Probabilistic evaluation of the material properties of the in vivo subject-specific articular surface using a computational model

Kyoung Tak Kang, Sung Hwan Kim, Juhyun Son, Young Han Lee, Shinil Kim, Heoung Jae Chun

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

This article used probabilistic analysis to evaluate material properties of the in vivo subject-specific tibiofemoral (TF) joint model. Sensitivity analysis, based on a Monte Carlo (MC) method, was performed using a subject-specific finite element (FE) model generated from in vivo computed tomography (CT) and magnetic resonance imaging (MRI) data, subjected to two different loading conditions. Specifically, the effects of inherent uncertainty in ligament stiffness, horn attachment stiffness, and articular surface material properties were assessed using multifactorial global sensitivity analysis. The MRI images were taken before and after axial compression, and when the flexion condition had been maintained at up to 90 degree flexion in the subject-specific knee joint. The loading conditions of the probabilistic subject-specific FE model (axial compression and 90 degree flexion) were similar to the MRI acquisition setup. We were able to detect the influence of material parameters while maintaining the potential effect of parametric interactions. Throughout the in silico property optimization, a subject-specific FE model was used and less sensitive parameters were eliminated in the global sensitivity method. Soft tissue material properties were estimated using an optimization procedure that involved the minimization of the differences between the kinematics predicted by the subject-specific model and those obtained through in vivo subject-specific data. The results of this approach suggest that the articular surface mechanical properties could be found by using in vivo measurements, which clarifies the valuable tool for future subject-specific studies related to TF joint scaffolds, allografts and biologics.

Original languageEnglish
Pages (from-to)1390-1400
Number of pages11
JournalJournal of Biomedical Materials Research - Part B Applied Biomaterials
Volume105
Issue number6
DOIs
Publication statusPublished - 2017 Aug

Fingerprint

Materials properties
Magnetic resonance
Axial compression
Imaging techniques
Sensitivity analysis
Stiffness
Ligaments
Biological Products
Scaffolds
Tomography
Allografts
Kinematics
Monte Carlo methods
Tissue
Mechanical properties

All Science Journal Classification (ASJC) codes

  • Biomaterials
  • Biomedical Engineering

Cite this

@article{1d050934a0a64c388304d2517f584b09,
title = "Probabilistic evaluation of the material properties of the in vivo subject-specific articular surface using a computational model",
abstract = "This article used probabilistic analysis to evaluate material properties of the in vivo subject-specific tibiofemoral (TF) joint model. Sensitivity analysis, based on a Monte Carlo (MC) method, was performed using a subject-specific finite element (FE) model generated from in vivo computed tomography (CT) and magnetic resonance imaging (MRI) data, subjected to two different loading conditions. Specifically, the effects of inherent uncertainty in ligament stiffness, horn attachment stiffness, and articular surface material properties were assessed using multifactorial global sensitivity analysis. The MRI images were taken before and after axial compression, and when the flexion condition had been maintained at up to 90 degree flexion in the subject-specific knee joint. The loading conditions of the probabilistic subject-specific FE model (axial compression and 90 degree flexion) were similar to the MRI acquisition setup. We were able to detect the influence of material parameters while maintaining the potential effect of parametric interactions. Throughout the in silico property optimization, a subject-specific FE model was used and less sensitive parameters were eliminated in the global sensitivity method. Soft tissue material properties were estimated using an optimization procedure that involved the minimization of the differences between the kinematics predicted by the subject-specific model and those obtained through in vivo subject-specific data. The results of this approach suggest that the articular surface mechanical properties could be found by using in vivo measurements, which clarifies the valuable tool for future subject-specific studies related to TF joint scaffolds, allografts and biologics.",
author = "Kang, {Kyoung Tak} and Kim, {Sung Hwan} and Juhyun Son and Lee, {Young Han} and Shinil Kim and Chun, {Heoung Jae}",
year = "2017",
month = "8",
doi = "10.1002/jbm.b.33666",
language = "English",
volume = "105",
pages = "1390--1400",
journal = "Journal of Biomedical Materials Research - Part B Applied Biomaterials",
issn = "1552-4973",
number = "6",

}

Probabilistic evaluation of the material properties of the in vivo subject-specific articular surface using a computational model. / Kang, Kyoung Tak; Kim, Sung Hwan; Son, Juhyun; Lee, Young Han; Kim, Shinil; Chun, Heoung Jae.

