RESUMO
PURPOSE: To evaluate the effect of the slice thickness and the size of region of interest (ROI) on CT number using quantitative CT phantom MATERIALS AND METHODS: The phantom containing 150 mg/cc, 75 mg/cc and 0 mg/cc calcium hydroxyapatite was scanned with 1, 3, 5 and 10mm slice thicknesses by single energy quantitative computed tomography (QCT). CT numbers were measured on center position of the phantom. Shape of ROI was circular and sizes were 1, 3, 5, 11, 16, 21, 26 and 33mm2 . ANOVA and Tukey's multiple comparison method were performed for statistical compari-son of CT numbers according to different slice thicknesses. Coefficient of variation of CT number measured in each size of ROI was evaluated in same slice thickness. RESULTS: CT numbers had statistically significant difference according to slice thicknesses (p<0.05). As the slice thickness increased, CT number also increased. As the density of phantom became lower and the size of ROI became smaller, the coefficient of variation of CT number increased. When the size of ROI was more than 11mm2 in 1mm slice thickness, 5mm2 in 3mm slice thickness and 3mm2 in 5mm slice thickness, the coefficient of variation became consistent. In 10mm slice thickness, the size of ROI had little effect on the coefficient of variation. CONCLUSION: CT number had variation according to the slice thickness and the size of ROI although the object was homogeneous. The slice thickness and the size of ROI are critical factors in precision of the CT number measurements.
Assuntos
DurapatitaRESUMO
PURPOSE: To evaluate the effect of the slice thickness and the size of region of interest (ROI) on CT number using quantitative CT phantom MATERIALS AND METHODS: The phantom containing 150 mg/cc, 75 mg/cc and 0 mg/cc calcium hydroxyapatite was scanned with 1, 3, 5 and 10mm slice thicknesses by single energy quantitative computed tomography (QCT). CT numbers were measured on center position of the phantom. Shape of ROI was circular and sizes were 1, 3, 5, 11, 16, 21, 26 and 33mm2 . ANOVA and Tukey's multiple comparison method were performed for statistical compari-son of CT numbers according to different slice thicknesses. Coefficient of variation of CT number measured in each size of ROI was evaluated in same slice thickness. RESULTS: CT numbers had statistically significant difference according to slice thicknesses (p<0.05). As the slice thickness increased, CT number also increased. As the density of phantom became lower and the size of ROI became smaller, the coefficient of variation of CT number increased. When the size of ROI was more than 11mm2 in 1mm slice thickness, 5mm2 in 3mm slice thickness and 3mm2 in 5mm slice thickness, the coefficient of variation became consistent. In 10mm slice thickness, the size of ROI had little effect on the coefficient of variation. CONCLUSION: CT number had variation according to the slice thickness and the size of ROI although the object was homogeneous. The slice thickness and the size of ROI are critical factors in precision of the CT number measurements.