切换至 "中华医学电子期刊资源库"
高级检索
中华肝脏外科手术学电子杂志

中华肝脏外科手术学电子杂志 ›› 2022, Vol. 11 ›› Issue (05) : 469 -475. doi: 10.3877/cma.j.issn.2095-3232.2022.05.009

临床研究

深度学习算法结合三低技术在上腹部动脉CT血管造影中的应用
张可1, 张悦1, 蒋伟1, 郭月飞1, 郭焯欣1, 孟占鳌1,†()   
  1. 1. 510630　广州，中山大学附属第三医院放射科
  • 收稿日期:2022-05-11 出版日期:2022-10-10
  • 通信作者: 孟占鳌
  • 基金资助:
    广州市科技计划项目(202007030007)

Application of deep learning algorithm combined with "triple-low" technique in CT angiography of upper abdominal arteries

Ke Zhang1, Yue Zhang1, Wei Jiang1, Yuefei Guo1, Zhuoxin Guo1, Zhan'ao Meng1,†()   

  1. 1. Department of Radiology, the Third Affiliated Hospital of Sun Yat-sen University, Guangzhou 510630, China
  • Received:2022-05-11 Published:2022-10-10
  • Corresponding author: Zhan'ao Meng
全文 ( ) 下载PDF ( ) 导出引用
引用本文：

张可, 张悦, 蒋伟, 郭月飞, 郭焯欣, 孟占鳌. 深度学习算法结合三低技术在上腹部动脉CT血管造影中的应用[J]. 中华肝脏外科手术学电子杂志, 2022, 11(05): 469-475.

Ke Zhang, Yue Zhang, Wei Jiang, Yuefei Guo, Zhuoxin Guo, Zhan'ao Meng. Application of deep learning algorithm combined with "triple-low" technique in CT angiography of upper abdominal arteries[J]. Chinese Journal of Hepatic Surgery(Electronic Edition), 2022, 11(05): 469-475.

目的

探讨深度学习图像重建（DLIR）结合低辐射剂量、低对比剂剂量、低对比剂注射速度（三低技术）在上腹部动脉CT血管造影（CTA）中的应用价值。

方法

本研究对象为2021年6月至2021年10月在中山大学附属第三医院接受上腹部动脉CTA检查的60例患者。患者均签署知情同意书，符合医学伦理学规定。其中男33例，女27例；年龄19~86岁，中位年龄49岁。按扫描方案分为标准方案组（S组，30例）和三低方案组（L组，30例）。记录两组的有效辐射剂量（ED）、对比剂剂量、对比剂注射速度。对S组进行60%ASIR-V（S-AV60）图像重建；对L组进行60%ASIR-V（L-AV60）、80%ASIR-V（L-AV80）、DLIR-M（L-DM）和DLIR-H（L-DH）图像重建。客观图像质量评价参数包括上腹部动脉CT值、标准差（SD）值、信噪比（SNR）和对比噪声比（CNR）；主观图像质量由两名放射科医师对重建图像进行双盲法评分。两组ED、对比剂剂量等比较采用t检验；SD、SNR、CNR等客观图像质量评价参数比较采用单因素方差分析；主观图像质量评分比较采用Kruskal-Wallis检验；采用Kappa检验分析两名放射科医师主观评分的一致性。

结果

L组ED为（5.1±1.3）mSv，明显低于S组的（10.5±2.1）mSv（t=-12.397，P<0.05）；对比剂剂量为（65±11）ml，明显低于S组的（100±21）ml（t=-8.150，P<0.05）；与S组注射速度5.0 ml/s相比，L组3.5 ml/s降低30%。对于上腹部CTA成像，L-DH组SD最小，SNR和CNR最大（P<0.05）。L-DH组重建图像的清晰度、图像噪声、图像伪影、小分支显示、临床诊断5个参数的主观图像质量评分分别为（4.7±0.5）、（4.6±0.5）、（4.8±0.4）、（4.5±0.5）、（4.7±0.5）分。5组重建方式中，L-DH组的主观图像质量评分最高（H=118.424，114.258，113.367，121.463，118.778；P<0.05）。两名放射科医师对5组上腹部动脉CTA的主观图像质量评分有较好的一致性（κ=0.672，P<0.05）。

结论

在上腹部动脉CTA中，三低技术可显著降低辐射剂量、对比剂剂量和对比剂注射速度。与推荐的60%ASIR-V标准方案相比，高级别DLIR结合三低技术可进一步改善图像质量，是较好的重建算法。

关键词: 深度学习图像重建, 自适应迭代重建-V, CT血管造影
Objective

To evaluate the application value of deep learning image reconstruction (DLIR) combined with low radiation dose, low contrast agent dose and low contrast agent injection speed (triple-low technique) in the CT angiography (CTA) of upper abdominal arteries.