In: Journal of Biomedical Materials Research - Part B Applied Biomaterials, Vol. 105, No. 6, 08.2017, p. 1390-1400.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Probabilistic evaluation of the material properties of the in vivo subject-specific articular surface using a computational model

AU - Kang, Kyoung Tak

AU - Kim, Sung Hwan

AU - Son, Juhyun

AU - Lee, Young Han

AU - Kim, Shinil

AU - Chun, Heoung Jae

PY - 2017/8

Y1 - 2017/8

N2 - This article used probabilistic analysis to evaluate material properties of the in vivo subject-specific tibiofemoral (TF) joint model. Sensitivity analysis, based on a Monte Carlo (MC) method, was performed using a subject-specific finite element (FE) model generated from in vivo computed tomography (CT) and magnetic resonance imaging (MRI) data, subjected to two different loading conditions. Specifically, the effects of inherent uncertainty in ligament stiffness, horn attachment stiffness, and articular surface material properties were assessed using multifactorial global sensitivity analysis. The MRI images were taken before and after axial compression, and when the flexion condition had been maintained at up to 90 degree flexion in the subject-specific knee joint. The loading conditions of the probabilistic subject-specific FE model (axial compression and 90 degree flexion) were similar to the MRI acquisition setup. We were able to detect the influence of material parameters while maintaining the potential effect of parametric interactions. Throughout the in silico property optimization, a subject-specific FE model was used and less sensitive parameters were eliminated in the global sensitivity method. Soft tissue material properties were estimated using an optimization procedure that involved the minimization of the differences between the kinematics predicted by the subject-specific model and those obtained through in vivo subject-specific data. The results of this approach suggest that the articular surface mechanical properties could be found by using in vivo measurements, which clarifies the valuable tool for future subject-specific studies related to TF joint scaffolds, allografts and biologics.

AB - This article used probabilistic analysis to evaluate material properties of the in vivo subject-specific tibiofemoral (TF) joint model. Sensitivity analysis, based on a Monte Carlo (MC) method, was performed using a subject-specific finite element (FE) model generated from in vivo computed tomography (CT) and magnetic resonance imaging (MRI) data, subjected to two different loading conditions. Specifically, the effects of inherent uncertainty in ligament stiffness, horn attachment stiffness, and articular surface material properties were assessed using multifactorial global sensitivity analysis. The MRI images were taken before and after axial compression, and when the flexion condition had been maintained at up to 90 degree flexion in the subject-specific knee joint. The loading conditions of the probabilistic subject-specific FE model (axial compression and 90 degree flexion) were similar to the MRI acquisition setup. We were able to detect the influence of material parameters while maintaining the potential effect of parametric interactions. Throughout the in silico property optimization, a subject-specific FE model was used and less sensitive parameters were eliminated in the global sensitivity method. Soft tissue material properties were estimated using an optimization procedure that involved the minimization of the differences between the kinematics predicted by the subject-specific model and those obtained through in vivo subject-specific data. The results of this approach suggest that the articular surface mechanical properties could be found by using in vivo measurements, which clarifies the valuable tool for future subject-specific studies related to TF joint scaffolds, allografts and biologics.

UR - http://www.scopus.com/inward/record.url?scp=84963624344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963624344&partnerID=8YFLogxK

U2 - 10.1002/jbm.b.33666

DO - 10.1002/jbm.b.33666

M3 - Article

C2 - 27080324

AN - SCOPUS:84963624344

VL - 105

SP - 1390

EP - 1400

JO - Journal of Biomedical Materials Research - Part B Applied Biomaterials

JF - Journal of Biomedical Materials Research - Part B Applied Biomaterials

SN - 1552-4973

IS - 6

ER -