Bone mineral density (BMD) measured by dual energy X-ray absorptiometry (DXA) is currently used to predict fracture risk and monitor progression of osteoporosis. However, reliable prediction of fractures for individuals has been hampered by the significant overlap in BMD between people with and without osteoporotic fractures. HRpQCT and μMRI are two new clinical imaging modalities that are capable of assessing bone microstructure and are believed to have great potential in clinical research of osteoporosis. The objectives of this thesis are to develop, validate, and apply the high resolution imaging based analysis techniques that can provide quantitative assessment of bone quality.

Compared with the gold standard μCT, μMRI and HR-pQCT have limited resolution and lower signal-to-noise ratio. However, the implications of such limitations on the imaging based micro structural and mechanical measurements were not clear. To validate μMRI and HR-pQCT based measurements, ex vivo μMRI and HR-pQCT based measurements from the clinical scan protocol were compared with the corresponding measurements obtained from the registered oCT images. It was found that the absolute values of the two clinical imaging modalities based microstructural and mechanical measurements differ significantly from the gold standard. However, most measurements from the clinical imaging modalities correlate significantly with their corresponding gold standards. These data lend added credence to the microstructural and mechanical measurements derived from in vivo μMR and HR-pQCT images and ensure the suitability of these uMRI and HR-pQCT measurements for clinical assessment of bone quality. The validated patient-specific uMRI based mechanical and microstructural analyses were then applied to longitudinal studies for eugonadal and hypogonadal men before and after testosterone treatment. Differences in estimated elastic moduli and morphological parameters were detected between the eugonadal and hypogonadal groups, and improvements were found upon treatment in the hypogonadal group. It is further noted that after 24 months of treatment the restoration in anisotropic elastic properties was not uniform among different directions. This observation suggest that the effect of testosterone treatment on mechanical properties is anisotropic, in that the estimated elastic modulus of tibial trabecular bone with higher initial loss in testosterone deficiency is less recoverable with treatment.

The validated patient-specific HR-pQCT based mechanical and microstructural analyses were applied to study the underlying mechanisms of idiopathic osteoporosis (IOP) patients in young women. Significant differences in anisotropic elastic material constants of the trabecular volume of interest as well as in the stiffness of the whole bone segment were detected between the IOP patients and control subjects. Significant differences were also detected in most of the microstructural parameters between the two groups. Morphological analyses indicate differences of microstructural deteriorations and suggest that boss loss occurred in the radius of IOP patients by thinning of the trabecular elements and loss of the entire trabecular elements, while the tibial bone was weakened by thinning of the cortical shell and loss of the entire trabecular elements.