目的 探讨基于人工智能辅助（AI）的冠状动脉CT血管成像（CCTA）多模态成像技术在冠状动脉解剖畸形诊断中的应用价值。 方法 回顾性分析2023年1月至2024年12月行冠状动脉CTA检查且冠状动脉解剖存在畸形的161例患者临床资料。采用“AI辅助诊断软件”对CCTA影像资料自动进行冠状动脉解剖畸形诊断，以医师组诊断为金标准，采用Kappa一致性检验评价AI软件与医师组对冠状动脉解剖畸形诊断结果的一致性。 结果 共随访冠状动脉7075例，其中解剖畸形患者161例，发病率约2.28%，AI 组与医师组在冠状动脉解剖畸形诊断检测结果上一致性高，Kappa=0.967（P<0.001）。 结论 AI软件对冠状动脉解剖畸形的检出率高，与人工诊断存在较高的一致性，可较好辅助影像医师对冠状动脉解剖畸形的诊断。
Abstract
Objective To explore the application value of artificial intelligence-assisted coronary CT angiography (CCTA) multimodal imaging technology in the diagnosis of coronary artery anatomical malformations. Methods The clinical data of 161 patients who underwent coronary artery CTA examination from January 2023 to December 2024 and had coronary artery anatomical malformations were retrospectively analyzed. Artificial Intelligence (AI) Coronary Assisted Diagnosis Software was used to automatically diagnose coronary artery anatomical malformations on the CCTA imaging data, with the diagnosis of physician group as gold standard, and Kappa consistency test was used to evaluate the consistency of the diagnosis results of coronary artery anatomical malformations between AI group and physician group. Results A total of 7075 cases of coronary arteries were followed up, including 161 patients with anatomical malformations, with a prevalence rate of about 2.28%, and there was a high degree of consistency between AI group and the physicians' group in the diagnostic test results of coronary artery anatomical malformations, with a Kappa=0.967 (P<0.001). Conclusions AI has a high detection rate of coronary artery anatomical malformations, and there is a high consistency with manual diagnosis, which can better assist imaging physicians in the diagnosis of coronary artery anatomical malformations.
关键词
人工智能 /
/
/
CT血管成像 /
/
/
冠状动脉 /
/
/
解剖畸形
Key words
Artificial intelligence (AI) /
/
/
CT angiography /
/
/
Coronary artery /
/
/
Anatomical malformation
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Bansal A, Sarkar PG, Gupta MD, et al. Prevalence and patterns of coronary artery anomalies in 28,800 adult patients undergoing angiography in a large tertiary care centre in India [J]. Monaldi Arch Chest Dis, 2021, 92(3): 2066. DOI:10.4081/monaldi.2021.2066.
[2] Balfour CP, Gonzalez AJ, Kramer MC. Non-invasive assessment of low- and intermediate-risk patients with chest pain[J]. Trends Cardiovasc Med, 2017, 27(3): 182-189. DOI: 10.1016/j.tcm.2016.08.006.
[3] 黄瑞, 李旭, 孙菁菁, 等. CT血管成像在冠状动脉疾病中应用价值[J]. 心脏杂志 ,2024, 36(1): 54-58. DOI: 10.12125/j.chj.202304036.
Huang R, Li X, Sun JJ, et al. Application value of CT angiography in coronary artery disease[J]. Heart Journal, 2024, 36(1):54-58. DOI: 10.12125/j.chj.202304036.
[4] Du M, He S, Liu J, et al. Artificial Intelligence in CT Angiography for the Detection of Coronary Artery Stenosis and Calcified Plaque: A Systematic Review and Meta-analysis[J]. Acad Radiol, 2025, 32(7): 3776-3787. DOI: 10.1016/j.acra.2025.03.054.
[5] 李顶, 汪艳芳, 李永欣, 等. 人工智能在医学影像诊断中的应用研究[J]. 中国临床解剖学杂志, 2020, 38(1):110-113. DOI: 10.13418/j.issn.1001-165x.2020.01.023.
Li D, Wang YF, Li YX, et al. Application research of artificial intelligence in medical image diagnosis[J]. Chinese Journal of Clinical Anatomy, 2020, 38(1):110-113. DOI: 10.13418/j.issn.1001-165x.2020. 01.023.
[6] Jiang B, Guo N, Ge Y, et al. Development and application of artificial intelligence in cardiac imaging[J]. Br J Radiol, 2020, 93(1113):20190812. DOI: 10.1259/bjr.20190812.
[7] 安琪, 李守军. 先天性心脏病外科治疗中国专家共识（十二）：先天性冠状动脉异常[J].中国胸心血管外科临床杂志, 2020, 27(12):1375-1381. DOI: 10.7507/1007-4848.202008031.
An Q, Li SJ, Chinese expert consensus on surgical treatment of congenital heart disease (12): Congenital coronary artery anomalies[J]. Chinese Journal of Clinical Thoracic and Cardiovascular Surgery, 2020, 27(12):1375-1381. DOI: 10.7507/1007-4848.202008031.
[8] 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. DOI: 10.13609/j.cnki.1000-0313.2021.08.009.
Zhang XH, Liu JB, Fan LJ. Feasibility of applying artificial intelligence technology to post-processing process of coronary CT angiography images[J]. Radiology Practice, 2021, 36(8): 994-999. DOI: 10.13609/j.cnki.1000-0313.2021.08.009.
[9] Li Y, Wu Y, He J, et al. Automatic coronary artery segmentation and diagnosis of stenosis by deep learning based on computed tomographic coronary angiography[J]. Eur Radiol, 2022, 32(9):6037-6045. DOI: 10.1007/s00330-022-08761-z.
[10]Yaman S, Aslan O, Güler H, et al. Deep learning techniques for automated coronary artery segmentation and coronary artery disease detection: A systematic review of the last decade (2013-2024) [J]. Comput Methods Programs Biomed, 2025, 268: 108858. DOI: 10.1016/j.cmpb.2025.108858.
[11]Fuenzalida VJJ, Rodriguez BSE, Muñoz QSA, et al. Anatomical Variants of the Origin of the Coronary Arteries: A Systematic Review and Meta-Analysis of Prevalence[J]. Diagnostics, 2024, 14(13):1458. DOI: 10.3390/diagnostics14131458.
[12]Liu CY, Tang CX, Zhang XL, et al. Deep learning powered coronary CT angiography for detecting obstructive coronary artery disease: The effect of reader experience, calcification and image quality [J]. Eur J Radiol, 2021, 142: 109835. DOI: 10.1016/j.ejrad.2021.109835.
[13]Shiri I, Baj G, Kazaj MP, et al.AI-based detection and classification of anomalous aortic origin of coronary arteries using coronary CT angiography images[J]. Nat Commun, 2025, 16(1): 3095. DOI:10.1038/s41467-025-58362-9.
[14]Khan AB, Iqbal F, Gul M, et al. A rare symptomatic case of congenital origin of right coronary artery from left coronary sinus[J]. Cureus, 2022, 14(5): e25358. DOI: 10.7759/cureus.25358.
基金
河北省卫生健康委员会医学科学研究课题计划项目（20250944）
PDF(2772 KB)
京ICP备08002354号-3
