Methods: A reference phantom is based on a base resin composed of urethane dimethacrylate and triethylene glycol dimethacrylate, hydroxyapatite (Hoya, Tokyo, Japan) for radiopaque filler, and poly (methyl methacrylate) for dispersant. The phantom has 2.0 mm thickness and calcium density of 0, 100, 200, 300, 400, 500, 600 and 740 mg/cm3. These calcium densities were determined from cancellous and cortical bone density and that contained in a conventional CT phantom for diagnosis of osteoporosis. The reference phantom was measured by micro CT (SMX-100CT, Shimadzu, Japan). The measurement condition of the tube voltage was 100 kV, voxel size was 0.0493 mm and FOV was 25.00 mm. After processing, the image was exported to DICOM file. The voxel value of the reference phantom was analyzed using the image analyzing software (Image J ver.1.44, National Institutes of Health, USA). A relationship between calcium density and voxel value was determined using regression analysis.
Results: Voxel values (x103) ±s.d. of the reference phantom containing calcium density of 0, 100, 200, 300, 400, 500, 600 and 740 mg/cm3 were 41.1±0.1, 42.8±0.2, 45.4±0.3, 46.4±0.4, 51.0±0.2, 53.8±0.3 ,55.6±0.2 and 60.1±0.2, respectively, and the regression line was y=26.3x+40202 (R2=0.99, P<0.001).
Conclusions: There was a significant linear relationship between the calcium density and the voxel value of a new reference phantom image using micro CT. This suggested that the phantom might be available to measure the accurate bone density.