OBJECTIVES: The aim of this in vitro study was to investigate the accuracy of fractal analysis and morphometry for bone quality assessment as measured with dual energy X-ray absorptiometry (DXA). MATERIAL AND METHODS: Nineteen mandibular bone samples were used for the creation of artificial bone lesions (n=5) or decalcification (n=12) to simulate osteoporosis; two samples were used as controls. Cone beam computed tomography (CBCT) and DXA scans were made before and after processing the samples. The image data obtained from the CBCT scans were used to calculate the mean fractal dimension (FD), bone area and density (morphometric analysis) of the samples. Bone mineral density (BMD) was obtained from the DXA scans and set as a reference value for bone quality. The correlation between BMD and FD and between BMD and morphometric results were calculated. RESULTS: A significant correlation between FD and BMD (rho=+0.71 to +0.75; P<0.05) was observed. Bone area and BMD of the specimens (rho=+0.69 to +0.85; P<0.05) were also significantly related, in contrast to the density analysis, for which no significant correlation to BMD was found. CONCLUSIONS: The results of this study suggest that fractal analysis and bone area measurement have potential to evaluate bone quality on CBCT images, while density measurement does not seem to be valid.
OBJECTIVES: The aim of this in vitro study was to investigate the accuracy of fractal analysis and morphometry for bone quality assessment as measured with dual energy X-ray absorptiometry (DXA). MATERIAL AND METHODS: Nineteen mandibular bone samples were used for the creation of artificial bone lesions (n=5) or decalcification (n=12) to simulate osteoporosis; two samples were used as controls. Cone beam computed tomography (CBCT) and DXA scans were made before and after processing the samples. The image data obtained from the CBCT scans were used to calculate the mean fractal dimension (FD), bone area and density (morphometric analysis) of the samples. Bone mineral density (BMD) was obtained from the DXA scans and set as a reference value for bone quality. The correlation between BMD and FD and between BMD and morphometric results were calculated. RESULTS: A significant correlation between FD and BMD (rho=+0.71 to +0.75; P<0.05) was observed. Bone area and BMD of the specimens (rho=+0.69 to +0.85; P<0.05) were also significantly related, in contrast to the density analysis, for which no significant correlation to BMD was found. CONCLUSIONS: The results of this study suggest that fractal analysis and bone area measurement have potential to evaluate bone quality on CBCT images, while density measurement does not seem to be valid.
Authors: R Pauwels; O Nackaerts; N Bellaiche; H Stamatakis; K Tsiklakis; A Walker; H Bosmans; R Bogaerts; R Jacobs; K Horner Journal: Br J Radiol Date: 2013-01 Impact factor: 3.039