|
|
|
| The value of artificial intelligence in post-processing coronary CTA images and diagnosing coronary artery stenosis |
| QI Dong, YAO Chuanshun, HU Shumin, HE Xingyi, YAO Muzi, SHEN Chaojun |
(Department of Medical Imaging, the First People′s Hospital of Bengbu, Bengbu Anhui 233000, China)
|
|
|
|
Abstract Objective： To analyze the value of artificial intelligence (AI) for coronary CT angiography (CCTA) image postprocessing and diagnosis of coronary artery disease. Methods： The imaging data of 40 patients who both underwent CCTA and percutaneous coronary arteriography (CAG) from January 2020 to January 2022 were retrospectively collected and analyzed. Image postprocessing and diagnosis were divided into manual group and AI group to compare the differences in postprocessing time and subjective image quality scores and to assess the differences in coronary plaque properties. Using CAG results as the gold standard, the differences in sensitivity, specificity, positive predictive value, negative predictive value and accuracy in the diagnosis of coronary artery stenosis between the two were compared on a vessel-by-vessel basis, and the consistency of the results was evaluated using the Kappa test. Results： The postprocessing and diagnosis time in the AI group was (236.57±20.66)s, which was approximately 70.04% shorter than that in the manual group (789.74±63.38)s. The difference was statistically significant (P<0.05); The difference between the objective and subjective scores of image quality obtained by the two methods was not statistically significant (all P>0.05). The overall accuracy of plaque detection in the AI group was 96.32% (131/136). The differences in the detection of calcified coronary plaque, noncalcified plaque and mixed plaque between the manual and AI groups were not statistically significant (P>0.05) and were in good agreement (Kappa=0.901, P<0.001). On a vesselbyvessel basis, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy of the AI group in diagnosing coronary artery stenosis were 87.72% (50/57), 94.12% (48/51), 94.34% (50/53), 87.27% (50/53) and 90.74% (98/108), respectively, and the CAG in diagnosing coronary artery stenosis with good consistency (Kappa=0.815, P<0.001). Conclusion： AI has certain advantages in CCTA image post-processing efficiency, plaque nature identification and coronary stenosis, and can be used as an effective aid for physicians to analyze and diagnose coronary artery disease.
|
|
Received: 13 June 2022
|
|
|
|
［1］盛寿, 高润霖, 刘力生, 等. 《中国心血管病报告2018》概要［J］. 中国循环杂志, 2019, 34(3): 209-220.
［2］Dreisbach JG, Nicol ED, Roobottom CA, et al. Challenges in delivering computed tomography coronary angiography as the firstline test for stable chest pain［J］. Heart, 2018, 104(11): 921-927.
［3］Motwani M, Dey D, Berman DS, et al. Machine learning for prediction of allcause mortality in patients with suspected coronary artery disease: a 5year multicentre prospective registry analysis［J］. Eur Heart J, 2017, 38(7): 500-507.
［4］Zreik M, Lessmann N, Van Hamersveh RW, et al. Deep learning analysis of the myocardium in coronary CT angiog raphy for identification of patients with functionally significant coronary artery stenosis［J］. Med Image Anal, 2018, 44(2): 72-85.
［5］赵小英, 李露露, 吴兴旺, 等. 宽体探测器CT追踪冻结技术对提高冠状动脉图像质量的应用评价［J］. 安徽医学, 2019, 40(6): 606-609.
［6］蒋杏芳, 应明亮. 256层与128层螺旋CT图像后处理技术对冠脉狭窄诊断效能的对比分析［J］. 浙江实用医学, 2018, 23(2): 120, 153.
［7］Newman JD, Rockman CB, Kosiborod M, et al. Diabetes mellitus is a coronary heart disease risk equivalent for peripheral vascular disease［J］. Am Heart J, 2017, 184: 114-120.
［8］Eipsic J, Abbara S, Achenbach S, et al. SCCT guidelines for the interpretation and reporting of coronary CT angiography: A report of the society of cardiovascular computed tomography guidelines committee［J］. J Cardiovasc Comput Tomogr, 2014, 8(5): 342-358.
［9］金晨望, 郭佑民. 人工智能辅助诊断技术在低剂量CT肺结节筛查中的应用及质控［J］. 中国放射学杂志, 2019, 53(1): 6-8.
［10］中华医学会放射学分会心胸学组. 心脏冠状动脉CT血管成像技术规范化应用中国指南［J］. 中华放射学杂志, 2017, 51(10): 732-743.
［11］马景旭, 陈欢, 王红. 分析肺部影像人工智能诊断系统鉴别诊断肺结节性质［J］. 临床肺科杂志, 2021, 26(6): 842-846.
［12］刘丰伟, 李汉军, 张逸鹤, 等. 人工智能在医学影像诊断中的应用［J］. 北京生物医学工程, 2019, 38(2): 206-211.
［13］赵明娟, 单春辉, 黄晓颖, 等. 人工智能结构化报告评价冠状动脉狭窄的应用研究［J］. 实用放射学杂志, 2021, 37(5): 758-762.
［14］萧毅, 刘士远. 客观看待人工智能在医学影像中的作用［J］. 放射学实践, 2018, 33(10): 992-994.
［15］张晓浩, 刘军波, 范丽娟. 人工智能技术应用于冠状动脉CTA图像后处理的可行性［J］. 放射学实践, 2021, 36(8): 994-999.
［16］Stary HC, Chandler AB, Dinsmore RE, et al. A definition of advanced types of atherosclerotic lesions and a histological classification of atherosclerosis. A report from the committee on vascular lesions of the council on arteriosclerosis, American heart association［J］. Arterioscler Thromb Vasc Biol, 1995, 15(9): 1512-1531.
［17］冯长超, 付敏. CT血管造影术在评估冠脉狭窄程度及其与斑块钙化程度关系中的应用［J］. 中国临床研究, 2018, 31(10): 1347-1350.
［18］王平, 龙晓宙, 黄刚, 等. 冠状动脉CT成像人工智能辅助诊断的应用价值［J］. 甘肃医药, 2020, 39(2): 122-125.
［19］王亚娟, 辛娟, 刘文亚, 等. 人工智能辅助诊断系统对冠状动脉狭窄的诊断价值［J］. 中华实用诊断与治疗杂志, 2021, 35(4): 408-411.
［20］李浚利, 黄益龙, 韩丹, 等. 冠状动脉CT血管成像中人工智能诊断冠心病的准确性［J］. 中国医学影像技术, 2021, 37(1): 59-62.
［21］黄增发, 王翔. 人工智能冠状动脉CT血管成像在冠心病诊断中的应用［J］. 放射学实践, 2018, 33(10): 1017-1021.
|
|
|
|
