VALIDATION OF A BONE MINERAL DENSITY CALIBRATION PROTOCOL FOR MICRO-COMPUTED TOMOGRAPHY

Micro-computed tomography (micro-CT) is widely used for in vitro studies to characterize bone structure at the resolution of 10-100 microns. However, a densitometric calibration protocol is necessary to convert the X-ray attenuation coefficient provided by micro-CT in bone mineral density (BMD). The lastest one has an important role to improve the accuracy of subject-specific finite element models. This work presents a simple calibration protocol based on the use of solid hydroxyapatite phantoms with the correction of the beam hardening effect. The method was validated in comparison to ashing measures of cortical and trabecular human bone. In addition, bone samples tissue mineral density (TMD) was calculated with two different methods. The correlation between ash density and BMD was linear both for cortical (R-2 = 0.91) and trabecular bone (R-2 = 0.91). The analysis stratified by tissue type versus the pooled analysis confirmed the validity of a common linear model for both types of tissue (R-2 = 0.99). Despite its simplicity, the correlation obtained in this work does not depend on the acquisition settings of the micro-CT. TMD was shown to be dependent on the tissue investigated, with values in the range of 1.15-1.21 mg/mm(3) for trabecular bone, and 1.19-1.29 mg/mm(3) for cortical bone. Results are of some interest for generating micro finite elements models.