Almohiy Hussain, Alasar Emadeldin Mohamed Mukhtar, Saade Charbel
Department of Radiological Sciences, College of Applied Medical Sciences, King Khalid University, Abha, Saudi Arabia.
Department of Radiological Sciences, College of Applied Medical Sciences, King Khalid University, Abha, Saudi Arabia; Department of Diagnostic Radiology, American University of Beirut Medical Center, Beirut, Lebanon.
J Med Imaging Radiat Sci. 2016 Sep;47(3):235-242. doi: 10.1016/j.jmir.2016.05.002. Epub 2016 Jun 30.
To evaluate the impact of patient centering and radiation dose during intracranial computed tomography (ICT) on quantitative and qualitative image quality.
A total of 500 consecutive patients who underwent ICT were retrospectively reviewed using a 128-slice CT scanner (Definition AS+, Siemens, Germany). Patients were subjected in equal numbers to one of two positioning protocols: group A, poorly centered; and group B involved accurate centering before imaging. Gray-white matter (GWM) conspicuity, contrast-to-noise ratio (CNR), and signal-to-noise ratio (SNR) in each group were calculated. Qualitative image quality in terms of GWM differentiation, distinctness of posterior fossa contents, and overall diagnostic acceptability were evaluated by 2 neuroradiologists. The dose length product, CNR, SNR, and noise were measured between each group and data generated were compared using Mann-Whitney U nonparametric statistics. Visual grading characteristic and Kappa analyses were performed.
The mean noise index was significantly lower in group B (2.61 ± 0.29) compared with A (2.66 ± 0.21; P < .02). The mean attenuation of GWM, SNR, and CNR in the frontal lobe (A, 1:0.77, 0.84, 8.70 ± 1.36; and B, 1:0.65, 0.85, 15.32 ± 1.21; P < .02), occipital lobe (A, 1:1.10, 1.18, 10.79±2.11; and B, 1:0.94, 0.64, 14.41 ± 3.09; P < .04), and cerebellum (A, 1:0.79, 0.90, 12.56 ± 4.08; and B, 1:0.82, 0.87, 14.07 ± 2.28; P < .04) were significantly higher in group B compared with A, while the globus pallidus, caudate nucleus, and optic track in the basal ganglia demonstrated no difference in each group (P > .05). Mean dose length product demonstrated no significance between each group (A, 1312.03 ± 133.92; B, 1298.11 ± 130.61). The qualitative analyses demonstrated significant increases in visual grading characteristic for each reader (P < .02) and interobserver agreement was significantly increased in protocol B (k = 0.81) compared with A (k = 0.62).
Correct patient centering increases the CNR and SNR in both GWM in the left and right hemispheres of the brain during ICT.
评估颅内计算机断层扫描（ICT）期间以患者为中心及辐射剂量对定量和定性图像质量的影响。
使用128层CT扫描仪（Definition AS +，西门子，德国）对500例连续接受ICT检查的患者进行回顾性研究。将患者平均分为两组，分别采用两种定位方案：A组，定位不佳；B组，成像前精确定位。计算每组的灰白质（GWM）清晰度、对比噪声比（CNR）和信噪比（SNR）。由2名神经放射科医生评估GWM区分度、后颅窝内容物清晰度及整体诊断可接受性方面的定性图像质量。测量每组之间的剂量长度乘积、CNR、SNR和噪声，并使用Mann-Whitney U非参数统计比较生成的数据。进行视觉分级特征和Kappa分析。
B组的平均噪声指数（2.61±0.29）显著低于A组（2.66±0.21；P <.02）。额叶（A组，1:0.77、0.84、8.70±1.36；B组，1:0.65、0.85、15.32±1.21；P <.02）、枕叶（A组，1:1.10、1.18、10.79±2.11；B组，1:0.94、0.64、14.41±3.09；P <.04）和小脑（A组，1:0.79、0.90、12.56±4.08；B组，1:0.82、0.87、14.07±2.28；P <.04）中，B组的GWM平均衰减、SNR和CNR显著高于A组，而基底节中的苍白球、尾状核和视束在每组中无差异（P>.05）。每组之间的平均剂量长度乘积无显著差异（A组，1312.03±133.92；B组，1298.11±130.61）。定性分析表明，每位读者的视觉分级特征显著增加（P <.02），与A组（κ = 0.62）相比，B组观察者间的一致性显著增加（κ = 0.81）。
ICT期间正确的患者定位可提高大脑左右半球GWM的CNR和SNR。
