首页|基于人工智能的冠状动脉周围脂肪衰减指数联合CT血流储备分数预测冠心病患者不良心血管事件风险的价值

基于人工智能的冠状动脉周围脂肪衰减指数联合CT血流储备分数预测冠心病患者不良心血管事件风险的价值

The Value of Predicting the Risk of Adverse Cardiovascular Events in Patients with Coronary Heart Disease Using Artificial Intelligence Based Pericoronal Fat Attenuation Index Combined with CT-FFR Technology

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
目的 探讨基于人工智能的冠状动脉周围脂肪衰减指数(FAI)联合CT的血流储备分数(CT-FFR)技术预测冠状动脉粥样硬化性心脏病(冠心病)患者不良心血管事件(MACE)风险的价值.方法 选取2022年6月至2023年3月我院行冠状动脉CT血管造影(CCTA)检查的冠心病患者180例,经排除标准筛选后最终纳入128例,根据MACE发生情况,分为MACE组(n=35)、非MACE组(n=93).采用Logistic回归分析MACE发生风险因素,并比较不同冠状动脉狭窄程度、不同斑块类型的冠心病患者冠周FAI、CT-FFR水平,经受试者工作特征曲线(ROC)曲线分析FAI、CT-FFR对MACE的联合预测价值.结果 糖尿病史、冠状动脉狭窄程度、斑块类型、FAI、CT-FFR为冠心病患者发生MACE的风险因素(P＜0.05);中度冠状动脉狭窄患者FAI低于重度冠状动脉狭窄患者,CT-FFR高于重度冠状动脉狭窄患者(P＜0.05);非钙化斑块患者FAI高于钙化斑块患者,CT-FFR低于钙化斑块患者(P＜0.05);FAI、CT-FFR联合预测MACE发生的ROC曲线下面积(AUC)为0.882,优于各指标单独预测价值(P＜0.05).结论 糖尿病史、冠状动脉狭窄程度、斑块类型、FAI、CT-FFR为MACE发生影响因素,FAI、CT-FFR对冠心病患者MACE发生具有良好预测价值.
Objective Exploring the value of artificial intelligence based pericoronal fat attenuation index(FAI)com-bined with CT based blood flow reserve fraction(CT-FFR)technology in predicting the risk of adverse cardiovascular e-vents(MACE)in patients with coronary heart disease.Methods From June 2022 to March 2023,180 consecutive pa-tients with coronary heart disease who underwent coronary CT angiography(CCTA)in our hospital were selected,and 128 cases were finally included after exclusion criteria screening.Based on the incidence of MACE,they were divided into MACE group(n=35)and non MACE group(n=93).Logistic regression analysis was used to analyze the risk factors for MACE,and the levels of pericoronal FAI and CT-FFR were compared among coronary heart disease patients with different degrees of coronary stenosis and plaque types.The ROC curve was used to analyze the combined predictive value of peri-coronal FAI and CT-FFR on MACE.Results The history of diabetes,the degree of coronary stenosis,plaque type,peri-coronal FAI,CT-FFR were risk factors for MACE in patients with coronary heart disease(P＜0.05).Patients with moderate coronary artery stenosis had lower pericoronal FAI than those with severe coronary artery stenosis,while CT-FFR was higher than those with severe coronary artery stenosis(P＜0.05).Patients with non calcified plaques had higher pericoronal FAI than those with calcified plaques,while CT-FFR was lower than those with calcified plaques(P＜0.05).The combined pre-diction of COVID-19 FAI and CT-FFR for AUC in MACE was 0.882,which was superior to the individual predictive value of each indicator(P＜0.05).Conclusion The history of diabetes,degree of coronary stenosis,plaque type,pericoronal FAI and CT-FFR are the influencing factors of MACE,and pericoronal FAI and CT-FFR have good predictive value for MACE in patients with coronary heart disease.

Coronary heart diseaseArtificial intelligencePericoronal fat attenuation indexBlood flow reserve fractionAdverse cardiovascular eventsPredictive value

解福友、刘艺超、邱晓晖、王利、涂胜

展开 >

236800 安徽,亳州市人民医院影像中心

亳州市人民医院心内科

冠状动脉粥样硬化性心脏病 人工智能 冠状动脉周围脂肪衰减指数 血流储备分数 不良心血管事件 预测价值

2020年度高校科学研究项目2023年度科研项目

KJ2020A0336by2023032

2024

临床放射学杂志
黄石市医学科技情报所

临床放射学杂志

CSTPCD北大核心
影响因子：0.872
ISSN：1001-9324
年,卷(期)：2024.43(8)
  • 17