McKnight Colin D, Watcharotone Kuanwong, Ibrahim Mohannad, Christodoulou Emmanuel, Baer Aaron H, Parmar Hemant A
Department of Radiology, University of Michigan, Taubman Center/B1/132 F, 1500 E Medical Center Drive, Ann Arbor, MI, 48105, USA.
Pediatr Radiol. 2014 Aug;44(8):997-1003. doi: 10.1007/s00247-014-2943-y. Epub 2014 Apr 3.
Over the last decade there has been escalating concern regarding the increasing radiation exposure stemming from CT exams, particularly in children. Adaptive statistical iterative reconstruction (ASIR) is a relatively new and promising tool to reduce radiation dose while preserving image quality. While encouraging results have been found in adult head and chest and body imaging, validation of this technique in pediatric population is limited.
The objective of our study was to retrospectively compare the image quality and radiation dose of pediatric head CT examinations obtained with ASIR compared to pediatric head CT examinations without ASIR in a large patient population.
Retrospective analysis was performed on 82 pediatric head CT examinations. This group included 33 pediatric head CT examinations obtained with ASIR and 49 pediatric head CT examinations without ASIR. Computed tomography dose index (CTDIvol) was recorded on all examinations. Quantitative analysis consisted of standardized measurement of attenuation and the standard deviation at the bilateral centrum semiovale and cerebellar white matter to evaluate objective noise. Qualitative analysis consisted of independent assessment by two radiologists in a blinded manner of gray-white differentiation, sharpness and overall diagnostic quality.
The average CTDIvol value of the ASIR group was 21.8 mGy (SD = 4.0) while the average CTDIvol for the non-ASIR group was 29.7 mGy (SD = 13.8), reflecting a statistically significant reduction in CTDIvol in the ASIR group (P < 0.01). There were statistically significant reductions in CTDI for the 3- to 12-year-old ASIR group as compared to the 3- to 12-year-old non-ASIR group (21.5 mGy vs. 30.0 mGy; P = 0.004) as well as statistically significant reductions in CTDI for the >12-year-old ASIR group as compared to the >12-year-old non-ASIR group (29.7 mGy vs. 49.9 mGy; P = 0.0002). Quantitative analysis revealed no significant difference in the homogeneity of variance in the ASIR group compared to the non-ASIR group. Radiologist assessment of gray-white differentiation, sharpness and overall diagnostic quality in ASIR examinations was not substantially different compared to non-ASIR examinations.
The use of ASIR in pediatric head CT examinations allows for a 28% CTDIvol reduction in the 3- to 12-year-old age group and a 48% reduction in the >12-year-old age group without substantially affecting image quality.
在过去十年中，人们对CT检查导致的辐射暴露增加日益关注，尤其是在儿童中。自适应统计迭代重建（ASIR）是一种相对新颖且有前景的工具，可在保持图像质量的同时降低辐射剂量。虽然在成人头部、胸部和身体成像中已发现令人鼓舞的结果，但该技术在儿科人群中的验证有限。
我们研究的目的是在大量患者群体中，回顾性比较采用ASIR的儿科头部CT检查与未采用ASIR的儿科头部CT检查的图像质量和辐射剂量。
对82例儿科头部CT检查进行回顾性分析。该组包括33例采用ASIR的儿科头部CT检查和49例未采用ASIR的儿科头部CT检查。记录所有检查的计算机断层扫描剂量指数（CTDIvol）。定量分析包括对双侧半卵圆中心和小脑白质的衰减及标准差进行标准化测量，以评估客观噪声。定性分析包括两名放射科医生以盲法独立评估灰白区分度、清晰度和整体诊断质量。
ASIR组的平均CTDIvol值为21.8 mGy（标准差=4.0），而非ASIR组的平均CTDIvol为29.7 mGy（标准差=13.8），这表明ASIR组的CTDIvol有统计学意义的降低（P<0.01）。与3至12岁未采用ASIR的组相比，3至12岁采用ASIR的组的CTDI有统计学意义的降低（21.5 mGy对30.0 mGy；P=0.004），并且与大于12岁未采用ASIR的组相比，大于12岁采用ASIR的组的CTDI也有统计学意义的降低（分别为29. mGy对49.9 mGy；P=0.0002）。定量分析显示，与未采用ASIR的组相比，ASIR组的方差同质性无显著差异。放射科医生对ASIR检查中灰白区分度、清晰度和整体诊断质量的评估与未采用ASIR检查相比没有实质性差异。
在儿科头部CT检查中使用ASIR可使3至12岁年龄组的CTDIvol降低28%，大于12岁年龄组降低48%，且不会对图像质量产生实质性影响。
