Non-invasive characterization of bone tissue and modeling of bone at micron and sub-micron levels

Bone is a composite material with a hierarchical structure, spanning from a nanoscale to a whole bone level. We focus on modeling bone at two structural length scales, namely a micron level and a sub-micron level, and perform a feasibility study of using ultrasound to detect bone growth in a limb regeneration project. At the sub-micron level, bone consists of a single lamella which is formed by a collagen fiber network with fibers aligned in a preferentially oriented direction. We model such a fibrous network computationally using finite element software Abaqus by representing fibers as Timoshenko beams. We generate a random arrangement of fibers by using a Poisson process. We investigate the effects of fiber orientation, void volume fraction and window size on the constitutive elastic response of such a beam network. We show that generally C1111 and C1212 values decrease with increasing window size, higher scattering angle, and higher void volume fraction. In addition, there is less scatter for larger window sizes, higher scattering angles and higher void volume fractions. Percolation threshold is also determined. The effect of dangling fibers on the volume fraction and the resulting constitutive response is also investigated. Three ways of estimating volume fraction of fibers are proposed. The results show that the volume fraction may be overestimated due to dangling fibers, resulting in inaccurate constitutive response. At micron level, bone consists of osteons and interstitial lamella which are made of lamellae. We propose using classical laminate theory, namely Sun & Li’s three dimensional laminate model to model bone at micron level. The outputs are compared with nanoindentation results, obtained from literature, and other theoretical models. Overall, our modeling results are in the good agreement with experimental results reported in literature and predictions from other analytical models. The second part of my M.S. dissertation research focuses on the non-invasive characterization of bone using ultrasound. It involves imaging frog limbs with removed large portion of long bone in a tarsus to detect and assess the quality of regenerated bone tissue. In order to determine the regions of bone, cartilage and soft tissue, a specially written program which gives the backscatter coefficient (BSC) at a certain frequency is used here. The result shows that ultrasound hold promising potential to identify bone region and therefore it can be applied to capture growth of bone tissue in-vivo. In addition to ultrasound, the feasibility of applying optical coherence tomography (OCT) to detect bone growth in frog limb is performed. Due to the insufficient depth penetration, the existence of bone cannot be confirmed using this technique.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Mechanical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2010

Author dc:creator

• Lee, Yikhan

Contributors dc:contributor

• Jasiuk, Iwona M.

• fiber network
• osteon
• interstitial lamella
• modeling
• ultrasound

Statement dc:rights

• © 2010 Yikhan Lee

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/15992

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/15992