Methods

60 patients receiving CTA of the upper abdominal arteries in the Third Affiliated Hospital of Sun Yat-sen University from June to October 2021 were recruited in this study. The informed consents of all patients were obtained and the local ethical committee approval was received. Among them, 33 patients were male and 27 female, aged from 19 to 86 years, with a median age of 49 years. According to the scanning plan, all patients were divided into the standard plan group (group S, n=30) and triple-low plan group (group L, n=30). The effective dose (ED), contrast agent dose and contrast agent injection speed in two groups were recorded. 60%ASIR-V (S-AV60) image reconstruction was performed in group S, and 60%ASIR-V (L-AV60), 80%ASIR-V(L-AV80), DLIR-M (L-DM) and DLIR-H (L-DH) image reconstructions were conducted in group L. Parameters of objective evaluation of image quality included CT value, standard deviation (SD) value, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) of the upper abdominal artery. For the subjective image quality evaluation, the reconstructed images were scored with double-blind manner by two radiologists. ED and contrast agent dose were compared between two groups by t test. SD, SNR, CNR and other objective image quality evaluation parameters were compared by one-way ANOVA. Subjective image quality score was compared by Kruskal-Wallis test. The consistency of subjective score between two radiologists was assessed by Kappa test.

Results

In group L, ED was (5.1±1.3) mSv, significantly lower than (10.5±2.1) mSv in group S (t=-12.397, P<0.05). The contrast agent dose in group L was (65±11) ml, significantly lower than (100±21) ml in group S (t=-8.150, P<0.05). Compared with the injection speed of 5.0 ml/s in group S, the injection speed was3.5 ml/s in group L, which was decreased by 30%. For CTA image of the upper abdomen, the SD inL-DH group was the smallest, and the SNR and CNR were the largest (P<0.05). In L-DH group, the subjective image quality scores of five parameters including sharpness, image noise, image artifact, small branch display and clinical diagnosis were 4.7±0.5, 4.6±0.5, 4.8±0.4, 4.5±0.5 and 4.7±0.5, respectively. Among the five reconstructions, the subjective image quality score in the L-DH group was the highest (H=118.424, 114.258, 113.367, 121.463, 118.778; P<0.05). Fair consistency of 5 subjective image quality scores of CTA of upper abdominal arteries by 2 radiologists were observed (κ=0.672, P<0.05).

Conclusions

In the CTA of upper abdominal arteries, "triple-low" technique can significantly reduce the radiation dose, contrast agent dose and contrast agent injection speed. Compared with the recommended 60%ASIR-V standard plan, high-level DLIR combined with "triple-low" technique can further improve the image quality, which is a favorable reconstruction algorithm.

