CCTA Risk Prediction Model Analysis of Coronary Artery Calcification Score and Assessment of Coronary Heart Disease Risk
Objective To calculate the coronary artery calcification scores(CACs)obtained by coronary CT angiography(CCTA)and to predict the probability of coronary heart disease(CHD)by machine learning(ML)analysis combined with the factors influencing coronary artery calcification(CAC).Methods The CCTA data of suspected CHD in hospital were selected,the degree of CAC was quantified,and the risk of CHD was assessed based on CACs and clinical related factors.Results Among the 5 maximum likelihood models,RF had the best accuracy(78.96％),sensitivity(SN)(93.86％),specificity(51.13％),Matthew correlation coefficient(Mcc)(0.5192),and the best AUC area(0.8375),which was much better than the other 4 ML models.Conclusion Computer ML model analysis confirmed the importance of CACs in predicting the occurrence of coronary heart disease,especially the prominent RF model.
万方数据
维普
