Surrogate modeling of deformable joint contact using artificial neural networks

Highlights • We develop a surrogate contact modeling method based on artificial neural networks. • The method uses special sampling techniques to gather input–output data points. • Method evaluation showed that it was 1000 times faster than a conventional method. • The method can remove a computatio...

Full description

Saved in:
Bibliographic Details
Published in:Medical engineering & physics 2015-09, Vol.37 (9), p.885-891
Main Authors: Eskinazi, Ilan, Fregly, Benjamin J
Format: Article
Language:English
Subjects:
Biomechanics
Elastic contact
Humans
Knee contact
Knee Joint - physiopathology
Knee Prosthesis
Meta-model
Models, Biological
Motion
Neural networks
Neural Networks (Computer)
Radiology
Response surface
Surrogate modeling
Tibiofemoral joint
Time Factors
Torque
Walking
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Highlights • We develop a surrogate contact modeling method based on artificial neural networks. • The method uses special sampling techniques to gather input–output data points. • Method evaluation showed that it was 1000 times faster than a conventional method. • The method can remove a computational bottleneck from musculoskeletal models.
ISSN:1350-4533
1873-4030
DOI:10.1016/j.medengphy.2015.06.006