Ex vivo quantitative μCT imaging of bone composition

Galateia Kazakia, Andrew Burghardt, Sam Cheung, Sharmila Majumdar

Quality of bone tissue — in particular degree and distribution of mineralization — may play a role in bone mechanics undetected by bone mineral density (BMD) or architectural measures. High-resolution assessment of bone tissue mineral density (TMD) may therefore provide information critical to the understanding of bone biomechanics. High-resolution 3-D assessment of TMD has been demonstrated using synchrotron radiation microcomputed tomography (SRμCT); however, this imaging modality is relatively inaccessible due to the scarcity of synchrotron facilities. Conventional desktop μCT systems are widely available and have been used extensively to assess bone micro-architecture. However, the polychromatic source and cone-shaped beam geometry used in conventional μCT imaging complicate the assessment of TMD. To evaluate the use of conventional μCT images in mineralization studies, we evaluated μCT-based measurement of degree and distribution of bone mineralization in a quantitative, spatially-resolved manner. Specifically, μCT measures of bone mineral content (BMC) and TMD were compared to those performed by SRμCT. Trabecular structure measures were also compared.

Fifteen cylinders of trabecular bone were machined from human proximal femurs, vertebral bodies and proximal tibias. Cylinders were imaged on a polychromatic μCT system (μCT 40: Scanco Medical AG, Basserrsdorf, Switzerland) at an isotropic voxel size of 8 μm. Volumes were reconstructed using beam hardening correction algorithms. SRμCT imaging was performed at the National Synchrotron Light Source with an isotropic voxel size of 7.50 μm. From both sets of images, attenuation values were converted to HA concentration using a linear regression derived using a calibration phantom (see figure). BMC, TMD and structural measures were calculated.

In high volume fraction specimens, μCT analysis resulted in BMC and TMD values 16.7% and 15.0% lower, respectively, than SRμCT values. In low volume fraction specimens, μCT analysis resulted in BMC and TMD values 12.8% and 12.9% lower, respectively, than SRμCT values. High correlations were found between μCT and SRμCT BMC values (R2 = 0.97-1.00). μCT and SRμCT TMD values were well-correlated when volume fraction groups were
considered individually (R2 = 0.78-0.99). Spatially-resolved comparisons highlighted substantial geometric nonuniformity in the μCT data, which was reduced (but not eliminated) using a more powerful beam hardening correction and did not exist in the SRμCT data. Repeated measures ANOVA with Tukey testing revealed small (approximately 1%-2%) but significant differences between μCT and SRμCT data for all architecture parameters with the exception of trabecular separation. Regression statistics comparing μCT structural data to SRμCT data showed excellent correlations (R2 = 0.99-1.00).

Our results indicate that μCT mineralization measures are underestimated but well-correlated to SRμCT and gravimetric data, particularly when volume fraction groups are considered individually and beam hardening algorithms are applied. Structural measures calculated by conventional μCT are accurate and highly correlated to SRμCT, consistent with previously published results. For more details, please see “Assessment of Bone Mineralization by Conventional X-ray Microcomputed Tomography: Comparison with Synchrotron Radiation Microcomputed Tomography and Ash Measurements,” Kazakia et al., Medical Physics, July 2008.

Related Content

MQIR Research Directions