Noninvasive prediction of vertebral body compressive strength using finite element method and an image based technique

Abstract

Noninvasive prediction of the strength of vertebral bodies under various loading conditions is a valuable tool for the assessment of clinical fractures and related side effects. This paper presents an effective approach for the noninvasive prediction of human vertebral strength using a new nonlinear finite element model and an image based parameter based on the quantitative computed tomography (QCT). Three thoracolumbar vertebrae were excised, as the samples, from three cadavers with an average age of 42 years old. The samples were scanned using the QCT. Then, the segmentation technique was performed on each sectional image. The segmented images were directly converted into three dimensional finite element models for linear and nonlinear analyses. A new material model, which is more compatible with real mechanical behavior of trabecular bone, was implemented in our nonlinear model. A new more mechanically meaningful image based parameter (σu A)min was also used for the ultimate compressive strength prediction. Subsequently, the samples were destructively tested under uniaxial compression and their experimental ultimate compressive strengths were obtained. There was very good agreements between the results obtained with the FEM and that of the experiments. Our new implemented finite element model can predict the ultimate compressive strength of human vertebra better than other common methods. The new image based parameter presented was also well agreement with experimental results. © 2008 Springer-Verlag.