双能CT血管成像能谱纯化联合迭代重建技术对颅内动脉瘤图像质量和辐射剂量的影响

Influence of selective photon shield combined with sinogram affirmed iterative reconstruction on image quality and radiation dose of dual-energy CT angiography for the diagnosis of intracranial aneurysms

目的 探讨能谱纯化技术（SPS）联合正弦图确认的迭代重组（SAFIRE）技术对双能量CTA诊断颅内动脉瘤图像质量和辐射剂量的影响.方法 回顾性分析临床诊断为自发性蛛网膜下腔出血,经三维DSA及手术证实为颅内动脉瘤,且行脑CTA检查的80例患者.将患者按照CTA扫描方案分为常规组及试验组,每组各40例.患者均行头部增强双能量模式扫描,且均采用SPS技术.常规组扫描参数：管电压分别为80 kV和附加0.5 mm锡板滤过的140 kV（sn140 kV）,管电流分别为300、150 mAs,原始数据行滤波反投影法（FBP）重组;试验组扫描参数：管电压分别为80 kV和sn140 kV,管电流分别为200、100 mAs,原始数据行SAFIRE重组.图像质量的评价,客观评价指标包括各部位（枕叶、大脑前动脉,大脑中动脉及大脑后动脉）平均CT值、背景噪声、信噪比（SNR）和对比噪声比（CNR）;并对图像质量进行主观评分.记录扫描辐射剂量.以DSA结果为金标准,计算试验组和常规组对颅内动脉瘤的检查符合率.常规组与试验组图像客观评价指标及辐射剂量的比较采用两独立样本t检验,图像主观评分的比较采用独立样本非参数等级资料检验.结果 常规组枕叶、大脑前动脉,大脑中动脉及大脑后动脉的噪声分别为（5.34±1.22）、（39.94±15.96）、（39.94±15.96）和（23.17±5.51）,试验组分别为（4.44±1.08）、（24.91±9.36）、（35.03±13.65）和（20.61±5.08）,差异均有统计学意义（t值分别为2.508、3.742、2.788和2.563,P均＜0.05）;常规组上述部位SNR分别为6.37±1.56、3.48±0.92、3.21±0.91和4.02±0.99,试验组分别为7.66±1.74、4.70±1.40、3.88±0.81和4.59±0.68,差异均有统计学意义（t值分别为2.504、3.248、2.067和1.029,P均＜0.05）.2组上述部位平均CT值和CNR的差异均无统计学意义（P均＞0.05）.常规组与试验组的图像质量评分分别为（4.86±0.07）和（4.67±0.21）分,差异无统计学意义（Z=-0.443,P=0.621）.常规组CTDIvol和ED分别为（20.61±1.07） mGy和（0.79±0.04） mSv,试验组分别为（13.59±1.27）mGy和（0.54±0.06） mSv,差异均有统计学意义（t值分别为24.039和15.813,P均＜0.01）.与三维DSA及手术结果比较,试验组均正确诊断,检出率为100.0％（29/29）,常规组诊断符合率为96.0％（25/26）.结论 SPS联合SAFIRE技术在双能量CTA诊断颅内动脉瘤中,能在保证图像质量的同时降低辐射剂量,且具有较高的诊断准确性.

Objective To investigate the influence of selective photon shield combined with sinogram affirmed iterative reconstruction（SAFIRE） on image quality and radiation dose of dual-energy CTA for the diagnosis of intracranial aneurysms.Methods The study retrospectively analyzed 80 patients with spontaneous subarachnoid hemorrhage,who were diangosed having intracranial aneurysms by three-dimensional DSA and surgery and underwent cerebral CTA examination.According to CTA scanning method,the patients were divided into conventional group and experimental group （n=40 for each）.All patients underwent head CTA in dual energy mode with selective photon shield （SPS）technology.The scanning protocols of conventional group were tube voltages 80 kV and sn 140 kV,tube currents 300 mAs and 150 mAs,filtered back projection （FBP） reconstruction.The scanning protocols of experimental group were tube voltages 80 kV,sn 140 kV,tube currents 200 mAs,100 mAs,SAFIRE reconstruction.Objective image quality evaluation indexes included the average CT value of all parts （occipital lobe,the anterior cerebral artery,middle cerebral artety and posterior cerebral artery）,the background noise,the signal-to-noise ratio （SNR） and contrast to noise ratio （CNR）,and subjective image quality was also assessed.Radiation dose was recorded.With 3D DSA or operation findings as gold standard,coincidence rate of intracranial aneurysm detection was calculated both in experimental group and conventional group.Objective image quality evaluation index and radiation dose both in conventional group and experimental group were compared with two independent two-sample t test,subjective image scores with non parameter test of independent samples.Results The noise of occipital lobe,the anterior cerebral artery,middle cerebral artery and posterior cerebral artery were （5.34 ± 1.22）,（39.94 ± 15.96）,（39.94± 15.96） and （23.17± 5.51） HU in conventional group,（4.44 ± 1.08）,（24.91 ± 9.36）,（35.03 ± 13.65） and （20.61 ± 5.08） in experimental group （the value of t test were 2.508,3.742,2.788 and 2.563,P＜0.05）.SNRs of above-mentioned locations were 6.37 ± 1.56,3.48 ±0.92,3.21 ±0.91 and 4.02±0.99 in conventional group,7.66±1.74,4.70±1.40,3.88±0.81 and 4.59±0.68 in experimental group （the value of t test were2.504,3.248,2.067 and 1.029,P＜0.05）.The average CT value and CNR of the abovelocations had no statistical difference in two groups （P＞0.05）.Image quality scores of the conventional group and experimental group were （4.86±0.07） and （4.67±0.21） without statistical difference （Z=-0.443,P=0.621）.CTDIvol and ED were （20.61 ±1.07） mGy and （0.79 ± 0.04） mSv in conventional group and （13.59 ± 1.27） mGy and （0.54 ± 0.06） mSv in experimental group （the value of t test were 24.039 and 15.813,P＜0.01）.With 3D DSA or operation findings as gold standard,intracranial aneurysm diagnosis coincidence of experimental group was 100.0％ （29/29）,while coincidence rate of conventional group was 96.0％ （25/26）.Conclusion SPS combined with SAFIRE techniques applied in dual energy CTA for the diagnosis of intracranial aneurysms could improve the quality of CT image and reduce the radiation dosage and have a high diagnosis accuracy.