ClearInfinity深度学习重建算法结合"双低"扫描技术在腹部CT血管成像中的临床应用价值  

Clinical value of ClearInfinity deep learning reconstruction algorithm combined with“double-low”scanning technology in abdominal CT angiography

作　　者：王赛 刘超[1] 梅莞翠 吕鸿泽 王冠 杨波[1] 陈文[1] WANG Sai;LIU Chao;MEI Wancui;LüHongze;WANG Guan;YANG Bo;CHEN Wen(Medical Imaging Center,Taihe Hospital Affiliated to Hubei University of Medicine,Shiyan,Hubei Province 442000,China;School of Biomedical Engineering,Hubei University of Medicine,Shiyan,Hubei Province 442000,China;CT Product Division of Neusoft Medical Systems Co.,Ltd.,Shenyang 110167,China)

机构地区：[1]湖北医药学院附属太和医院医学影像中心,湖北十堰442000 [2]湖北医药学院生物医学工程学院,湖北十堰442000 [3]东软医疗系统股份有限公司CT产品事业部,辽宁沈阳110167

出　　处：《实用放射学杂志》2025年第3期491-495,共5页Journal of Practical Radiology

基　　金：湖北省教育厅中青年人才项目(Q20222110)。

摘　　要：目的 探讨ClearInfinity深度学习重建算法在低千伏(kV)和低对比剂中对腹部CT血管成像(CTA)的图像质量及辐射剂量的影响.方法 选取100例行腹部CTA检查的患者,随机分为A、B 2组.A组:管电压70 kV,对比剂30~35 mL,根据重建算法分为A1和A2 2个亚组,A1组50%ClearInfinity,A2组50%ClearView迭代算法;B组:管电压100 kV,对比剂60~70 mL,50%ClearView.客观评价3组图像的腹主动脉、肝固有动脉、肠系膜上动脉、肾动脉和髂总动脉感兴趣区(ROI)的CT值和标准差(SD)值,计算信噪比(SNR)以及对比噪声比(CNR);主观评分由2名医师评估;分析 A、B 2组辐射剂量.结果 A组的容积CT剂量指数(CTDI_(vol))、剂量长度乘积(DLP)和有效剂量(ED)与B组相比显著降低(P<0.05);A1组和B组主观评分均高于 A2组(P<0.05),A1组和B组主观评分无差异(P>0.05);与A1组相比,A2组各血管的SNR、CNR均显著降低,B组腹主动脉和髂总动脉的CT值、髂总动脉和肠系膜上动脉的CNR显著增高、肾动脉的SNR显著降低(P<0.05).结论 ClearInfinity深度学习重建算法联合70 kV扫描技术在腹部CTA中可获得较好的图像质量,并有效降低患者辐射剂量和对比剂用量,具有较高的临床应用价值.Objective To investigate the effect of ClearInfinity deep learning reconstruction algorithm on image quality and radiation dose of abdominal computed tomography angiography(CTA)at low kV and low contrast medium.Methods One hundred patients who underwent abdominal CTA were selected and randomly divided into group A and group B.Group A:tube voltage 70 kV,contrast medium 30-35 mL,divided into A1 and A2 subgroups according to reconstruction algorithm,group A150%ClearInfinity,group A250%ClearView iterative algorithm;group B:tube voltage 100 kV,contrast medium 60-70 mL,50%ClearView.CT values and standard deviation(SD)values of region of interest(ROI)of abdominal aorta,proper hepatic artery,superior mesenteric artery,renal artery and common iliac artery were evaluated objectively,while signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR)were calculated;subjective scores were evaluated by two physicians;radiation doses of groups A and B were analyzed.Results Volume CT dose index(CTDI_(vol)),dose length product(DLP)and effective dose(ED)in group A were significantly lower than those in group B(P<0.05),subjective scores in group A1 and group B were higher than those in group A2(P<0.05),and there was no difference between group A1 and group B(P>0.05).Compared with group A1,SNR and CNR of all vessels in group A2 were significantly decreased.CT values of abdominal aorta and common iliac artery,CNR of common iliac artery and superior mesenteric artery in group B were significantly increased,SNR of renal artery was significantly decreased(P<0.05).Conclusion ClearInfinity deep learning reconstruction algorithm combined with 70 kV scanning technology can obtain better abdominal CTA image quality,and effectively reduce the radiation dose and contrast medium of patients,which has high clinical application value.

关 键 词：深度学习 图像质量 腹部 CT血管成像 

分 类 号：R814.42[医药卫生—影像医学与核医学] R816.5[医药卫生—放射医学] R814.43[医药卫生—临床医学]