Abstract:

Objective To discuss the value of improving the quality of the image and the removal of hardening artifact for the usage of GE Discovery computed tomography(CT)750 HD 70 keV monochromatic image in the posterior cranial fossa imaging. Methods ①From June 2011 to October 2012, 87 patients were taken spectrum scan mode head CT scan in our hospital, 400 patients were taken conventional mixed energy head CT scan using GE Discovery CT750 HD, their image informations were used to create database respectively. ②From each database, 20 cases were extracted using random number method at random respectively. CT value and its standard deviation(SD) of 20 mm2 region of interest(ROI) were measured for different position of brain tissue inside posterior cranial fossa in the cases selected, including:midpoint of the bilateral internal auditory meatus, pons center and cerebellar white matter in the level with maximum 4th ventricle. Image quality was measured by the noise level(CT value standard deviation SD), and statistical analysis between the 2 groups was taken. In addition, from each database, 30 cases were extracted using random number method at random respectively. Then for the cases selected, score for the posterior fossa artifact was taken and a statistical analysis was made. Results All cases of spectrum scan mode 70 keV monochromatic image in our study can meet the diagnosis need of posterior cranial fossa. Median noise level of bilateral internal auditory meatus midpoint, pons center and cerebellar white matter in the level with maximum 4th ventricle are(unit:HU):70 keV monochromatic image group3.96(3.60, 4.24)、3.66(3.40, 3.92)、3.37(3.06, 3.65), mixed energy head CT group 5.52(5.16, 5.80)、4.52(4.15, 5.36)、4.26(3.95, 4.74), respectively. Compared to mixed energy head CT, noise index of different brain tissue position inside the posterior fossa is decreased significantly, there are significant statistical differences, all P<0.001. As to posterior cranial fossa artifact, median score of 70 keV monochromatic image(4, 3-4) is higher than the mixed energy CT image(3, 2-3), and there are significant statistical differences, P<0.001. Conclusions Spectrum scan mode 70 keV monochromatic image can meet the diagnosis need of posterior cranial fossa after initial studies. Compared with conventional mixed energy CT image, 70 keV monochromatic image can significantly reduce image noise of the posterior fossa brain parenchyma, improve image quality, and inhibit the artifact of posterior cranial fossa.

Key words: Computed tomography; Spectral scan; Posterior fossa

摘要：

目的 探讨宝石能谱计算机断层扫描（computed tomography，CT）70 keV单能量图像在后颅窝成像中提高图像质量及去除硬化伪影的价值。 方法 ①将2011年6月至2012年10月在首都医科大学附属北京天坛医院进行头部宝石能谱CT 70 keV单能量模式平扫的87例以及同期头部宝石能谱CT常规混合能量CT平扫400例患者影像学资料分别建立数据库。②从上述两个数据库中采用随机序数法分别抽取20例患者资料，测定双侧内听道中点、四脑室最大层面桥脑中心及小脑白质20 mm2感兴趣区CT值及其标准差，图像质量用噪声水平（CT值标准差）衡量，比较两组图像质量。另外，从上述两个数据库中采用随机序数法分别抽取30例患者资料，进行后颅窝伪影评分，比较两组评分差异。 结果 能谱扫描模式下70 keV单能量图像所测得两组双侧内听道中点、四脑室最大层面桥脑中心及小脑白质的噪声水平中位数（四分位数）分别为（单位：HU）：70 keV单能量图像3.96（3.60，4.24）、3.66（3.40，3.92）、3.37（3.06，3.65），混合能量图像5.52（5.16，5.80）、4.52（4.15，5.36）、4.26（3.95，4.74），与混合能量组比较，70 keV单能量组图像后颅窝脑组织不同位置噪声水平均降低，差异有显著性，P均<0.001。后颅窝伪影方面，70 keV单能量组图像评分高于混合能量组，两组中位数和四分位数分别为单能量组4（3，4）和混合能量组3（2，3），差异有显著性（P<0.001）。 结论 能谱扫描模式下70 keV单能量图像经初步研究显示，能够满足后颅窝诊断需要，并且与常规混合能量图像相比较，能够显著降低后颅窝脑实质图像噪声，提高后颅窝图像质量、抑制后颅窝伪影。

关键词: 计算机断层扫描; 能谱扫描; 后颅窝