You Jia, Dai Yue, Huang Ning, Li Jing-Jing, Cheng Li, Zhang Xiu-Li, Liu Qi, Liu Ying, Xu Kai
From the *Department of Radiology, Affiliated Hospital of Xuzhou Medical College, Xuzhou, China; and †GE Life Science, Beijing, China.
J Comput Assist Tomogr. 2015 Sep-Oct;39(5):774-80. doi: 10.1097/RCT.0000000000000260.
To assess radiation dose and image quality using Adaptive Statistical Iterative Reconstruction (ASIR) in craniocervical computed tomographic angiography and to further evaluate the impact of body mass index (BMI) on image quality.
A total of 178 consecutive patients (112 men, 66 women; age range, 25-79 years) were enrolled in this prospective study and randomly divided into 2 groups: group A (conventional group): 120 kV, filtered back-projection reconstruction, and group B (low-dose group): 100 kV, 40% ASIR reconstruction. Radiation dose and image quality between groups A and B were compared. Correlation coefficients were calculated between quantitative image quality measurement and BMI, and between radiation dose and BMI using Pearson correlation. Two experienced radiologists independently evaluated the image quality with 4-point scores, and interrater reliability was calculated using κ analysis.
The volume CT dose index, dose-length product, and effective dose of group B were lower than those of group A (each P < 0.01), with decreases of 23.99%, 25.15% and 25.47% respectively. Positive correlations existed between radiation dose and BMI for both groups A and B (each P < 0.01). Group B had lower image noise, higher attenuation, higher signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), and higher subjective score than did group A (each P < 0.01). Computed tomographic values had negative correlations with BMI for the head, neck, and shoulders in both groups A and B (each P < 0.01). Image noise, SNR, and CNR had no correlations with BMI for the head and neck in both groups (each P > 0.01). On the contrary, image noise had a positive correlation, and SNR and CNR had a negative correlations with BMI for the shoulders in group A (each P < 0.01), but in contrast, no such correlations were found in group B (each P > 0.01).
In craniocervical CTA, ASIR can improve the image quality and reduce radiation dose in patients. Furthermore, ASIR reduced the variances of image quality of different body sizes in the shoulders.
在颅颈计算机断层血管造影中使用自适应统计迭代重建（ASIR）评估辐射剂量和图像质量，并进一步评估体重指数（BMI）对图像质量的影响。
本前瞻性研究共纳入178例连续患者（112例男性，66例女性；年龄范围25 - 79岁），随机分为2组：A组（传统组）：120 kV，滤波反投影重建；B组（低剂量组）：100 kV，40% ASIR重建。比较A组和B组之间的辐射剂量和图像质量。使用Pearson相关性分析计算定量图像质量测量与BMI之间以及辐射剂量与BMI之间的相关系数。两位经验丰富的放射科医生独立使用4分制对图像质量进行评估，并使用κ分析计算评分者间信度。
B组的容积CT剂量指数、剂量长度乘积和有效剂量均低于A组（各P < 0.01），分别降低了23.99%、25.15%和25.47%。A组和B组的辐射剂量与BMI之间均存在正相关（各P < 0.01）。B组的图像噪声更低、衰减更高、信噪比（SNR）和对比噪声比（CNR）更高，主观评分也高于A组（各P < 0.01）。A组和B组头部、颈部和肩部的计算机断层值与BMI均呈负相关（各P < 0.01）。两组头部和颈部的图像噪声、SNR和CNR与BMI均无相关性（各P > 0.01）。相反，A组肩部的图像噪声与BMI呈正相关，SNR和CNR与BMI呈负相关（各P < 0.01），但相比之下，B组未发现此类相关性（各P > 0.01）。
在颅颈CTA中，ASIR可提高患者的图像质量并降低辐射剂量。此外，ASIR减少了肩部不同体型患者图像质量的差异。
