Click here to

Session: CT and CBCT: New Technologies, Algorithms, and Emerging Applications II [Return to Session]

Comparative Analysis of a Resolution-Limiting X-Ray System and a Micro-CT System for the Quantification of Bone Mineral Density in Rat Models

C Warioba*, EL Marshall, C Haxton, K Takahashi, University of Chicago, Chicago, IL


SU-F-201-6 (Sunday, 7/10/2022) 2:00 PM - 3:00 PM [Eastern Time (GMT-4)]

Room 201

Purpose: To assess if a resolution limiting x-ray (XRAD) system is a feasible substitute for micro-CT small animal imaging in comparing bone mineral density (BMD) values from both systems by defining a procedure for computation of BMD in Sprague-Dawley (SD) rat models.

Methods: One postmortem SD rat was imaged via XRAD (isotropic voxel size of 50 micrometers and via micro-CT (isotropic voxel size of 25 micrometers). Six calibration phantom inserts (pure water, 100% cortical bone (CB), 50% CB, 30% CB, liver and adipose) from a Gammex 467 CT density phantom were imaged under identical conditions. A linear fit was defined between the hydroxyapatite (HA) density of the calibration phantom inserts for CB (100%, 50%, and 30%) and liver (assumed 0% CB) and their image derived voxel values. Threshold voxel values for soft tissue and bone were identified via ROI selection within the rat images. Values exceeding two standard deviations from the soft tissue mean were labeled bone. Voxels within the rat images containing bone were segmented then mapped to HA density via linear fit. BMD was computed from total density of HA divided by total number of bone voxels.

Results: The linear fit between the HA density of the calibration phantom inserts and their voxel values (VV) was HA density = 0.3265(VV) – 7593 (micro-CT) and 0.4209(VV) + 26.697 (XRAD). Voxel values of 24994.8 and 102.2 were defined as lower bounds for bone classification, values exceeding threshold were included within BMD quantification. BMD values of the rat were 1696.5 mg/cc and 637.9 mg/cc from micro-CT and XRAD systems, respectively.

Conclusion: An effective workflow for BMD quantification using micro-CT and XRAD imaging systems was defined. The inadequacy of the XRAD system in quantifying BMD was demonstrated. Small animal imaging must be conducted on the micro-CT until XRAD data inconsistencies are rectified.

Funding Support, Disclosures, and Conflict of Interest: Parent grant: R03DE028395, Supplement: R03DE028395-02S1 && A grant from University of Chicago institute of translational medicine.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email