Mechanical strains experienced at the tissue level are intimately related to the mechanical integrity of whole bones and their response to environmental and interventional stimulus. Techniques for measurement of trabecular strains in an entire cancellous bone volume have been developed for laboratory studies of excised bones that can be performed using high resolution imaging systems such as microcomputed tomography (CT). These methods involve comparing high-resolution images of bones taken under loaded and no-load conditions and, through advanced mathematical computations, calculation of tissue strains (digital volume correlation, DVC). Thus far, it has not been possible to apply DVC methodologies to human spines in vivo, due to issues with resolution, radiation exposure and the need for a safe, yet effective mechanical loading protocol within the clinical imaging modality. Modern digital tomosynthesis (DTS) systems have the characteristics needed to be able to perform strain mapping in vivo but this has never been tried. With the overall hypothesis that DTS-based strain mapping is feasible and informative, the following aims are proposed to rigorously optimize, validate and demonstrate utility in human spine through a set of in vitro and in vivo experiments:
Aim 1 : Identify the strain levels that can be measured with DTS-DVC under physiologically relevant load magnitudes by using human cadaveric vertebral bodies and comparison to CT based DVC. CT will serve as the gold standard for optimizing values of DVC analysis parameters for a DTS application, determining thresholds for admissible strain values to maximize measurement accuracy and precision and identifying strain components to be further validated in the later stages of the research.
Aim 2 : Determine the extent to which DTS-DVC performed under a clinically applicable loading protocol predict vertebral strength and energy to failure independent from bone density. Using in vitro destructive mechanical test results as gold standard outcome, this aim will determine the relative efficacy of strain components for improving prediction of mechanical failure in cadaveric vertebrae and define margins of error for these predictions.
Aim 3 : Determine the range of in vivo vertebral strains measured with DTS-DVC in a sample of human subjects with a vertebral deformity and those without. Testing the ability of DTS-DVC to discriminate between cases with a predictable strain outcome from those that are normal will provide in vivo proof of concept. Through comparison of in vivo and in vitro strains, it will be possible to estimate the error in predictive models to be tested in future clinical studies. Development of the DTS-DVC methodology through the proposed aims is expected to substantially improve understanding of etiologies of age- and disease-related bone and joint degeneration, assessment of fracture risk and assessment of efficacy of therapeutic and surgical interventions aiming to restore bone function.
This project is designed to develop a new method, combining biomechanical loading, tomosynthesis imaging and advanced computational image correlation techniques, for direct in vivo measurement of how mechanical strains are distributed within the structure of vertebral bone under load. This new method will enable direct assessment of the mechanical efficacy of bone structure and material, thereby improving our ability to predict those at increased risk of spine fractures and those who are responding or likely to respond well to surgical, exercise and therapeutic interventions. As such, the knowledge gained form this research may substantially change our clinical approach for preventing and treating osteoporosis, a disease with a large impact on the elderly population.
| Oravec, Daniel; Kim, Woong; Flynn, Michael J et al. (2018) The relationship of whole human vertebral body creep to geometric, microstructural, and material properties. J Biomech 73:92-98 |
| Yeni, Yener N; Kim, Woong; Oravec, Daniel et al. (2018) Assessment of vertebral wedge strength using cancellous textural properties derived from digital tomosynthesis and density properties from dual energy X-ray absorptiometry and high resolution computed tomography. J Biomech 79:191-197 |
