Decoding Simultaneous Multi-DOF Wrist Movements From Raw EMG Signals Using a Convolutional Neural Network

Yang, Wei; Yang, Dapeng; Liu, Yu; Liu, Hong

doi:10.1109/thms.2019.2925191

Cited by 55 publications

(29 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Ameri et al proposed a CNN-based regression model which estimated wrist angles more accurately than support vector regression (SVR) and achieved better performances in the Fitts' law test [9]. Yang et al investigated data-augmentation methods for CNN, and observed that CNN outperformed SVR significantly in the decoding of wrist kinetics [10,11]. Moreover, CNN can also work as the deep feature extractor in the hybrid CNN-RNN (RNN denotes recurrent neural networks) scheme to further increase the estimation accuracy [12,13].…”

Section: Introductionmentioning

confidence: 99%

Inter-Subject Domain Adaptation for CNN-Based Wrist Kinematics Estimation Using sEMG

Bao

Zaidi

Xie

et al. 2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Recently, convolutional neural network (CNN) has been widely investigated to decode human intentions using surface Electromyography (sEMG) signals. However, a pre-trained CNN model usually suffers from severe degradation when testing on a new individual, and this is mainly due to domain shift where characteristics of training and testing sEMG data differ substantially. To enhance inter-subject performances of CNN in the wrist kinematics estimation, we propose a novel regression scheme for supervised domain adaptation (SDA), based on which domain shift effects can be effectively reduced. Specifically, a two-stream CNN with shared weights is established to exploit source and target sEMG data simultaneously, such that domaininvariant features can be extracted. To tune CNN weights, both regression losses and a domain discrepancy loss are employed, where the former enable supervised learning and the latter minimizes distribution divergences between two domains. In this study, eight healthy subjects were recruited to perform wrist flexion-extension movements. Experiment results illustrated that the proposed regression SDA outperformed fine-tuning, a stateof-the-art transfer learning method, in both single-single and multiple-single scenarios of kinematics estimation. Unlike finetuning which suffers from catastrophic forgetting, regression SDA can maintain much better performances in original domains, which boosts the model reusability among multiple subjects.

show abstract

Section: Introductionmentioning

confidence: 99%

Inter-Subject Domain Adaptation for CNN-Based Wrist Kinematics Estimation Using sEMG

Bao

Zaidi

Xie

et al. 2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…Jiang designs a proportional and simultaneous controlled prosthetic hand by a dedicated multi-layer perceptron networks [15] . Yang proposes a convolutional neural network structure to decode simultaneous movements with three DOFs [16] . However, most of the research focuses on supervised learning, that is, while collecting myoelectric signals, it has to collect hand motion signals by other sensors such as camera or data glove at the same time, which increases the difficulty of collecting training samples.…”

Section: Introductionmentioning

confidence: 99%

Real-time Continuous Hand Motion Myoelectric Decoding by Automated Data Labeling

Zeng

Chen

et al. 2019

Preprint

View full text Add to dashboard Cite

In this paper an automated data labeling (ADL) neural network was proposed to streamline dataset collecting for real-time predicting the continuous motion of hand and wrist, these gestures are only decoded from a surface electromyography (sEMG) array of eight channels. Unlike collecting both the bio-signals and hand motion signals as samples and labels in supervised learning, this algorithm only collects the unlabeled sEMG into an unsupervised neural network, in which the hand motion labels are auto-generated. The coefficient of determination (r2) for three DOFs, i.e. wrist flex/extension, wrist pro/supination, hand open/close, was 0.86, 0.89 and 0.87 respectively. The comparison between real motion labels and auto-generated labels shows that the latter has earlier response than former. The results of Fitts' law test indicate that ADL has capability of controlling multi-DOFs simultaneously even though the training set only contains sEMG data from single DOF gesture. Moreover, no more hand motion measurement needed which greatly helps upper-limb amputee imagine the gesture of residual limb to control a dexterous prosthesis.

show abstract

“…Since there are 20 hidden units in this layer, the dimension of deep features is 20. Compared with CNN architectures in many previous studies [15,16,21], the dimension of our last FC Block is smaller since we empirically found that a too large dimension might not be able to benefit the performances of CNN and CNN-LSTM significantly. TABLE IV illustrates the R 2 values of CNN and CNN-LSTM when using different number of hidden units in the 2 nd FC layer, which indicates that in our experiments 20 dimensions can be a good choice for both two models.…”

Section: G Comparison Of Deep Feature Dimensionsmentioning

confidence: 97%

“…For instance, Ameri et al investigated a CNNbased regression technique which outperformed a traditional SVR-based scheme in an online Fitts' law test [19]. Yang et al presented several data-augmentation approaches for CNN in decoding 3-DoF wrist movements [20], and verified that the proposed CNN structure outperformed SVR significantly when confounding factors were involved [21]. Although CNN is good at extracting spatial correlations of multi-channel sEMG signals, it inherently ignores the temporal information during continuous muscle contractions.…”

Section: Introductionmentioning

confidence: 99%

A CNN-LSTM Hybrid Model for Wrist Kinematics Estimation Using Surface Electromyography

Bao

Zaidi

Xie

et al. 2021

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

Convolutional neural network (CNN) has been widely exploited for simultaneous and proportional myoelectric control due to its capability of deriving informative, representative and transferable features from surface electromyography (sEMG). However, muscle contractions have strong temporal dependencies but conventional CNN can only exploit spatial correlations. Considering that long short-term memory neural network (LSTM) is able to capture long-term and non-linear dynamics of time-series data, in this paper we propose a CNN-LSTM hybrid model to fully explore the temporal-spatial information in sEMG. Firstly, CNN is utilized to extract deep features from sEMG spectrum, then these features are processed via LSTM-based sequence regression to estimate wrist kinematics. Six healthy participants are recruited for the participatory collection and motion analysis under various experimental setups. Estimation results in both intra-session and inter-session evaluations illustrate that CNN-LSTM significantly outperforms CNN, LSTM and several representative machine learning approaches, particularly when complex wrist movements are activated. Index Terms-sEMG, wrist kinematics estimation, deep learning, convolutional neural network, long short-term memory network, hybrid model.

show abstract

Decoding Simultaneous Multi-DOF Wrist Movements From Raw EMG Signals Using a Convolutional Neural Network

Cited by 55 publications

References 30 publications

Inter-Subject Domain Adaptation for CNN-Based Wrist Kinematics Estimation Using sEMG

Inter-Subject Domain Adaptation for CNN-Based Wrist Kinematics Estimation Using sEMG

Real-time Continuous Hand Motion Myoelectric Decoding by Automated Data Labeling

A CNN-LSTM Hybrid Model for Wrist Kinematics Estimation Using Surface Electromyography

Contact Info

Product

Resources

About