Probabilistic evaluation of predicted force sensitivity to muscle attachment and glenohumeral stability uncertainty

Jaclyn N. Chopp-Hurley, Joseph E. Langenderfer, Clark R. Dickerson

Research output: Contribution to journalArticlepeer-review

13 Scopus citations

Abstract

A major benefit of computational modeling in biomechanics research is its ability to estimate internal muscular demands given limited input information. However, several assumptions regarding model parameters and constraints may influence model outputs. This research evaluated the influence of model parameter variability, specifically muscle attachment locations and glenohumeral stability thresholds, on predicted rotator cuff muscle force during internal and external axial humeral rotation tasks. Additionally, relative sensitivity factors assessed which parameters were more contributory to output variability. Modest model parameter variation resulted in considerable variability in predicted force, with origin-insertion locations being particularly influential. Specifically, the scapula attachment site of the subscapularis muscle was important for modulating predicted force, with sensitivity factors ranging from α = 0.2 to 0.7 in a neutral position. The largest variability in predicted forces was present for the subscapularis muscle, with average differences of 33.0 ± 9.6% of normalized muscle force (1-99% CI), and a maximal difference of 51% in neutral exertions. Infraspinatus and supraspinatus muscles elicited maximal differences of 15.0 and 20.6%, respectively, between confidence limits. Overall, origin and insertion locations were most influential and thus incorporating geometric variation in the prediction of rotator cuff muscle forces may provide more representative population estimates.

Original languageEnglish
Pages (from-to)1867-1879
Number of pages13
JournalAnnals of Biomedical Engineering
Volume42
Issue number9
DOIs
StatePublished - Sep 2014

Keywords

  • Model parameters
  • Moment arm
  • Rotator cuff
  • Shoulder
  • Stochastic modeling

Fingerprint

Dive into the research topics of 'Probabilistic evaluation of predicted force sensitivity to muscle attachment and glenohumeral stability uncertainty'. Together they form a unique fingerprint.

Cite this