Finger Movement Pattern Recognition from Surface EMG Signals Using Machine Learning Algorithms

Krishnan, Shravan; Akash, R.; Kumar, Dilip; Jain, Rishab; Rathai, Karthik Murali Madhavan; Patil, Shantanu

doi:10.1007/978-981-13-1477-3_7

Cited by 5 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the limitations of neuromuscular interfaces for the comfortable and reliable acquisition of EMG signals in this context remain. Existing commercial systems currently use metallic electrodes or electrodes made of polymeric films to interface with the neuromuscular system which are not flexible or breathable [6,7], and as a result, are not comfortable for continuous use [3], and may contribute to skin complications [8].…”

Section: Introductionmentioning

confidence: 99%

A Mass-Producible Washable Smart Garment with Embedded Textile EMG Electrodes for Control of Myoelectric Prostheses: A Pilot Study

Alizadeh-Meghrazi

Sidhu²,

Jain³

et al. 2022

Sensors

View full text Add to dashboard Cite

Electromyography (EMG) is the resulting electrical signal from muscle activity, commonly used as a proxy for users’ intent in voluntary control of prosthetic devices. EMG signals are recorded with gold standard Ag/AgCl gel electrodes, though there are limitations in continuous use applications, with potential skin irritations and discomfort. Alternative dry solid metallic electrodes also face long-term usability and comfort challenges due to their inflexible and non-breathable structures. This is critical when the anatomy of the targeted body region is variable (e.g., residual limbs of individuals with amputation), and conformal contact is essential. In this study, textile electrodes were developed, and their performance in recording EMG signals was compared to gel electrodes. Additionally, to assess the reusability and robustness of the textile electrodes, the effect of 30 consumer washes was investigated. Comparisons were made between the signal-to-noise ratio (SNR), with no statistically significant difference, and with the power spectral density (PSD), showing a high correlation. Subsequently, a fully textile sleeve was fabricated covering the forearm, with 14 textile electrodes. For three individuals, an artificial neural network model was trained, capturing the EMG of 7 distinct finger movements. The personalized models were then used to successfully control a myoelectric prosthetic hand.

show abstract

Section: Introductionmentioning

confidence: 99%

A Mass-Producible Washable Smart Garment with Embedded Textile EMG Electrodes for Control of Myoelectric Prostheses: A Pilot Study

Alizadeh-Meghrazi

Sidhu²,

Jain³

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Then, the root mean square (RMS; Kundu et al, 2018 ; Le Sant et al, 2019 ) is extracted as a feature of sEMG signals. Compared with other features, such as waveform length ( Phinyomark et al, 2009 ; Arief et al, 2015 ), and autoregressive model features ( Subasi, 2012 ; Krishnan et al, 2019 ), it has been verified that the RMS feature obtain the best result under different lengths of sampling moving window ( Luo et al, 2020 ).…”

Section: Methodsmentioning

confidence: 98%

A multimodal human-robot sign language interaction framework applied in social robots

Zhong

Wang

2023

Front. Neurosci.

View full text Add to dashboard Cite

Deaf-mutes face many difficulties in daily interactions with hearing people through spoken language. Sign language is an important way of expression and communication for deaf-mutes. Therefore, breaking the communication barrier between the deaf-mute and hearing communities is significant for facilitating their integration into society. To help them integrate into social life better, we propose a multimodal Chinese sign language (CSL) gesture interaction framework based on social robots. The CSL gesture information including both static and dynamic gestures is captured from two different modal sensors. A wearable Myo armband and a Leap Motion sensor are used to collect human arm surface electromyography (sEMG) signals and hand 3D vectors, respectively. Two modalities of gesture datasets are preprocessed and fused to improve the recognition accuracy and to reduce the processing time cost of the network before sending it to the classifier. Since the input datasets of the proposed framework are temporal sequence gestures, the long-short term memory recurrent neural network is used to classify these input sequences. Comparative experiments are performed on an NAO robot to test our method. Moreover, our method can effectively improve CSL gesture recognition accuracy, which has potential applications in a variety of gesture interaction scenarios not only in social robots.

show abstract

“…Designing PH to perform accurate hand movement is very difficult. 1 PH can be controlled using electroencephalogram (EEG) signals or electromyogram (EMG) signals. Retrieval of EEG signals is complex as it requires proper positioning of electrodes on the scalp.…”

Section: Introductionmentioning

confidence: 99%

Performance evaluation of pattern recognition networks using electromyography signal and time-domain features for the classification of hand gestures

Vasanthi

Jayasree

2020

Proc Inst Mech Eng H

View full text Add to dashboard Cite

The problem of classifying individual finger movements of one hand is focused in this article. The input electromyography signal is processed and eight time-domain features are extracted for classifying hand gestures. The classified finger movements are thumb, middle, index, little, ring, hand close, thumb index, thumb ring, thumb little and thumb middle and the hand grasps are palmar class, spherical class, hook class, cylindrical class, tip class and lateral class. Four state-of-the-art classifiers namely feed forward artificial neural network, cascaded feed forward artificial neural network, deep learning neural network and support vector machine are selected for this work to classify the finger movements and hand grasps using the extracted time-domain features. The experimental results show that the artificial neural network classifier is stabilized at 6 epochs for finger movement dataset and at 4 epochs for hand grasps dataset with low mean square error. However, the support vector machine classifier attains the maximum accuracy of 97.3077% for finger movement dataset and 98.875% for hand grasp dataset which is significantly greater than feed forward artificial neural network, cascaded feed forward artificial neural network and deep learning neural network classifiers.

show abstract

Finger Movement Pattern Recognition from Surface EMG Signals Using Machine Learning Algorithms

Cited by 5 publications

References 15 publications

A Mass-Producible Washable Smart Garment with Embedded Textile EMG Electrodes for Control of Myoelectric Prostheses: A Pilot Study

A Mass-Producible Washable Smart Garment with Embedded Textile EMG Electrodes for Control of Myoelectric Prostheses: A Pilot Study

A multimodal human-robot sign language interaction framework applied in social robots

Performance evaluation of pattern recognition networks using electromyography signal and time-domain features for the classification of hand gestures

Contact Info

Product

Resources

About