摘要
目的:探讨应用16层螺旋CT(MSCT)、采用适宜噪声指数的自动毫安调节技术获得质量稳定的新生儿头颅CT扫描图像,应用模拟噪声指数降低射线剂量并应用于临床的可行性。方法:选取100例临床需要进行头颅CT检查的新生儿患者,男61例,女39例,其中32例为早产儿,年龄3 h^1个月,平均7天,头围29~37 mm,平均33.6 mm。应用16层螺旋CT扫描仪,Axial方式扫描,管电压100 kV,管电流范围为10~150 mA,噪声指数3.0,层厚、层间距均为5 mm,球管旋转时间0.8 s,扫描野Ped Head,进行头颅CT扫描获取原始图像(A组);应用GE公司研发的噪声指数软件对A组图像进行模拟噪声处理,分别模拟扫描剂量减少20%(B组)、30%(C组)、40%(D组),每例患者选取鞍上池层面、基底节层面、半卵园中心层面3个层面,将对应的不同模拟噪声指数添加到A组图像上获取新的图像序列;分别对经不同噪声指数处理的图像测量脑灰白质的CT值及标准差,记录各幅图像的辐射剂量参数(mA);每组图像的CT值比较采用2χ检验,标准差比较采用t检验。由2名医师独立对图像进行评估,图像质量评定使用好、较好、一般、差4级评分制进行评价,评定为一般认为符合临床诊断要求,采用Kappa方法评价不同观察者评定结果间的一致。Pearson correlation法用于分析患儿头围和辐射剂量参数(mA)之间的统计学意义,P<0.05为差异具有统计学意义。结果:A、B、C、D 4组图像灰质CT值为25.12±0.6、25.13±0.7、25.17±0.6、25.11±0.8,白质CT值为12.97±0.3、12.99±0.4、13.02±0.4、13.01±0.4;A、B、C、D 4组图像标准差测量分别为2.71±0.10、2.98±0.12、3.14±0.13、3.37±0.12,差异无统计学意义(P>0.05);A、B、C、D 4组图像扫描剂量分别为(135.6±2.1)mAs、(108.8±2.0)mAs、(95.2±1.8)mAs、(81.4±1.9)mAs,差异有统计学意义(P<0.05);A、B、C、D 4组平均图像质量评分分别为3.63±0.21、3.26±0.20、2.92±0.26、2.52±0.23,所得图像均能满足临床诊断要求,2名医师的评估结果获得了中等一致性(Kappa值分别为0.503和0.576),患儿头围和辐射剂量参数有相关性(P<0.05)。结论:应用16层螺旋CT进行新生儿头颅CT自动毫安扫描,设定适宜的噪声指数在保证图像质量的基础上还能显著降低射线剂量,减少对新生儿的辐射。
Objective：To explore the scanning images of neonatal cranial CT with stable quality derived from 16-MSCT with appropriate noise index,auto-milliampere regulating technology and low ray dose by simulation noise index,and explore the feasibility in clinic.Methods：100 neonates receiving cranial CT scanning were selected,including 61 boys and 39 girls,32 neonates were premature infants,the range of their ages was from 3 hours to one month,7 days on average;the range of their head circumferences was from 29 mm to 37 mm,33.6 mm on average;16-MSCT scanner was used,Axial mode was adopted,the tube voltage was 100 kV,the range of tube current was from 10 mA to 150 mA,the noise index was 3.0,the slice thickness was 5 mm,the layer spacing was 5 mm;the time of bulb tube rotation was 0.8 seconds,the vision of scanning was Ped Head,original images were obtained by cranial scanning（group A）.Noise index software developed by GE company was used for simulation noise treatment,the ray doses were reduced by 20%（group B）,30%（group C） and 40%（group D）,three layers including suprasellar cistern layer,basal ganglia layer and semioval center layer were selected from all the cases,new image sequences were obtained by adding different simulation noise indexes to images in group A;CT values and standard deviations of ectocinerea and brain white matter after treatment by different simulation noise indexes were measured,the ray dose parameters（mA） of all the images were recorded;χ2 test was used to compare CT values and t test was used to compare standard deviation;2 doctors were selected to evaluate the images independently,the standards of evaluation included good,relatively good,general and poor;Kappa method was used to evaluate the coincidence of results from different doctors;Pearson correlation analysis method was used to analyze the correlation between head circumference of infants and ray dose parameters;P〈0.05 was the standard of significant difference.Results：The CT values of grey matter in group A,group B,group C and group D were（25.12±0.6）,（25.13±0.7）,（25.17±0.6） and（25.11±0.8）,respectively;the CT values of white matter in group A,group B,group C and group D were（12.97±0.3）,（12.99±0.4）,（13.02±0.4） and（13.01±0.4）,respectively;the standard deviations of images in group A,group B,group C and group D were（2.71±0.10）,（2.98±0.12）,（3.14±0.13） and（3.37±0.12）,respectively,there was no significant difference（P〉0.05）;the scanning doses of images in group A,group B,group C and group D were（135.6±2.1） mAs,（108.8±2.0） mAs,（95.2±1.8） mAs and（81.4±1.9） mAs,respectively,there was significant difference（P〈0.05）.The average scores of quality of images in group A,group B,group C and group D were（3.63±0.21）,（3.26±0.20）,（2.92±0.26） and（2.52±0.23）,respectively,all the images satisfied the demand of clinical diagnosis,the evaluation results of the two doctors had medium coincidence（the Kappa values were 0.503 and 0.576,respectively）,there was a correlation between head circumference of infants and ray dose parameters（P〈0.05）.Conclusion：Neonatal cranial CT automatic milliampere scanning of 16MSCT and appropriate noise index can reduce ray dose significantly and decrease the radiation to neonates on the basis of ensuring the quality of images.
出处
《中国妇幼保健》
CAS
北大核心
2010年第34期5033-5036,共4页
Maternal and Child Health Care of China
基金
广西壮族自治区卫生厅计划科研项目〔Z2008349〕
