A comparative analysis of parametric and patient-specific finite element (FE) modeling of spinal motion segments.
To develop patient-specific FE models of spinal motion segments using mesh-morphing methods applied to a parametric FE model. To compare strain and displacement patterns in parametric and morphed models for both healthy and metastatically involved vertebrae.
Parametric FE models may be limited in their ability to fully represent patient-specific geometries and material property distributions. Generation of multiple patient-specific FE models has been limited because of computational expense. Morphing methods have been successfully used to generate multiple specimen-specific FE models of caudal rat vertebrae.
FE models of a healthy and a metastatic T6–T8 spinal motion segment were analyzed with and without patient-specific material properties. Parametric and morphed models were compared using a landmark-based morphing algorithm.
Morphing of the parametric FE model and including patient-specific material properties both had a strong impact on magnitudes and patterns of vertebral strain and displacement.
Small but important geometric differences can be represented through morphing of parametric FE models. The mesh-morphing algorithm developed provides a rapid method for generating patient-specific FE models of spinal motion segments.
Finite element (FE) models representing a healthy and a metastatic spinal motion segment were generated from a previously developed parametric model. Patient-specific FE models were created based on the parametric models using a mesh-morphing algorithm. Successful morphing yielded notable differences in strain and displacement patterns in the healthy and metastatically involved vertebrae.
From the *Orthopaedic and Biomechanics Laboratory, Sunnybrook Health Sciences Centre, and †University of Toronto, Ontario.
Acknowledgment date: January 10, 2008. First revision date: March 10, 2008. Second revision date: March 13, 2008. Acceptance date: March 17, 2008.
The manuscript submitted does not contain information about medical device(s)/drug(s).
Federal and Foundation funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript.
Supported by Canadian Institutes for Health Research (68991) and Canadian Breast Cancer Research Alliance (14337).
Address correspondence and reprint requests to Cari Whyne, PhD, Orthopaedic and Biomechanics Lab, Sunnybrook Health Sciences Centre, 2075 Bayview Ave., Room UB-19, Toronto, Ontario, M4N 3M5 Canada; E-mail: firstname.lastname@example.org