摘要
目的探讨基于冠状动脉CT血管成像(CCTA)的人工智能(AI)诊断系统及CT无创血流储备分数(CT-FFR)在评估高海拔地区冠状动脉临界病变结构及功能学中的应用价值。方法前瞻性收集2022年1月~2023年10月青海大学附属医院冠心病临界病变患者164例,按居住地海拔进行分组,其中2000~3000m为A组(n=83),3000m以上为B组(n=81),再将两组患者按冠脉狭窄程度细分为50%~60%亚组(n=84)和61%~70%亚组(n=80)。将患者冠状动脉CT血管成像数据导入AI辅助诊断及CT-FFR测量系统,以冠脉造影及冠脉传统血流储备分数(FFR)为金标准,分别评价AI及CT-FFR在高海拔地区冠脉临界病变诊断中的应用。结果以FFR为金标准,CT-FFR与FFR的一致性为83.75%。B组钙化斑块、易损斑块高于A组(P=0.037、0.020);B组冠状动脉多支病变、61%~70%狭窄程度发生率均高于A组(P<0.05);A组、B组在61%~70%亚组钙化斑块、易损斑块发生率均高于50%~60%亚组(P<0.05)。B组CT-FFR值低于A组(0.76±0.04 vs 0.88±0.05,P<0.01);A、B两组在61%~70%亚组CT-FFR值≤0.80、<0.70的发生率高于50%~60%亚组(P<0.05)。结论AI诊断系统及CT-FFR对评估高海拔地区冠状动脉临界病变的结构特征及血流动力学改变的结果与冠脉造影、FFR一致性高,具有较高的诊断敏感度和特异度。
Objective To explore the application value of artificial intelligence(AI)diagnostic system based on coronary CT angiography and CT non-invasive blood flow reserve fraction(CT-FFR)in assessing the function of coronary artery critical lesions at high altitude.Methods A prospective collection was conducted on 164 patients with critical coronary artery disease at Qinghai University Affiliated Hospital from January 2022 to October 2023.They were grouped according to their residential altitude,with group A at 2000-3000 m(n=83)and group B at>3000 m(n=81).The two groups of patients were further divided into subgroups of 50%-60%(n=84)and 61%-70%(n=80)based on the degree of coronary stenosis.Import patient CCTA data into AI assisted diagnosis and CT-FFR measurement systems,and evaluate the application of AI and CT-FFR in the diagnosis of coronary critical lesions in high-altitude areas using coronary angiography and traditional coronary FFR as gold standards.Results Using FFR as the gold standard,the consistency between CT-FFR and FFR was 83.75%.The calcified and vulnerable plaques in group B were higher than those in group A(P=0.037,0.020);The incidence of multi branch coronary artery disease and 61%-70%stenosis degree in group B was higher than that in group(P<0.05);The incidence of calcified and vulnerable plaques in the 61%-70%subgroups of group A and group B was higher than that in the 50%-60%subgroups(P<0.05).The CT-FFR value of group B was significantly lower than that of group A(0.76±0.04 vs 0.88±0.05,P<0.01);The incidence of CT-FFR values≤0.80 and<0.70 in the 61%-70%subgroups of group B was higher than that in the 50%-60%subgroups(P<0.05).Conclusion CT-FFR diagnostic system based on AI has a high consistency with FFR in evaluating coronary artery characteristics and hemodynamic changes in patients with critical coronary artery lesions at different altitudes,and has a high diagnostic sensitivity and specificity,which significantly improves the diagnostic efficiency.
作者
王雪燕
曹云太
韩千程
颜梅
韩玲
温生宝
WANG Xueyan;CAO Yuntai;HAN Qiancheng;YAN Mei;HAN Ling;WEN Shengbao(Medical Image Center,Qinghai University Affiliated Hospital,Xining 810000,China)
出处
《分子影像学杂志》
2024年第6期616-621,共6页
Journal of Molecular Imaging
基金
青海省医药卫生科技项目(2021-wjzdx-40)。
关键词
人工智能
冠状动脉
CT无创血流储备分数
CT血管成像
冠脉临界病变
artificial intelligence
coronary artery
fractional flow reserve derived from computed tomography angiography
CT angiography
critical lesion