Intelligent Prediction of Human Lower Extremity Joint Moment: An Artificial Neural Network Approach

Baoping Xiong, Nianyin Zeng*, Han Li, Yuan Yang, Yurong Li, Meilan Huang, Wuxiang Shi, Min Du, Yudong Zhang

*Corresponding author for this work

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Human joint moment plays an important role in quantitative rehabilitation assessment and exoskeleton robot control. However, the existing moment prediction methods require kinematic and kinetic data of human body as input, and the measurement of them needs special equipment, which makes them unable to be used in an unconstrained environment. According to the situation, this paper develops a novel method where a small number of input variables selected by Elastic Net are used as the input of artificial neural network (ANN) to predict joint moments, which makes the prediction in daily life possible. The method is tested on the experimental data collected from eight healthy subjects that are running on a treadmill at a speed of 2, 3, 4, and 5 m/s, respectively. Taking the right lower limb's 10 electromyography (EMG) and 5 joints angle data as candidate variable sets, Elastic Net is used to obtain the variable coefficients of the right lower limb's four joint moments. The inputs of the ANN determined by the variable coefficients are used to train and predict the joint moments. Prediction accuracy is evaluated by using the normalized root-mean-square error (NRMSE %) and cross correlation coefficient ( >amp;rho>amp; ) between the predicted joint moment and multi-body dynamics moment. Results of our study suggest that the method can accurately predict joint moment (NRMSE < 7.89%, >amp;rho >0.9633$>amp; ) with only 5-6 EMG signals. In conclusion, this method can effectively reduce the input variables while keeping a certain precision, which makes the joint moment prediction simple and out of equipment limitation. This method may facilitate the researches on real-Time gait analysis and exoskeleton robot control in motor rehabilitation.

Original languageEnglish (US)
Article number8648489
Pages (from-to)29973-29980
Number of pages8
JournalIEEE Access
Volume7
DOIs
StatePublished - Jan 1 2019

Fingerprint

Neural networks
Electromyography
Patient rehabilitation
Robots
Gait analysis
Exercise equipment
Mean square error
Kinematics
Kinetics
Exoskeleton (Robotics)

Keywords

  • Joint moment prediction
  • artificial neural network
  • elastic net

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)

Cite this

Xiong, Baoping ; Zeng, Nianyin ; Li, Han ; Yang, Yuan ; Li, Yurong ; Huang, Meilan ; Shi, Wuxiang ; Du, Min ; Zhang, Yudong. / Intelligent Prediction of Human Lower Extremity Joint Moment : An Artificial Neural Network Approach. In: IEEE Access. 2019 ; Vol. 7. pp. 29973-29980.
@article{33684993605846f19e335c7eaf8805c1,
title = "Intelligent Prediction of Human Lower Extremity Joint Moment: An Artificial Neural Network Approach",
abstract = "Human joint moment plays an important role in quantitative rehabilitation assessment and exoskeleton robot control. However, the existing moment prediction methods require kinematic and kinetic data of human body as input, and the measurement of them needs special equipment, which makes them unable to be used in an unconstrained environment. According to the situation, this paper develops a novel method where a small number of input variables selected by Elastic Net are used as the input of artificial neural network (ANN) to predict joint moments, which makes the prediction in daily life possible. The method is tested on the experimental data collected from eight healthy subjects that are running on a treadmill at a speed of 2, 3, 4, and 5 m/s, respectively. Taking the right lower limb's 10 electromyography (EMG) and 5 joints angle data as candidate variable sets, Elastic Net is used to obtain the variable coefficients of the right lower limb's four joint moments. The inputs of the ANN determined by the variable coefficients are used to train and predict the joint moments. Prediction accuracy is evaluated by using the normalized root-mean-square error (NRMSE {\%}) and cross correlation coefficient ( >amp;rho>amp; ) between the predicted joint moment and multi-body dynamics moment. Results of our study suggest that the method can accurately predict joint moment (NRMSE < 7.89{\%}, >amp;rho >0.9633$>amp; ) with only 5-6 EMG signals. In conclusion, this method can effectively reduce the input variables while keeping a certain precision, which makes the joint moment prediction simple and out of equipment limitation. This method may facilitate the researches on real-Time gait analysis and exoskeleton robot control in motor rehabilitation.",
keywords = "Joint moment prediction, artificial neural network, elastic net",
author = "Baoping Xiong and Nianyin Zeng and Han Li and Yuan Yang and Yurong Li and Meilan Huang and Wuxiang Shi and Min Du and Yudong Zhang",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/ACCESS.2019.2900591",
language = "English (US)",
volume = "7",
pages = "29973--29980",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

