目的:基于电子计算机断层扫描(CT)影像组学构建儿童普通肺炎进展为重症肺炎的列线图模型。方法:选取 234例儿童肺炎作为研究对象，根据重症肺炎的发生情况分为重症肺炎组和普通肺炎组。应用Logistic回归分析筛选儿童普通肺炎进展为重症肺炎的危险因素，采用 R 软件建立儿童普通肺炎进展为重症肺炎的列线图模型。结果:儿童普通肺炎进展为重症肺炎的列线图模型的校正曲线预测效能较佳，模型的ROC曲线下面积 0。950。结论:基于 CT 影像组学的儿童普通肺炎进展为重症肺炎的列线图模型准确率和临床应用价值较高，能够用于重症肺炎的预测。
Predictive value of clinical-imaging nomogram model for progression of common pneumonia to severe pneumonia in children
Objective:To construct a column chart model of children with common pneumonia progressing to severe pneumonia based on electronic computed tomography(CT)imaging omics.Method:234 children with pneumonia were selected as the research subjects,and were divided into severe pneumonia group and common pneumonia group based on the occurrence of severe pneumonia.Using logistic regression analysis to screen for risk factors for the progression of pediatric common pneumonia to severe pneumonia,and using R software to establish a column chart model for the progression of pediatric common pneumonia to severe pneumonia.Result:The column chart model of children with common pneumonia progressing to severe pneumonia had better predictive performance on the correction curve,with an area under the ROC curve of 0.950.Conclusion:The column chart model based on CT imaging omics has high accuracy and clinical application value in predicting the progression from common pneumonia to severe pneumonia in children,and can be used for the prediction of severe pneumonia.
万方数据
维普
