[Computed tomography colonography image assessment using colon phantom-effects of reconstructed slice thickness].

Computed tomography colonography (CTC) is a robust and reliable imaging test of the colon. Recent studies show good sensitivity for the identification of nonpolypoid (flat) lesions as well. The purpose of this study was to determine the accuracy and reproducibility of a volume-rendering (VR) and virtual gross pathology (VGP) technique for detecting a polypoid lesion phantom by varying slice thickness. The scan of a simulated house-made phantom was performed using a 16-slice CT scanner with varying combinations of tube voltage (120 kVp), effective exposure (100 mAs), detector configuration (16×0.75 mm), rotation time (0.75 s), helical pitch (0.688, 0.938, 1.066 and 1.188), reconstruction kernel (A, B and C), and section thickness/reconstruction interval (0.8/0.4, 1.0/0.5 and 1.5/0.75 mm). All image data were transferred to a three-dimensional workstation to assess multi-planar reformation (MPR), VR and VGP. Accuracy of volume measurement using the VR technique for quantitative analysis was compared using a paired t-test. Four radiological technologists also independently evaluated the visual score using the VGP technique for qualitative analysis, and their evaluations were compared using one-way analysis of variance with Fisher's protected least significant difference post-hoc test. There was a statistically significant difference in reproducibility between the three different slice thicknesses as to volume measurement and observer performance test (p<0.01 and p<0.05, respectively). Furthermore, the reproducibility improved when using thinner slices. In conclusion, VR and VGP techniques using a slice thickness of 0.8 mm made it possible to maintain accuracy and reproducibility when using CTC to detect polypoid lesions.