Publication Date

8-2016

Date of Final Oral Examination (Defense)

6-6-2016

Type of Culminating Activity

Thesis - Boise State University Access Only

Degree Title

Master of Science in Mechanical Engineering

Department

Mechanical and Biomechanical Engineering

Major Advisor

Trevor Lujan, Ph.D.

Advisor

John F. Gardner, Ph.D.

Advisor

David Estrada, Ph.D.

Abstract

The mechanical behavior of soft connective tissue is governed by a dense network of fibrillar proteins in the extracellular matrix. Characterization of this fibrous network requires the accurate extraction of descriptive structural parameters from imaging data, including fiber dispersion and mean fiber orientation. Multiscale analysis of the fibrillar organization can be used to predict the mechanical behavior of these soft tissues, however the specific scale that is most relevant to the soft tissue mechanics remains unclear. In this study, we compared the ability of micro- and macro-scale structural organization of collagen fiber networks to predict the mechanical behavior of bovine ligament.

Common methods to quantify fiber parameters include fast Fourier transforms (FFT) and structure tensors at the micro-scale and quantitative polarized light imaging (QPLI) at the macro-scale. However, information is limited on the accuracy of the micro-scale analysis methods. In this study, we compared the FFT and structure tensor methods using test images of fiber networks with varying topology. The FFT method with a band-pass filter was the most accurate, with an error of 0.71 ± 0.43 degrees in measuring mean fiber orientation and an error of 7.4 ± 3.0% in measuring fiber dispersion in the test images. The accuracy of the structure tensor method was approximately 4 times worse than the FFT band-pass method when measuring fiber dispersion. A free software application, FiberFit, was then developed that utilizes an FFT band-pass filter to fit fiber orientations to a semicircular von Mises distribution utilizing parameters: mean fiber orientation μ and fiber dispersion k. Macro- and micro-scale organization parameters of bovine ligament, analyzed with QPLI and FiberFit, revealed moderate correlations of k ( = 0.50) and μ ( = 0.25) parameters.

A fiber crimp model was used to determine which scale (micro- or macro-) was more appropriate for the prediction of ligament behavior. Two healthy bovine ligaments with varying collagen network organization -- the medial collateral sesamoid ligament (MCSL) and palmar annular ligament (PAL) -- were mechancially tested to characterize the variation in mechanics. Macro-scale collagen organization was quantified using QPLI, while micro-scale organization was imaged using confocal microscopy and quantified using the validated FiberFit software program (micro-scale field of view was about 0.1% of the macro-scale field of view). Specimen specific semicircular von Mises parameters, including fiber dispersion (k) and mean orientation (μ) were incorporated into the fiber crimp model. The model was fit to experimental data of the MCSL using a least-squares curve fitting algorithm and used to predict the behavior of the PAL. Micro-scale imaging results at revealed significantly different k values between the MCSL and PAL tissue, where k = 2.6 ± 0.2 for MCSL, k = 1.5 ± 0.4 for PAL using FiberFit. Macro-scale dispersion parameters were not different between groups (k = 5.1 ± 2.2 for MCSL and k = 1.6 ± 1.7 for PAL using QPLI). The macro-scale parameters from QPLI correlated better to the mechanical results for PAL, when compared to the micro-scale parameters from FFT analysis of confocal images (ϵ = 5.04 and ϵ = 9.38, respectively). Results indicate a discrepancy between the model and mechanical behavior, suggesting that the fiber organization alone does not account for the entirety of mechanical changes observed between the MCSL and PAL.

This model can be utilized as a tool to develop correlations between ligament structure and function. An investigation of other structural differences between the MCSL and PAL should be explored, to improve model predictions. Different microstructural imaging modalities should also be investigated to verify that macro-scale organization is a better predictor of the observed mechanical behavior. By developing a validated software application that facilitates the automated analysis of fiber organization and comparing multiscale fiber network imaging techniques, this study builds a framework for future work that can help advance a mechanistic understanding of collagen networks, clarifying the mechanobiology of soft tissue remodeling and repair.

Share

COinS