A Trap Motion in Validating Muscle Activity Prediction from Musculoskeletal Model using EMG

A. D. Wibawa, N. Verdonschot, J.P.K. Halbertsma, J.G.M. Burgerhof, R.L. Diercks, G. J. Verkerke

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Musculoskeletal modeling nowadays is becoming the most common tool for studying and analyzing human motion. Besides its potential in predicting muscle activity and muscle force during active motion, musculoskeletal modeling can also calculate many important kinetic data that are difficult to measure in vivo, such as joint force or ligament force. This paper will validate muscle activity predicted by the model during a static motion like knee flexion motion (squat motion). In this experiment, knee flexion motion was performed by 5 healthy subjects and modeled by using Gait Lower Extremity model from AnyBody Modeling System (AMS). Eight lower limb muscle activity prediction from the model will be validated by 8 EMG electrodes that measured the same muscles during squat motion. Muscle activity pattern and the position of onset would be used as a key factor in this validation study. Pearson correlation coefficient will be used to compare the pattern of both graphs. Knee joint force prediction from the model will also be compared with the literature studies. The result showed that 3 muscles showed high correlation coefficient, while the other 4 muscles showed slightly medium and one showed low correlation. Time delay of muscle activation between the model and EMG was recorded from Vastus Medialis muscle (18.38 ms) and Vastus Lateralis (22.8 ms), with muscle activation from the model was late compared to EMG. In conclusion, this statistical study has shown some detail differences between EMG and muscle activity prediction from the model. Knee flexion motion can be used as a trap motion when validating muscle activity of a musculoskeletal model, because the model will activate muscle activity based on motion data of markers, while in knee-flexed position, there was no marker’s movement, but the EMG was highly active due to the posture of the subjects in maintaining the knee-flexed position. However, the knee compressive force prediction from the model has showed positive confirmation from the literatures.

Original languageEnglish
Pages (from-to)61-72
Number of pages12
JournalInternational Journal of Bio-Science and Bio-Technology
Volume8
Issue number6
DOIs
Publication statusPublished - 2016

Fingerprint

Muscle
Muscles
Knee
Quadriceps Muscle
Lower Extremity
Chemical activation
Validation Studies
Ligaments
Knee Joint
Posture
Gait
Time delay
Healthy Volunteers
Electrodes

Keywords

  • EMG
  • Inverse dynamics analysis
  • Knee flexion motion
  • Muscle activity validation
  • Musculoskeletal modeling

Cite this

@article{fb849fded7b94420a1f6fc9e7c7b3f32,
title = "A Trap Motion in Validating Muscle Activity Prediction from Musculoskeletal Model using EMG",
abstract = "Musculoskeletal modeling nowadays is becoming the most common tool for studying and analyzing human motion. Besides its potential in predicting muscle activity and muscle force during active motion, musculoskeletal modeling can also calculate many important kinetic data that are difficult to measure in vivo, such as joint force or ligament force. This paper will validate muscle activity predicted by the model during a static motion like knee flexion motion (squat motion). In this experiment, knee flexion motion was performed by 5 healthy subjects and modeled by using Gait Lower Extremity model from AnyBody Modeling System (AMS). Eight lower limb muscle activity prediction from the model will be validated by 8 EMG electrodes that measured the same muscles during squat motion. Muscle activity pattern and the position of onset would be used as a key factor in this validation study. Pearson correlation coefficient will be used to compare the pattern of both graphs. Knee joint force prediction from the model will also be compared with the literature studies. The result showed that 3 muscles showed high correlation coefficient, while the other 4 muscles showed slightly medium and one showed low correlation. Time delay of muscle activation between the model and EMG was recorded from Vastus Medialis muscle (18.38 ms) and Vastus Lateralis (22.8 ms), with muscle activation from the model was late compared to EMG. In conclusion, this statistical study has shown some detail differences between EMG and muscle activity prediction from the model. Knee flexion motion can be used as a trap motion when validating muscle activity of a musculoskeletal model, because the model will activate muscle activity based on motion data of markers, while in knee-flexed position, there was no marker’s movement, but the EMG was highly active due to the posture of the subjects in maintaining the knee-flexed position. However, the knee compressive force prediction from the model has showed positive confirmation from the literatures.",
keywords = "EMG, Inverse dynamics analysis, Knee flexion motion, Muscle activity validation, Musculoskeletal modeling",
author = "Wibawa, {A. D.} and N. Verdonschot and J.P.K. Halbertsma and J.G.M. Burgerhof and R.L. Diercks and Verkerke, {G. J.}",
year = "2016",
doi = "10.14257/ijbsbt.2016.8.6.07",
language = "English",
volume = "8",
pages = "61--72",
journal = "International Journal of Bio-Science and Bio-Technology",
issn = "2233-7849",
publisher = "Science and Engineering Research Support Society",
number = "6",

}

A Trap Motion in Validating Muscle Activity Prediction from Musculoskeletal Model using EMG. / Wibawa, A. D.; Verdonschot, N.; Halbertsma, J.P.K.; Burgerhof, J.G.M.; Diercks, R.L.; Verkerke, G. J.

