Use this URL to cite or link to this record in EThOS:
Title: A shape analysis approach to prediction of bone stiffness using FEXI
Author: Pisharody, Sandhya
ISNI:       0000 0001 3491 9216
Awarding Body: University of Hull
Current Institution: University of Hull
Date of Award: 2007
Availability of Full Text:
Access from EThOS:
Access from Institution:
The preferred method of assessing the risk of an osteoporosis related fracture is currently a measure of bone mineral density (BMD) by dual energy X-ray absorptiometry (DXA). However, other factors contribute to the overall risk of fracture, including anatomical geometry and the spatial distribution of bone. Finite element analysis can be performed in both two and three dimensions, and predicts the deformation or induced stress when a load is applied to a structure (such as a bone) of defined material composition and shape. The simulation of a mechanical compression test provides a measure of whole bone stiffness (N mm−1). A simulation system was developed to study the sensitivity of BMD, 3D and 2D finite element analysis to variations in geometric parameters of a virtual proximal femur model. This study demonstrated that 3D FE and 2D FE (FEXI) were significantly more sensitive to the anatomical shape and composition of the proximal femur than conventional BMD. The simulation approach helped to analyse and understand how variations in geometric parameters affect the stiffness and hence strength of a bone susceptible to osteoporotic fracture. Originally, the FEXI technique modelled the femur as a thin plate model of an assumed constant depth for finite element analysis (FEA). A better prediction of tissue depth across the bone, based on its geometry, was required to provide a more accurate model for FEA. A shape template was developed for the proximal femur to provide this information for the 3D FE analysis. Geometric morphometric techniques were used to procure and analyse shape information from a set of CT scans of excised human femora. Generalized Procrustes Analysis and Thin Plate Splines were employed to analyse the data and generate a shape template for the proximal femur. 2D Offset and Depth maps generated from the training set data were then combined to model the three-dimensional shape of the bone. The template was used to predict the three-dimensional bone shape from a 2D image of the proximal femur procured through a DXA scan. The error in the predicted 3D shape was measured as the difference in predicted and actual depths at each pixel. The mean error in predicted depths was found to be 1.7mm compared to an average bone depth of 34mm. 3D FEXI analysis on the predicted 3D bone along with 2D FEXI for a stance loading condition and BMD measurement were performed based on 2D radiographic projections of the CT scans and compared to bone stiffness results obtained from finite element analysis of the original 3D CT scans. 3D FEXI provided a significantly higher correlation (R2 = 0.85) with conventional CT derived 3D finite element analysis than achieved with both BMD (R2 = 0.52) and 2D FEXI (R2 = 0.44).
Supervisor: Phillips, Roger Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: Computer Science