Key words: Deep learning image reconstruction, Adaptive statistical iterative reconstruction-V, Computed tomographic angiography
表1 L组和S组上腹部动脉CTA检查患者一般资料比较
组别 例数 性别（例） 年龄（岁，±s） 身高（cm，±s） 体重（kg，±s） BMI（kg/m2，±s）
男 女
L组 30 17 13 49±16 163±7 65±11 24±3
S组 30 16 14 50±16 165±10 67±14 24±4
统计值   χ2=0.067 t=-0.073 t=-0.690 t=-0.525 t=-0.182
P值   0.795 0.942 0.493 0.602 0.857
表2 L组和S组上腹部动脉CTA检查患者对比剂应用情况比较(±s)
组别 例数 对比剂剂量（ml） CTDIvol（mGy） DLP（mGy·cm） ED（mSv）
L组 30 65±11 12±3 339±84 5.1±1.3
S组 30 100±21 22±4 703±137 10.5±2.1
t值   -8.150 -9.817 -12.397 -12.397
P值   <0.001 <0.001 <0.001 <0.001
表3 五组上腹部动脉CTA图像重建方式的客观图像质量评价参数比较(±s)
参数   S-AV60 L-AV60 L-AV80 L-DM L-DH F值 P值
SD AO 19.6±4.2 23.0±4.6 20.1±4.3 16.2±3.2 11.6±2.3 55.632 <0.001
  CA 21.1±5.1 24.5±5.4 19.7±7.1 18.5±6.1 13.7±4.4 14.437 <0.001
CHA 19.6±4.5 23.2±5.4 17.7±6.5 17.2±4.9 13.4±3.7 14.771 <0.001
SA 20.9±7.2 23.3±7.1 20.5±7.1 20.2±6.6 16.7±5.6 3.750 0.006
SMA 20.3±5.9 18.3±7.2 15.5±6.4 15.0±5.0 11.4±3.7 14.325 <0.001
SNR AO 23±6 18±5 20±5 25±7 35±9 24.889 <0.001
  CA 21±7 16±3 22±9 22±8 31±12 15.728 <0.001
CHA 22±6 17±4 23±9 24±8 30±9 17.353 <0.001
SA 21±8 17±6 20±8 21±8 25±10 3.789 0.006
SMA 23±9 25±13 30±16 29±14 38±17 5.226 0.001
CNR AO 31±7 24±7 31±8 27±8 36±10 9.336 <0.001
  CA 28±8 21±6 27±8 24±6 32±8 10.926 <0.001
CHA 27±6 20±5 26±7 24±6 32±8 14.063 <0.001
SA 27±8 20±6 26±7 23±6 31±7 11.395 <0.001
SMA 29±7 21±6 28±9 24±7 33±8 11.138 <0.001
图1 一例患者上腹部血管MIP和VR重建图注：a组为VR重建，b组为MIP重建；相同窗宽窗位下，从左到右依次为S-AV60、L-AV60、L-AV80、L-DM、L-DH组；S-AV60、L-AV60、L-AV80噪声点密集，L-DM、L-DH组噪声最佳；S-AV60、L-AV60、L-AV80图像在肝总动脉边缘模糊，颗粒感，细小分支血管模糊，有蜡状伪影，斑点状外观；L-DM、L-DH门静脉主干清晰锐利，无颗粒感，细小分支血管纹理清楚，无蜡状伪影；MIP为最大强度投影，VR为容积再现，ASIR-V为自适应统计迭代重建V，DLIR-M、DLIR-H分别为中等级别、高等级别深度学习图像重建，L组为三低方案组，S组为标准方案组，S-AV60为S组60%ASIR-V，L-AV60为L组60%ASIR-V，L-AV80为L组80%ASIR-V，L-DM为L组DLIR-M，L-DH为L组DLIR-H
表4 五组上腹部动脉CTA主观图像质量评分比较（分，±s）
参数 S-AV60 L-AV60 L-AV80 L-DM L-DH H值 P值
清晰度 3.00±0.26 2.23±0.57 2.18±0.80 3.80±0.41 4.67±0.48 118.424 <0.001
图像噪声 2.92±0.49 2.15±0.57 2.15±0.69 3.62±0.49 4.57±0.50 114.258 <0.001
图像伪影 3.20±0.43 2.05±0.67 2.00±0.67 3.80±0.41 4.82±0.36 113.367 <0.001
小分支显示 3.32±0.53 2.03±0.41 2.02±0.43 3.33±0.48 4.48±0.50 121.463 <0.001
临床诊断 3.32±0.41 2.00±0.69 1.97±0.72 3.67±0.48 4.70±0.47 118.778 <0.001
表5 两名放射科医师对上腹部动脉CTA的主观图像质量评分一致性分析（分，±s）
医师 S-AV60 L-AV60 L-AV80 L-DM L-DH H值 P值
医师1 3.12±1.00 2.08±0.19 2.01±0.29 3.64±0.18 4.65±0.18 137.104 <0.001
医师2 3.13±0.13 2.11±0.19 2.07±0.24 3.65±0.18 4.64±0.18 136.829 <0.001
κ值 0.451 0.814 0.567 0.951 0.793    
P值 0.004 <0.001 <0.001 <0.001 <0.001    
[1]
