way long- and short-memory model to make the computer recognize the graphic language more accurate and further explores the graphic language representation in the visual communication design based on the two-way long- and short-memory model.

show abstract

Section: Methodsmentioning

confidence: 99%

“…When this pattern is generally recognized, it can be seen through graphics that the purpose of guiding behavior becomes a visual effect. e capture and understanding of the visual effect by the final target object form a complete visual experience [9]. erefore, in visual communication design, we need to be extra cautious in the selection of graphics.…”

mentioning

confidence: 99%

Graphic Language Representation in Visual Communication Design Based on Two-Way Long- and Short-Memory Model

Zhang

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

show abstract

“…For example, in Reference [7], the authors proposed and used the CNN architecture to extract spatial information from sEMG signals and perform HGR classification. In addition to CNN-based architectures, some researches [33], [34] used Recurrent Neural Networks (RNNs) to extract the temporal features from the sEMG signals. RNNs are used because sEMG signals are sequential in the nature, and recurrent-based networks such as Long Short Term Memory (LSTM) can extract the patterns in a sequence of sEMG data treating HGR as a sequence modeling task.…”

Section: Related Workmentioning

confidence: 99%

TraHGR: Transformer for Hand Gesture Recognition via ElectroMyography

Zabihi¹,

Rahimian²,

Asif³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep learning-based Hand Gesture Recognition (HGR) via surface Electromyogram (sEMG) signals has recently shown considerable potentials for development of advanced myoelectric-controlled prosthesis. Although deep learning techniques can improve HGR accuracy compared to their classical counterparts, classifying hand movements based on sparse multichannel sEMG signals is still a challenging task. Furthermore, existing deep learning approaches, typically, include only one model as such can hardly maintain acceptable generalized performance in changing scenarios. In this paper, we aim to address this challenge by capitalizing on the recent advances of hybrid models and transformers. In other words, we propose a hybrid framework based on the transformer architecture, which is a relatively new and revolutionizing deep learning model. The proposed hybrid architecture, referred to as the Transformer forHand Gesture Recognition (TraHGR), consists of two parallel paths followed by a linear layer that acts as a fusion center to integrate the advantage of each module and provide robustness over different scenarios. We evaluated the proposed architecture TraHGR based on the commonly used second Ninapro dataset, referred to as the DB2. The sEMG signals in the DB2 dataset are measured in the real-life conditions from 40 healthy users, each performing 49 gestures. We have conducted extensive set of experiments to test and validate the proposed TraHGR architecture, and compare its achievable accuracy with more than five recently proposed HGR classification algorithms over the same dataset. W have also compared the results of the proposed TraHGR architecture with each individual path and demonstrated the distinguishing power of the proposed hybrid architecture. The recognition accuracies of the proposed TraHGR architecture are 86.18%, 88.91%, 81.44%, and 93.84%, which are 2.48%, 5.12%, 8.82%, and 4.30% higher than the state-ofthe-art performance for DB2 (49 gestures), DB2-B (17 gestures), DB2-C (23 gestures), and DB2-D (9 gestures), respectively.

show abstract

“…Simply speaking, it means that LSTM can have better performance in longer sequences compared with normal RNN. It is suitable for processing bioelectrical signals with time-series characteristics [64,65]. Further, it is also suitable for the time-series feature RMS we employ here.…”

Section: Grip Force Prediction Algorithm and Hand Rehabilitation Cont...mentioning

confidence: 99%

Study on Flexible sEMG Acquisition System and Its Application in Muscle Strength Evaluation and Hand Rehabilitation

Liu

Li²,

Zhang³

et al. 2022

Micromachines

View full text Add to dashboard Cite

Wearable devices based on surface electromyography (sEMG) to detect muscle activity can be used to assess muscle strength with the development of hand rehabilitation applications. However, conventional acquisition devices are usually complicated to operate and poorly comfortable for more medical and scientific application scenarios. Here, we report a flexible sEMG acquisition system that combines a graphene-based flexible electrode with a signal acquisition flexible printed circuit (FPC) board. Our system utilizes a polydimethylsiloxane (PDMS) substrate combined with graphene transfer technology to develop a flexible sEMG sensor. The single-lead sEMG acquisition system was designed and the FPC board was fabricated considering the requirements of flexible bending and twisting. We demonstrate the above design approach and extend this flexible sEMG acquisition system to applications for assessing muscle strength and hand rehabilitation training using a long- and short-term memory network training model trained to predict muscle strength, with 98.81% accuracy in the test set. The device exhibited good flexion and comfort characteristics. In general, the ability to accurately and imperceptibly monitor surface electromyography (EMG) signals is critical for medical professionals and patients.

show abstract

Temporal Dilation of Deep LSTM for Agile Decoding of sEMG: Application in Prediction of Upper-Limb Motor Intention in NeuroRobotics

Cited by 31 publications

References 49 publications

Graphic Language Representation in Visual Communication Design Based on Two-Way Long- and Short-Memory Model

Graphic Language Representation in Visual Communication Design Based on Two-Way Long- and Short-Memory Model

TraHGR: Transformer for Hand Gesture Recognition via ElectroMyography

Study on Flexible sEMG Acquisition System and Its Application in Muscle Strength Evaluation and Hand Rehabilitation

Contact Info

Product

Resources

About