An Evaluation of Hand-Force Prediction Using Artificial Neural-Network Regression Models of Surface EMG Signals for Handwear Devices

Masayuki Yokoyama, Ryohei Koyama, Masao Yanagisawa

    Research output: Contribution to journalArticle

    5 Citations (Scopus)

    Abstract

    Hand-force prediction is an important technology for hand-oriented user interface systems. Specifically, surface electromyography (sEMG) is a promising technique for hand-force prediction, which requires a sensor with a small design space and low hardware costs. In this study, we applied several artificial neural-network (ANN) regression models with different numbers of neurons and hidden layers and evaluated handgrip forces by using a dynamometer. A handwear with dry electrodes on the dorsal interosseous muscles was used for our evaluation. Eleven healthy subjects participated in our experiments. sEMG signals with six different levels of forces from 0 N to 200 N and maximum voluntary contraction (MVC) are measured to train and test our ANN regression models. We evaluated three different methods (intrasession, intrasubject, and intersubject evaluation), and our experimental results show a high correlation (0.840, 0.770, and 0.789 each) between the predicted forces and observed forces, which are normalized by the MVC for each subject. Our results also reveal that ANNs with deeper layers of up to four hidden layers show fewer errors in intrasession and intrasubject evaluations.

    Original languageEnglish
    Article number3980906
    JournalJournal of Sensors
    Volume2017
    DOIs
    Publication statusPublished - 2017 Jan 1

    Fingerprint

    Electromyography
    regression analysis
    Neural networks
    evaluation
    Dynamometers
    predictions
    electromyography
    User interfaces
    Neurons
    Muscle
    Hardware
    Electrodes
    contraction
    Sensors
    dynamometers
    Costs
    Experiments
    muscles
    neurons
    hardware

    ASJC Scopus subject areas

    • Control and Systems Engineering
    • Instrumentation
    • Electrical and Electronic Engineering

    Cite this

    An Evaluation of Hand-Force Prediction Using Artificial Neural-Network Regression Models of Surface EMG Signals for Handwear Devices. / Yokoyama, Masayuki; Koyama, Ryohei; Yanagisawa, Masao.

    In: Journal of Sensors, Vol. 2017, 3980906, 01.01.2017.

    Research output: Contribution to journalArticle

    @article{6b66dde20f3d42358ee6690adff5ec64,
    title = "An Evaluation of Hand-Force Prediction Using Artificial Neural-Network Regression Models of Surface EMG Signals for Handwear Devices",
    abstract = "Hand-force prediction is an important technology for hand-oriented user interface systems. Specifically, surface electromyography (sEMG) is a promising technique for hand-force prediction, which requires a sensor with a small design space and low hardware costs. In this study, we applied several artificial neural-network (ANN) regression models with different numbers of neurons and hidden layers and evaluated handgrip forces by using a dynamometer. A handwear with dry electrodes on the dorsal interosseous muscles was used for our evaluation. Eleven healthy subjects participated in our experiments. sEMG signals with six different levels of forces from 0 N to 200 N and maximum voluntary contraction (MVC) are measured to train and test our ANN regression models. We evaluated three different methods (intrasession, intrasubject, and intersubject evaluation), and our experimental results show a high correlation (0.840, 0.770, and 0.789 each) between the predicted forces and observed forces, which are normalized by the MVC for each subject. Our results also reveal that ANNs with deeper layers of up to four hidden layers show fewer errors in intrasession and intrasubject evaluations.",
    author = "Masayuki Yokoyama and Ryohei Koyama and Masao Yanagisawa",
    year = "2017",
    month = "1",
    day = "1",
    doi = "10.1155/2017/3980906",
    language = "English",
    volume = "2017",
    journal = "Journal of Sensors",
    issn = "1687-725X",
    publisher = "Hindawi Publishing Corporation",

    }

    TY - JOUR

    T1 - An Evaluation of Hand-Force Prediction Using Artificial Neural-Network Regression Models of Surface EMG Signals for Handwear Devices

    AU - Yokoyama, Masayuki

    AU - Koyama, Ryohei

    AU - Yanagisawa, Masao

    PY - 2017/1/1

    Y1 - 2017/1/1

    N2 - Hand-force prediction is an important technology for hand-oriented user interface systems. Specifically, surface electromyography (sEMG) is a promising technique for hand-force prediction, which requires a sensor with a small design space and low hardware costs. In this study, we applied several artificial neural-network (ANN) regression models with different numbers of neurons and hidden layers and evaluated handgrip forces by using a dynamometer. A handwear with dry electrodes on the dorsal interosseous muscles was used for our evaluation. Eleven healthy subjects participated in our experiments. sEMG signals with six different levels of forces from 0 N to 200 N and maximum voluntary contraction (MVC) are measured to train and test our ANN regression models. We evaluated three different methods (intrasession, intrasubject, and intersubject evaluation), and our experimental results show a high correlation (0.840, 0.770, and 0.789 each) between the predicted forces and observed forces, which are normalized by the MVC for each subject. Our results also reveal that ANNs with deeper layers of up to four hidden layers show fewer errors in intrasession and intrasubject evaluations.

    AB - Hand-force prediction is an important technology for hand-oriented user interface systems. Specifically, surface electromyography (sEMG) is a promising technique for hand-force prediction, which requires a sensor with a small design space and low hardware costs. In this study, we applied several artificial neural-network (ANN) regression models with different numbers of neurons and hidden layers and evaluated handgrip forces by using a dynamometer. A handwear with dry electrodes on the dorsal interosseous muscles was used for our evaluation. Eleven healthy subjects participated in our experiments. sEMG signals with six different levels of forces from 0 N to 200 N and maximum voluntary contraction (MVC) are measured to train and test our ANN regression models. We evaluated three different methods (intrasession, intrasubject, and intersubject evaluation), and our experimental results show a high correlation (0.840, 0.770, and 0.789 each) between the predicted forces and observed forces, which are normalized by the MVC for each subject. Our results also reveal that ANNs with deeper layers of up to four hidden layers show fewer errors in intrasession and intrasubject evaluations.

    UR - http://www.scopus.com/inward/record.url?scp=85042449353&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85042449353&partnerID=8YFLogxK

    U2 - 10.1155/2017/3980906

    DO - 10.1155/2017/3980906

    M3 - Article

    AN - SCOPUS:85042449353

    VL - 2017

    JO - Journal of Sensors

    JF - Journal of Sensors

    SN - 1687-725X

    M1 - 3980906

    ER -