Development and Validation of an Image-Based Computed Trabecular Network Model

The aging process often results in decreased bone health in the lumbar spine which can lead to increased risk of vertebral injury. Osteoporosis most affects the density and distribution of trabecular (spongy) bone, but current models overlook the heterogeneous qualities of this region. Finite element (FE) modeling offers a noninvasive method to assess fracture risk using metrics of bone quality measured from quantitative computed tomography (qCT) scans. However, due to current limitations of CT scans, it can be difficult to accurately quantify bone degradation in the spine clinically. The primary objectives of this study are to develop a novel computed trabecular matrix for lumbar vertebrae trabeculae that will be used to extract additional information from CT scans. The computed trabecular matrix that results from this study could provide additional insight into the relationship between image data and bone strength. Bone morphology and bone volume fraction (BV/TV) will be measured in qCT scans of cadaveric lumbar spine. Forces from muscles will be applied at attachment points, and axial compressive force from body weight will be applied the superior endplate. The magnitude and direction of these forces will be estimated from methods using a previously established musculoskeletal model (C. Favier et al., 2021; Favier, McGregor, & Phillips, 2021). These metrics will be input to Optistruct (Altair Engineering, Troy, MI), which will then perform topology optimization to optimally distribute the load throughout the trabecular region to minimize bone stress while staying below the prescribed BV/TV. These vertebral models will be compressed until failure. To validate the biomechanical properties of the computed trabecular matrix, the same cadaveric vertebrae will be removed and compressed to failure using a servohydraulic uniaxial loading system. The novel computed trabecular matrix will then be used to model the effects of aging on bone quality degradation. Bone quality metrics of the lumbar spine will be obtained from 30 adults between 50-79 years of age who underwent baseline and follow-up CT scans 12-48 months apart. This research will employ an interdisciplinary approach by using radiology, biomechanics, and orthopaedics to study age-related bone decrement. The project will yield a novel computed trabecular matrix to extract additional clinically relevant data from image series, further adding value to CT scans in the fields of geriatrics and orthopaedics.

This research is comprised of two parts:

- The first part of this study involves development of a topology optimized trabecular network for lumbar vertebrae and validation of the model through compression testing on cadaveric samples. This model will be developed from qCT scans.

- The second part of this study will investigate possible correlations between image data and model performance to predict age-related changes of lumbar vertebral fracture risk in the aging population.