Automated selection and entry of computed tomography data in finite element modelling of the human femur

Greening, Dawna Rae (1998) Automated selection and entry of computed tomography data in finite element modelling of the human femur. Masters thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (4MB)

Abstract

Researchers in the Biomedical Engineering Centre (BEC) at Memorial University of Newfoundland are studying the structural and biomechanical properties of the human femur in order to better model and test hip prostheses. The advent of computed tomography (CT) technology has allowed researchers to visualize and use 3D anatomical data for such studies. In an attempt to generate realistic and useful 3D computer models of the human femur, BEC researchers have been using geometric and densimetric data from CT imagery to construct finite element models. Initial construction methods required manual input and manipulation of the data. This proved tedious and labour intensive. -- The aim of this project was to reduce user involvement and develop a software package to automate the model building process. By simplifying the procedure, it is hoped that users will be able to build models quickly and easily with minimal training in modelling software. Software modules were developed using several languages and run on a personal computer networked to a UNIX system. The software reads and translates CT data, detects and tracks the inner and outer edges of the femoral cortical bone, and selects equidistant points along these edges which define the geometry. This geometric data. combined with densimetric data in the form of CT numbers, is used to generate input files for the ANSYS finite element modelling package, which generates a solid volume model of the femur used for further testing. -- With restricted access to specimens and CT scanning facilities, testing of the software was limited throughout the course of this project. In view of the small test base, the results must be interpreted with caution, but preliminary results are encouraging. Compared with manual procedures used prior to this project, user involvement and the time needed to generate a model are dramatically reduced. The models are smoother in appearance and are easier to mesh. Direct measurement comparisons between true edges, and manually and automatically selected model edges indicate that the automated models are equal, if not superior, to those generated using the manual technique.

Actions (login required)

View Item