| Objective: The purpose was to compare the quality of coronal multiplanarreformation images with that of direct transverse thin-section images using multislice CT (MSCT).To explore the essential and possibility of isotropic imaging in thorax,and to select the best combination of scaning parameters.Materials and Methods: MSCT was performed on a 16 SLICE helical scanner(SOMATOM sensation cardiac),Post-process was performed by Volume Wizard work station.Matrix 512X512,FOV 384mm. First procedure: scanning images.120 patients who did not have any history of pulmonanry disease in superior airway were choosen.They were divided into four groups at random, and each group with 30 patients . The scan parameters of each group were different. collimation = 0.75 mm, pitch=1; collimation = 0.75 mm, pitch =1.5; collimation =1.5 mm, pitch =1; collimation=1.5 mm, pitch =1.5. Second procedure: reconstruct imagesAfter scanning , 8 groups transverse images were reconstructed with following parameters:Group 1: collimation = 0.75 mm, pitch=l, increment = 0.75 mm.Group 2: collimation = 0.75 mm, pitch =1, increment = 0. 5 mmoGroup 3: collimation = 0.75 mm, pitch =1.5, increment = 0.75 mimGroup 4: collimation = 0.75 mm, pitch =1.5, increment = 0. 5 mm.Group 5: collimation =1.5 mm, pitch =1, increment =1.5 mm,Group 6: collimation =1.5 mm, pitch =1, increment = 1 mm.Group 7: collimation =1.5 mm, pitch =1.5, increment =1.5 mm?Group 8: collimation = 1.5 mm, pitch =1.5, increment = 1 mm,All transverse images are observed with lung windows (window width of 1200Hu and window level of -700Hu)Third procedure: coronal MPREach above-mentioned direct transverse images are reformatted into corresponding coronal multiplanar reformation (MPR) imagesCD 0.75 mm collimation group, choose reformation thickness as 0.8mm, reformation interval as 0.5 mm,1.5mm collimation group, choose reformation thickness as 2mm, reformation interval as 0.5 mm;(2) All images with the same image magnifying ratio Z=6;?All multiplanar refomation images are observed with standard lung window value (window width of 1200Hu and window level of — 600Hu) .Fourth procedure: using six anatomic structure which can present showing ability to evaluat and grade, in order to compare and analyze between different transverse images as well as image quality between transverse images and corresponding coronal reformatting imsges, six anatomic structure separately as following ? the interlobular septa ; (D the central bronchovascular bundles coursing vertically; (3) the central bronchovascular bundles coursing horizontally ; @ the peripheralbronchovascular bundles; ?the interlobar fissurres, and (6) the pulmonary parenchyma ?Choose three radiologists who have been engaged in medical image diagnosing work for more than 10 years , to grade all image quality independently with double blind method ? the scores are classifid as follws: 0 score show no anatomic structure in existence ; 1 score show there are some anatomic structure in existence ^ but they are not integrity or apparent metamorphose > distinctive blur in margin, stairstep artifact in evidence, there are false structure; 2 scor show anatomic structure in existence ^ integrity mostly, but blur in some brim, stairstep artifact form false structure gently; three score show anatomic structure display clearly , satisfy diagnosing require ? Put graded results apply to SPSS 11.0 statistics software for rank sum test and Ridit statistical analyze.Results: Results of statistics show:CD there are non significant variation image quality between 0.75mm collimation transverse images group and 1.5mm collimation group transverse images.(2) there are significant variation image quality between coronal MPR with same pitch ■> same reconstruction interval n diferrent collimation for ability showing anatomic structure in lung , the image quality of 0.75 mm collimation group is superior to 1.5 mm collimation group.(3) there are non significant variation image quality between of reconstruction interval 100% (0.75 mm) and 75% (0.5mm) coronal MPR with same collimation(0.75 mmX same pitch (1 \ differrent collimation for ability showing anatomic structure in lung.? there are significant variation image quality of coronal MPR with same collimation> same reconstruction interval (100%> 75%) for ability showing anatomic structure in lung , the image quality of pitch 1 group is superior to 1.5 pitchgroup o(5) there are significant variation image quality between transverse images and coronal image with same condition for ability showing the interlobar fissurres, coronal imagegroups is superior totransverse image groups at large, there are non significant variation between them for ability showing other 5 anatomic symbolWith the fast pace of advances in multidetector CT (MDCT), and the beginning clinical implementation of 16-row scanners, improvements in spatial resolution in the z-axis with near-isotropic imaging provide exquisite multiplanar reconstructions of thorax anatomy. The use of the CT scanner for thorax imaging is mainly influenced by the spatial and temporal resolution that can be achieved with the applied technologies and procedures. The data acquisition with 16 x 0.75 mm scan slice and 16X1.5mm slice. The introduction of multidetector helical computed tomography (CT) has revolutionized the field of thorax imaging and has created a new dimension of temporal and spatial resolution. The speed of multidetector CT can either be used to reduce the time needed to cover a given volume or to increase the resolution along the z-axis, thereby resulting in nearly isotropic voxelsjeading to an improvement in the diagnostic value of thorax CT.Conclusion: Both the fundamental clinical results manifest that isotropy can reach on visual effect with suitable scan parameter. One scan parameter combination is using collimation = 0.75 mm , pitch =1 , increment = 0.75 mm, The other parameter combination is using collimation = 0.75 mm , pitch =1 , increment = 5 mm.The image quality of coronal multiplanar reconstructions from isotropic voxel data obtained using 0.75-mm collimation, with or without overlapping reconstruction, is similar to that of direct transverse thin-section CT scans 0 there are non significant variation between coronal MPR images and transverse images with isotropy for showing lung structure ? For showing the interlobar fissurres, coronal image groups issuperior to transverse image groups at large.The best scan parameter combination is collimation 0.75 mm, increment 0.75mm pitch 1. |