Liu S, Li W, Shi H, et al. Low-dose scanning technology combined with low-concentration contrast material in renal computed tomography angiography (CTA): a preliminary study[J]. Med Sci Monit, 2017(23):4351-4359.
[2]
Malviya KK, Verma A, Nayak AK, et al. Unraveling variations in celiac trunk and hepatic artery by CT angiography to aid in surgeries of upper abdominal region[J]. Diagnostics, 2021, 11(12):2262.
[3]
Verhagen MV, Dikkers R, de Kleine RH, et al. Assessment of hepatic artery anatomy in pediatric liver transplant recipients: MR angiography versus CT angiography[J]. Pediatr Transplant, 2021, 25(4):e14002.
[4]
Zhou Z, Chen H, Wei W, et al. Low kilovoltage peak (kVp) with an adaptive statistical iterative reconstruction algorithm in computed tomography urography: evaluation of image quality and radiation dose[J]. Am J Transl Res, 2016, 8(9):3883-3892.
[5]
Saade C, Deeb IA, Mohamad M, et al. Contrast medium administration and image acquisition parameters in renal CT angiography: what radiologists need to know[J]. Diagn Interv Radiol, 2016, 22(2):116-124.
[6]
陈平, 宋芹霞, 史恒峰, 等. 双源CT三低扫描模式在头颈部血管造影中的应用价值[J]. 安徽医学, 2017, 38(11):1435-1439.
[7]
Sun Y, Hua Y, Wang M, et al. Evaluation of a high concentrated contrast media injection protocol in combination with low tube current for dose reduction in coronary computed tomography angiography: a randomized, two-center prospective study[J]. Acad Radiol, 2017, 24(12):1482-1490.
[8]
Geyer LL, Schoepf UJ, Meinel FG, et al. State of the art: iterative CT reconstruction techniques[J]. Radiology, 2015, 276(2):339-357.
[9]
Laurent G, Villani N, Hossu G, et al. Full model-based iterative reconstruction (MBIR) in abdominal CT increases objective image quality, but decreases subjective acceptance[J]. Eur Radiol, 2019, 29(8):4016-4025.
[10]
Laurent G, Villani N, Hossu G, et al. Full model-based iterative reconstruction (MBIR) in abdominal CT increases objective image quality, but decreases subjective acceptance[J]. Eur Radiol, 2019, 29(8):4016-4025.
[11]
Jensen CT, Liu X, Tamm EP, et al. Image quality assessment of abdominal CT by use of new deep learning image reconstruction: initial experience[J]. AJR Am J Roentgenol, 2020, 215(1):50-57.
[12]
Ren Z, Zhang X, Hu Z, et al. Reducing radiation dose and improving image quality in CT portal venography using 80 kV and adaptive statistical iterative reconstruction-V in slender patients[J]. Acad Radiol, 2020, 27(2):233-243.
[13]
Kulkarni NM, Uppot RN, Eisner BH, et al. Radiation dose reduction at multidetector CT with adaptive statistical iterative reconstruction for evaluation of urolithiasis: how low can we go?[J]. Radiology, 2012, 265(1):158-166.
[14]
刘义军, 刘爱连, 方鑫, 等. 70 kVp超低辐射剂量和低碘摄入量在低体质量指数患者腹部CTA的可行性[J]. 中国医学影像技术, 2017, 33(3):473-477.