In: International Journal of Bio-Science and Bio-Technology, Vol. 8, No. 6, 2016, p. 61-72.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - A Trap Motion in Validating Muscle Activity Prediction from Musculoskeletal Model using EMG

AU - Wibawa, A. D.

AU - Verdonschot, N.

AU - Halbertsma, J.P.K.

AU - Burgerhof, J.G.M.

AU - Diercks, R.L.

AU - Verkerke, G. J.

PY - 2016

Y1 - 2016

N2 - Musculoskeletal modeling nowadays is becoming the most common tool for studying and analyzing human motion. Besides its potential in predicting muscle activity and muscle force during active motion, musculoskeletal modeling can also calculate many important kinetic data that are difficult to measure in vivo, such as joint force or ligament force. This paper will validate muscle activity predicted by the model during a static motion like knee flexion motion (squat motion). In this experiment, knee flexion motion was performed by 5 healthy subjects and modeled by using Gait Lower Extremity model from AnyBody Modeling System (AMS). Eight lower limb muscle activity prediction from the model will be validated by 8 EMG electrodes that measured the same muscles during squat motion. Muscle activity pattern and the position of onset would be used as a key factor in this validation study. Pearson correlation coefficient will be used to compare the pattern of both graphs. Knee joint force prediction from the model will also be compared with the literature studies. The result showed that 3 muscles showed high correlation coefficient, while the other 4 muscles showed slightly medium and one showed low correlation. Time delay of muscle activation between the model and EMG was recorded from Vastus Medialis muscle (18.38 ms) and Vastus Lateralis (22.8 ms), with muscle activation from the model was late compared to EMG. In conclusion, this statistical study has shown some detail differences between EMG and muscle activity prediction from the model. Knee flexion motion can be used as a trap motion when validating muscle activity of a musculoskeletal model, because the model will activate muscle activity based on motion data of markers, while in knee-flexed position, there was no marker’s movement, but the EMG was highly active due to the posture of the subjects in maintaining the knee-flexed position. However, the knee compressive force prediction from the model has showed positive confirmation from the literatures.

AB - Musculoskeletal modeling nowadays is becoming the most common tool for studying and analyzing human motion. Besides its potential in predicting muscle activity and muscle force during active motion, musculoskeletal modeling can also calculate many important kinetic data that are difficult to measure in vivo, such as joint force or ligament force. This paper will validate muscle activity predicted by the model during a static motion like knee flexion motion (squat motion). In this experiment, knee flexion motion was performed by 5 healthy subjects and modeled by using Gait Lower Extremity model from AnyBody Modeling System (AMS). Eight lower limb muscle activity prediction from the model will be validated by 8 EMG electrodes that measured the same muscles during squat motion. Muscle activity pattern and the position of onset would be used as a key factor in this validation study. Pearson correlation coefficient will be used to compare the pattern of both graphs. Knee joint force prediction from the model will also be compared with the literature studies. The result showed that 3 muscles showed high correlation coefficient, while the other 4 muscles showed slightly medium and one showed low correlation. Time delay of muscle activation between the model and EMG was recorded from Vastus Medialis muscle (18.38 ms) and Vastus Lateralis (22.8 ms), with muscle activation from the model was late compared to EMG. In conclusion, this statistical study has shown some detail differences between EMG and muscle activity prediction from the model. Knee flexion motion can be used as a trap motion when validating muscle activity of a musculoskeletal model, because the model will activate muscle activity based on motion data of markers, while in knee-flexed position, there was no marker’s movement, but the EMG was highly active due to the posture of the subjects in maintaining the knee-flexed position. However, the knee compressive force prediction from the model has showed positive confirmation from the literatures.

KW - EMG

KW - Inverse dynamics analysis

KW - Knee flexion motion

KW - Muscle activity validation

KW - Musculoskeletal modeling

UR - http://www.scopus.com/inward/record.url?scp=85009110344&partnerID=8YFLogxK

U2 - 10.14257/ijbsbt.2016.8.6.07

DO - 10.14257/ijbsbt.2016.8.6.07

M3 - Article

VL - 8

SP - 61

EP - 72

JO - International Journal of Bio-Science and Bio-Technology

JF - International Journal of Bio-Science and Bio-Technology

SN - 2233-7849

IS - 6

ER -