OBJECTIVE: To investigate the improvement in postprocessing time and image quality of 3D negative-contrast CT cholangiography (nCTC) with minimum intensity projection (minIP). METHODS: Thirty-eight patients with suspected biliary obstruction who underwent portal phase helical CT were retrospectively studied. Three-dimensional nCTC was generated using 3D tools with manual cut function in three methods: one-slice, two-slice, and vari-slice cut slab, respectively. After adjusting the threshold level to about -20 HU, the erosion function with a value of one voxel was employed each time for further cleaning up hypodense interfering voxels such as fat and air that remained and the procedure was stopped before the biliary and pancreatic ducts were overeroded. RESULTS: Mean values of the total postprocessing time of the three manual cut methods for 3D nCTC were 57.8, 29.9, and 20.6 min, respectively. After the first erosion, interfering voxels were removed in more than half of the cases and they could be primarily eliminated following the second erosion. Meanwhile, the pancreaticobiliary ducts showed obvious erosion findings at the third erosion. CONCLUSIONS: The vari-slice manual cut method was shown to be the most timesaving postprocessing method and the image quality for 3D nCTC can be improved when one to two frequencies of erosion are applied.
OBJECTIVE: To investigate the improvement in postprocessing time and image quality of 3D negative-contrast CT cholangiography (nCTC) with minimum intensity projection (minIP). METHODS: Thirty-eight patients with suspected biliary obstruction who underwent portal phase helical CT were retrospectively studied. Three-dimensional nCTC was generated using 3D tools with manual cut function in three methods: one-slice, two-slice, and vari-slice cut slab, respectively. After adjusting the threshold level to about -20 HU, the erosion function with a value of one voxel was employed each time for further cleaning up hypodense interfering voxels such as fat and air that remained and the procedure was stopped before the biliary and pancreatic ducts were overeroded. RESULTS: Mean values of the total postprocessing time of the three manual cut methods for 3D nCTC were 57.8, 29.9, and 20.6 min, respectively. After the first erosion, interfering voxels were removed in more than half of the cases and they could be primarily eliminated following the second erosion. Meanwhile, the pancreaticobiliary ducts showed obvious erosion findings at the third erosion. CONCLUSIONS: The vari-slice manual cut method was shown to be the most timesaving postprocessing method and the image quality for 3D nCTC can be improved when one to two frequencies of erosion are applied.