Development of an image-based system for assessment of bone density, geometry, and micro-architecture suitable for human studies

Osteoporosis is highly associated with reduced bone mineral density (BMD), micro-structural deterioration, and enhanced fracture-risk. Bone micro-structural properties are strong determinants of bone strength and fracture-risk. And research shows that transverse trabeculae improve bone strength by arresting buckling of longitudinal trabeculae. The first aim of my PhD research was to develop and validate a new method of characterizing transverse and longitudinal trabeculae using CT and micro-CT scans of cadaveric ankle specimens and examine its application over human study data at in vivo imaging resolution.

Advanced in vivo imaging technology has been applied to hip site. A major challenge for hip volumetric images is the selection of standardized anatomic regions of interest (ROIs) for regional femur bone measurements. Therefore, my second aim was to develop and validate an active-shape-model (ASM) based method to automatically generate standardized anatomic ROIs that will adjust for subject-specific variations in femur geometry and shape. A novel geodesic-editor graphical-user-interface (GUI) has been designed and developed that allows drawing and manipulating lines and landmarks on arbitrary curved surfaces as compared with conventional GUIs. Also, a morphological smoothing algorithm has been developed to smooth image-based femur shape and reduce digital artifacts.

Computational failures can occur due to a wide range of unexpected conditions in human studies, such as missing of trabecular bone due to a higher threshold, leakage of bone segmentation, and incorrect ROI generation. My third aim was to develop a comprehensive manual quality control protocol for quantitative bone micro-structural imaging cascades for human studies.