SurfMyoAiR: A Surface Electromyography-Based Framework for Airwriting Recognition

Tripathi, Ayush; Prathosh, A P; Muthukrishnan, Suriya Prakash; Kumar, Lalan

doi:10.1109/tim.2023.3248084

Cited by 13 publications

(10 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These approaches, however, fail to leverage the temporal nature inherent to dynamic inputs. In turn, these strategies tend to fall apart when recognizing composite gestures, like air-drawn letters [31] or sign language words [32] and are susceptible to inadvertent activations [33]. Fortunately, over the past decade, the machine learning field of time-series classificationthe classification of non-stationary time series of varying lengths-has evolved exponentially [34][35][36].…”

Section: Discrete Controlmentioning

confidence: 99%

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition

Eddy,

Campbell,

Bateman

et al. 2024

J. Neural Eng.

View full text Add to dashboard Cite

Discrete myoelectric control-based gesture recognition has recently gained interest as a possible input modality for many emerging ubiquitous computing applications. Unlike the continuous control commonly employed in powered prostheses, discrete systems seek to recognize the dynamic sequences associated with gestures to generate event-based inputs. More akin to those used in general-purpose human-computer interaction, these could include, for example, a flick of the wrist to dismiss a phone call or a double tap of the index finger and thumb to silence an alarm. Moelectric control systems have been shown to achieve near-perfect classification accuracy, but in highly constrained offline settings. Real-world, online systems are subject to ``confounding factors'' (i.e., factors that hinder the real-world robustness of myoelectric control that are not accounted for during typical offline analyses), which inevitably degrade system performance, limiting their practical use. Although these factors have been widely studied in continuous prosthesis control, there has been little exploration of their impacts on discrete myoelectric control systems for emerging applications and use cases. Correspondingly, this work examines, for the first time, three confounding factors and their effect on the robustness of discrete myoelectric control: (1) limb position variability, (2) cross-day use, and a newly identified confound faced by discrete systems (3) gesture elicitation speed. Results from four different discrete myoelectric control architectures: (1) Majority Vote LDA, (2) Dynamic Time Warping, (3) an LSTM network trained with Cross Entropy, and (4) an LSTM network trained with Contrastive Learning, show that classification accuracy is significantly degraded (p<0.05) as a result of each of these confounds. This work establishes that confounding factors are a critical barrier that must be addressed to enable the real-world adoption of discrete myoelectric control for robust and reliable gesture recognition.

show abstract

Section: Discrete Controlmentioning

confidence: 99%

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition

Eddy,

Campbell,

Bateman

et al. 2024

J. Neural Eng.

View full text Add to dashboard Cite

show abstract

“…As we all know, deep learning develops rapidly. In the last few years, some novel models have been developed and used to classify the movements of the hand and wrist joints, for example, CNN-BiLSTM (Nguyen-Trong et al, 2021;Tripathi et al, 2022) and Graph Convolutional Network (GCN; Lai et al, 2021;Yang et al, 2022). However, to the best of our knowledge, they have not been used in studies to classify ankle movements.…”

Section: Limitations and Future Directionsmentioning

confidence: 99%

“…The introduction of deep learning methods in motion intention recognition using sEMG signals has led to good classification results (Huang et al, 2019;Gautam et al, 2020;Lai et al, 2021;Nguyen-Trong et al, 2021;Tripathi et al, 2022;Yang et al, 2022). Regarding ankle movement classification, Chen et al ( 2019) used a cerebellar model neural network (CMNN) to classify two ankle movements (IV and EV).…”

Section: Introductionmentioning

confidence: 99%

“…They tested this model on three healthy subjects with an average accuracy of 97.55%, which is a great improvement over traditional machine learning classifiers. However, many previous studies mainly focused on improving the accuracy of classification rather than the real-time performance of online classification (Cheng et al, 2020;Gautam et al, 2020;Lai et al, 2021;Nguyen-Trong et al, 2021;Tripathi et al, 2022;Yang et al, 2022), which is extremely important for bilateral training. The original signals are often directly fed into the deep learning models (Cheng et al, 2020;Gautam et al, 2020;Lai et al, 2021;Nguyen-Trong et al, 2021;Tripathi et al, 2022;Yang et al, 2022); therefore, even a slightly large amount of data can seriously affect the real-time performance (Joshi et al, 2013).…”

Section: Introductionmentioning

confidence: 99%

“…However, many previous studies mainly focused on improving the accuracy of classification rather than the real-time performance of online classification (Cheng et al, 2020;Gautam et al, 2020;Lai et al, 2021;Nguyen-Trong et al, 2021;Tripathi et al, 2022;Yang et al, 2022), which is extremely important for bilateral training. The original signals are often directly fed into the deep learning models (Cheng et al, 2020;Gautam et al, 2020;Lai et al, 2021;Nguyen-Trong et al, 2021;Tripathi et al, 2022;Yang et al, 2022); therefore, even a slightly large amount of data can seriously affect the real-time performance (Joshi et al, 2013). Moreover, these deep learning models have a large number of parameters when raw sEMG signals are used, which leads to high computational costs, making their application difficult for occasions requiring real-time performance (Xia et al, 2020).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A CNN-LSTM model for six human ankle movements classification on different loads

Wang²,

Yang³

et al. 2023

Front. Hum. Neurosci.

View full text Add to dashboard Cite

This study aims to address three problems in current studies in decoding the ankle movement intention for robot-assisted bilateral rehabilitation using surface electromyogram (sEMG) signals: (1) only up to four ankle movements could be identified while six ankle movements should be classified to provide better training; (2) feeding the raw sEMG signals directly into the neural network leads to high computational cost; and (3) load variation has large influence on classification accuracy. To achieve this, a convolutional neural network (CNN)—long short-term memory (LSTM) model, a time-domain feature selection method of the sEMG, and a two-step method are proposed. For the first time, the Boruta algorithm is used to select time-domain features of sEMG. The selected features, rather than raw sEMG signals are fed into the CNN-LSTM model. Hence, the number of model’s parameters is reduced from 331,938 to 155,042, by half. Experiments are conducted to validate the proposed method. The results show that our method could classify six ankle movements with relatively good accuracy (95.73%). The accuracy of CNN-LSTM, CNN, and LSTM models with sEMG features as input are all higher than that of corresponding models with raw sEMG as input. The overall accuracy is improved from 73.23% to 93.50% using our two-step method for identifying the ankle movements with different loads. Our proposed CNN-LSTM model have the highest accuracy for ankle movements classification compared with CNN, LSTM, and Support Vector Machine (SVM).

show abstract

An efficient multi-modal sensors feature fusion approach for handwritten characters recognition using Shapley values and deep autoencoder

Singh,

Chaturvedi

2024

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

SurfMyoAiR: A Surface Electromyography-Based Framework for Airwriting Recognition

Cited by 13 publications

References 45 publications

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition

Understanding the influence of confounding factors in myoelectric control for discrete gesture recognition

A CNN-LSTM model for six human ankle movements classification on different loads

An efficient multi-modal sensors feature fusion approach for handwritten characters recognition using Shapley values and deep autoencoder

Contact Info

Product

Resources

About