Xiong, B, Zeng, N, Li, H, Yang, Y, Li, Y, Huang, M, Shi, W, Du, M & Zhang, Y 2019, 'Intelligent Prediction of Human Lower Extremity Joint Moment: An Artificial Neural Network Approach', IEEE Access, vol. 7, 8648489, pp. 29973-29980. https://doi.org/10.1109/ACCESS.2019.2900591

Intelligent Prediction of Human Lower Extremity Joint Moment : An Artificial Neural Network Approach. / Xiong, Baoping; Zeng, Nianyin; Li, Han; Yang, Yuan; Li, Yurong; Huang, Meilan; Shi, Wuxiang; Du, Min; Zhang, Yudong.

In: IEEE Access, Vol. 7, 8648489, 01.01.2019, p. 29973-29980.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Intelligent Prediction of Human Lower Extremity Joint Moment

T2 - An Artificial Neural Network Approach

AU - Xiong, Baoping

AU - Zeng, Nianyin

AU - Li, Han

AU - Yang, Yuan

AU - Li, Yurong

AU - Huang, Meilan

AU - Shi, Wuxiang

AU - Du, Min

AU - Zhang, Yudong

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Human joint moment plays an important role in quantitative rehabilitation assessment and exoskeleton robot control. However, the existing moment prediction methods require kinematic and kinetic data of human body as input, and the measurement of them needs special equipment, which makes them unable to be used in an unconstrained environment. According to the situation, this paper develops a novel method where a small number of input variables selected by Elastic Net are used as the input of artificial neural network (ANN) to predict joint moments, which makes the prediction in daily life possible. The method is tested on the experimental data collected from eight healthy subjects that are running on a treadmill at a speed of 2, 3, 4, and 5 m/s, respectively. Taking the right lower limb's 10 electromyography (EMG) and 5 joints angle data as candidate variable sets, Elastic Net is used to obtain the variable coefficients of the right lower limb's four joint moments. The inputs of the ANN determined by the variable coefficients are used to train and predict the joint moments. Prediction accuracy is evaluated by using the normalized root-mean-square error (NRMSE %) and cross correlation coefficient ( >amp;rho>amp; ) between the predicted joint moment and multi-body dynamics moment. Results of our study suggest that the method can accurately predict joint moment (NRMSE < 7.89%, >amp;rho >0.9633$>amp; ) with only 5-6 EMG signals. In conclusion, this method can effectively reduce the input variables while keeping a certain precision, which makes the joint moment prediction simple and out of equipment limitation. This method may facilitate the researches on real-Time gait analysis and exoskeleton robot control in motor rehabilitation.

AB - Human joint moment plays an important role in quantitative rehabilitation assessment and exoskeleton robot control. However, the existing moment prediction methods require kinematic and kinetic data of human body as input, and the measurement of them needs special equipment, which makes them unable to be used in an unconstrained environment. According to the situation, this paper develops a novel method where a small number of input variables selected by Elastic Net are used as the input of artificial neural network (ANN) to predict joint moments, which makes the prediction in daily life possible. The method is tested on the experimental data collected from eight healthy subjects that are running on a treadmill at a speed of 2, 3, 4, and 5 m/s, respectively. Taking the right lower limb's 10 electromyography (EMG) and 5 joints angle data as candidate variable sets, Elastic Net is used to obtain the variable coefficients of the right lower limb's four joint moments. The inputs of the ANN determined by the variable coefficients are used to train and predict the joint moments. Prediction accuracy is evaluated by using the normalized root-mean-square error (NRMSE %) and cross correlation coefficient ( >amp;rho>amp; ) between the predicted joint moment and multi-body dynamics moment. Results of our study suggest that the method can accurately predict joint moment (NRMSE < 7.89%, >amp;rho >0.9633$>amp; ) with only 5-6 EMG signals. In conclusion, this method can effectively reduce the input variables while keeping a certain precision, which makes the joint moment prediction simple and out of equipment limitation. This method may facilitate the researches on real-Time gait analysis and exoskeleton robot control in motor rehabilitation.

KW - Joint moment prediction

KW - artificial neural network

KW - elastic net

UR - http://www.scopus.com/inward/record.url?scp=85064595647&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064595647&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2019.2900591

DO - 10.1109/ACCESS.2019.2900591

M3 - Article

AN - SCOPUS:85064595647

VL - 7

SP - 29973

EP - 29980

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

M1 - 8648489

ER -