[15]
Liu B, Gao S, Chang Z, et al. Lower extremity CT angiography at80 kVp using iterative model reconstruction[J]. Diagn Interv Imaging, 2018, 99(9):561-568.
[16]
Kim SY, Cho JY, Lee J, et al. Low-tube-voltage CT urography using low-concentration-iodine contrast media and iterative reconstruction: a multi-institutional randomized controlled trial for comparison with conventional CT urography[J]. Korean J Radiol, 2018, 19(6):1119-1129.
[17]
Chartrand G, Cheng PM, Vorontsov E, et al. Deep learning: a primer for radiologists[J]. Radiographics, 2017, 37(7):2113-2131.
[18]
Jensen CT, Liu X, Tamm EP, et al. Image quality assessment of abdominal CT by use of new deep learning image reconstruction: initial experience[J]. AJR Am J Roentgenol, 2020, 215(1):50-57.
[19]
Arndt C, Güttler F, Heinrich A, et al. Deep learning CT image reconstruction in clinical practice[J]. Rofo, 2021, 193(3):252-261.
[20]
Willemink MJ, Noël PB. The evolution of image reconstruction for CT-from filtered back projection to artificial intelligence[J]. Eur Radiol, 2019, 29(5):2185-2195.
[21]
Kim JH, Yoon HJ, Lee E, et al. Validation of deep-learning image reconstruction for low-dose chest computed tomography scan: emphasis on image quality and noise[J]. Korean J Radiol, 2021, 22(1):131-138.
[1] 吴蕊含, 孟箭. 根尖周炎致眶尖综合征合并大面积脑梗塞一例[J]. 中华口腔医学研究杂志(电子版), 2016, 10(03): 228-229.
[2] 常冬梅, 王鑫. 迭代重建技术对双低剂量肺动脉CT血管造影图像质量分析[J]. 中华肺部疾病杂志(电子版), 2022, 15(04): 515-517.
[3] 崔永奇, 郑宇, 张丽妍, 陈增爱, 吴学玲. 肺癌合并肺栓塞的胸部CT血管造影特点分析[J]. 中华肺部疾病杂志(电子版), 2020, 13(02): 134-138.
[4] 雷漫诗, 邓锶锶, 汪昕蓉, 黄锦彬, 向青, 熊安妮, 孟占鳌. 人工智能辅助压缩感知技术在上腹部T2WI压脂序列中的应用[J]. 中华肝脏外科手术学电子杂志, 2023, 12(05): 551-556.
[5] 孟占鳌, 张悦, 蒋伟, 郭月飞, 郭焯欣, 张可. DLIR算法结合前置ASIR-V技术在过重患者门静脉成像中的应用[J]. 中华肝脏外科手术学电子杂志, 2022, 11(04): 373-379.
[6] 高英超, 徐菲, 杜亚强, 秦洪梅, 王园园, 陈琳琳, 代拥军. CT血管造影在腹腔镜辅助结肠癌术前评估中的应用[J]. 中华结直肠疾病电子杂志, 2017, 06(03): 226-229.
[7] 中国神经科学学会神经损伤与修复分会, 卫健委脑卒中防治工程委员会专家委员会, 中国卒中学会急救医学分会. 移动CT脑血管造影技术操作专家共识2020[J]. 中华神经创伤外科电子杂志, 2020, 06(02): 76-80.
[8] 房树志, 王洪生, 郭英俊, 孟文博, 程月飞, 张威, 王娟, 赵佩林. 破裂的大脑中动脉镜像动脉瘤的诊断与治疗[J]. 中华神经创伤外科电子杂志, 2017, 03(05): 272-276.
[9] 黎鹏程, 任刚, 张继志, 周敬斌, 王木春, 邓东风. CTA与DSA诊断椎基底动脉狭窄存在差异一例报道并文献复习[J]. 中华脑科疾病与康复杂志(电子版), 2019, 09(03): 188-189.
[10] 尹燕伟, 朱敏, 董祥梅, 胡嘉航, 李永生. 冠状动脉CT血管造影图像标注方法的研究[J]. 中华消化病与影像杂志(电子版), 2022, 12(05): 281-285.
[11] 季鹏, 郭言言, 王超. CT灌注成像联合CT血管造影对TIA患者进展为急性脑梗死的预测[J]. 中华介入放射学电子杂志, 2023, 11(02): 128-132.
[12] 笪彬彬, 赵卫, 姜永能, 石潆. 支气管动脉CTA对介入栓塞治疗咯血的指导及应用价值[J]. 中华介入放射学电子杂志, 2016, 04(01): 13-16.
[13] 李良, 孙全伟, 孙悦. CCTA联合4D－STE对冠状动脉粥样硬化患者MACE的预测价值[J]. 中华心脏与心律电子杂志, 2018, 06(01): 40-42.
阅读次数
全文


摘要

版权所有 中华医学会 中华医学电子音像出版社有限责任公司  京ICP备14006079号-1  网络出版服务许可证：(署)网出证(京)字第075号