Machine learning algorithms for predicting scapular kinematics

Kristen F. Nicholson, R. Tyler Richardson, Elizabeth A.Rapp van Roden, R. Garry Quinton, Kert F. Anzilotti, James G. Richards

Research output: Contribution to journalArticle

Abstract

The goal of this study was to develop and validate a non-invasive approach to estimate scapular kinematics in individual patients. We hypothesized that machine learning algorithms could be developed using motion capture data to accurately estimate dynamic scapula orientation based on measured humeral orientations and acromion process positions. The accuracy of the algorithms was evaluated against a gold standard of biplane fluoroscopy using a 2D to 3D fluoroscopy/model matching process. Individualized neural networks were developed for nine healthy adult shoulders. These models were used to predict scapulothoracic kinematics, and the predicted kinematics were compared to kinematics obtained using biplane fluoroscopy to determine the accuracy of the machine learning algorithms. Results showed correlations between predicted kinematics and validation kinematics. Estimated kinematics were within 10 of validation kinematics. We concluded that individualized machine learning algorithms show promise for providing accurate, non-invasive measurements of scapulothoracic kinematics.

Original languageEnglish (US)
Pages (from-to)39-45
Number of pages7
JournalMedical Engineering and Physics
Volume65
DOIs
StatePublished - Mar 2019

Fingerprint

Biomechanical Phenomena
Learning algorithms
Learning systems
Kinematics
Fluoroscopy
Acromion
Machine Learning
Scapula
Data acquisition
Neural networks

All Science Journal Classification (ASJC) codes

  • Biophysics
  • Biomedical Engineering

Cite this

Nicholson, K. F., Richardson, R. T., van Roden, E. A. R., Quinton, R. G., Anzilotti, K. F., & Richards, J. G. (2019). Machine learning algorithms for predicting scapular kinematics. Medical Engineering and Physics, 65, 39-45. https://doi.org/10.1016/j.medengphy.2019.01.005
Nicholson, Kristen F. ; Richardson, R. Tyler ; van Roden, Elizabeth A.Rapp ; Quinton, R. Garry ; Anzilotti, Kert F. ; Richards, James G. / Machine learning algorithms for predicting scapular kinematics. In: Medical Engineering and Physics. 2019 ; Vol. 65. pp. 39-45.
@article{2572bce8b51d420aaa90837ae47cbf74,
title = "Machine learning algorithms for predicting scapular kinematics",
abstract = "The goal of this study was to develop and validate a non-invasive approach to estimate scapular kinematics in individual patients. We hypothesized that machine learning algorithms could be developed using motion capture data to accurately estimate dynamic scapula orientation based on measured humeral orientations and acromion process positions. The accuracy of the algorithms was evaluated against a gold standard of biplane fluoroscopy using a 2D to 3D fluoroscopy/model matching process. Individualized neural networks were developed for nine healthy adult shoulders. These models were used to predict scapulothoracic kinematics, and the predicted kinematics were compared to kinematics obtained using biplane fluoroscopy to determine the accuracy of the machine learning algorithms. Results showed correlations between predicted kinematics and validation kinematics. Estimated kinematics were within 10 of validation kinematics. We concluded that individualized machine learning algorithms show promise for providing accurate, non-invasive measurements of scapulothoracic kinematics.",
author = "Nicholson, {Kristen F.} and Richardson, {R. Tyler} and {van Roden}, {Elizabeth A.Rapp} and Quinton, {R. Garry} and Anzilotti, {Kert F.} and Richards, {James G.}",
year = "2019",
month = "3",
doi = "10.1016/j.medengphy.2019.01.005",
language = "English (US)",
volume = "65",
pages = "39--45",
journal = "Medical Engineering and Physics",
issn = "1350-4533",
publisher = "Elsevier BV",

}

Nicholson, KF, Richardson, RT, van Roden, EAR, Quinton, RG, Anzilotti, KF & Richards, JG 2019, 'Machine learning algorithms for predicting scapular kinematics', Medical Engineering and Physics, vol. 65, pp. 39-45. https://doi.org/10.1016/j.medengphy.2019.01.005

Machine learning algorithms for predicting scapular kinematics. / Nicholson, Kristen F.; Richardson, R. Tyler; van Roden, Elizabeth A.Rapp; Quinton, R. Garry; Anzilotti, Kert F.; Richards, James G.

In: Medical Engineering and Physics, Vol. 65, 03.2019, p. 39-45.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Machine learning algorithms for predicting scapular kinematics

AU - Nicholson, Kristen F.

AU - Richardson, R. Tyler

AU - van Roden, Elizabeth A.Rapp

AU - Quinton, R. Garry

AU - Anzilotti, Kert F.

AU - Richards, James G.

PY - 2019/3

Y1 - 2019/3

N2 - The goal of this study was to develop and validate a non-invasive approach to estimate scapular kinematics in individual patients. We hypothesized that machine learning algorithms could be developed using motion capture data to accurately estimate dynamic scapula orientation based on measured humeral orientations and acromion process positions. The accuracy of the algorithms was evaluated against a gold standard of biplane fluoroscopy using a 2D to 3D fluoroscopy/model matching process. Individualized neural networks were developed for nine healthy adult shoulders. These models were used to predict scapulothoracic kinematics, and the predicted kinematics were compared to kinematics obtained using biplane fluoroscopy to determine the accuracy of the machine learning algorithms. Results showed correlations between predicted kinematics and validation kinematics. Estimated kinematics were within 10 of validation kinematics. We concluded that individualized machine learning algorithms show promise for providing accurate, non-invasive measurements of scapulothoracic kinematics.

AB - The goal of this study was to develop and validate a non-invasive approach to estimate scapular kinematics in individual patients. We hypothesized that machine learning algorithms could be developed using motion capture data to accurately estimate dynamic scapula orientation based on measured humeral orientations and acromion process positions. The accuracy of the algorithms was evaluated against a gold standard of biplane fluoroscopy using a 2D to 3D fluoroscopy/model matching process. Individualized neural networks were developed for nine healthy adult shoulders. These models were used to predict scapulothoracic kinematics, and the predicted kinematics were compared to kinematics obtained using biplane fluoroscopy to determine the accuracy of the machine learning algorithms. Results showed correlations between predicted kinematics and validation kinematics. Estimated kinematics were within 10 of validation kinematics. We concluded that individualized machine learning algorithms show promise for providing accurate, non-invasive measurements of scapulothoracic kinematics.

UR - http://www.scopus.com/inward/record.url?scp=85060942240&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060942240&partnerID=8YFLogxK

U2 - 10.1016/j.medengphy.2019.01.005

DO - 10.1016/j.medengphy.2019.01.005

M3 - Article

C2 - 30733173

AN - SCOPUS:85060942240

VL - 65

SP - 39

EP - 45

JO - Medical Engineering and Physics

JF - Medical Engineering and Physics

SN - 1350-4533

ER -