基于胸部X线影像组学列线图诊断小儿肺炎支原体肺炎  

Radiomics nomogram based on chest X-ray in identifying mycoplasma pneumoniae pneumonia in children

作　　者：康嗣如 田荣华[1] KANG Si-ru;TIAN Rong-hua(Department of Radiology,Xiaogan Hospital affiliated to Wuhan University of Science and Technology,Hubei 432000,China)

机构地区：[1]武汉科技大学附属孝感医院放射影像科,湖北432000

出　　处：《放射学实践》2025年第2期265-272,共8页Radiologic Practice

基　　金：孝感市自然科学计划项目(XGKJ2022010002)。

摘　　要：目的:探讨基于胸部X线影像组学列线图诊断肺炎支原体肺炎(MPP)的效能及可行性。方法:回顾性分析2022年4至2023年10月间于本院行胸部X线正位片检查患者共计129例,在胸部X线正位片图像上勾画感兴趣区(ROI)并提取放射组学特征,将样本按8:2随机分为训练组(n=103)和测试组(n=26)。利用最小绝对收缩和选择算子(LASSO)降低数据维度、选择特征,利用逻辑回归(LR)建立预测MPP的临床模型、放射组学模型及联合模型,并绘制出联合模型对应的诺模图(Nomogrm);利用受试者工作特征(ROC)曲线下面积(AUC)、准确性、特异性、敏感性、阳性预测值和阴性预测值等指标评估组合模型的性能,使用DeLong检验比较模型间的预测效能,绘制模型的校准曲线并采用Hosmer-Lemeshow验证模型的拟合优度,采用决策曲线分析(DCA)评估各个模型的临床价值。结果:临床模型、影像组学模型及联合模型在训练组中的AUC分别为0.731[95%置信区间(CI):0.631~0.832]、0.804[95%CI:0.718~0.890]、0.808[95%CI:0.721~0.896]。在测试组中的AUC分别为0.839[95%CI:0.679~0.999]、0.881[95%CI:0.754~1.000]、0.899[95%CI:0.777~1.000]。DeLong检测结果显示联合模型与临床模型差异存在统计学意义(P<0.05),联合模型与影像组学模型差异无统计学意义(P>0.05)。Hosmer-Lemeshow检验显示模型校正良好(P=0.705、0.705、0.336)。DCA结果显示影像组学模型、联合模型预测MPP的临床价值均优于临床模型。结论:基于胸部X线影像组学列线图具有客观、准确诊断肺炎支原体肺炎的潜力。Objective:To investigate efficacy of Radiomics nomogram based on the chest X-ray in identifying Mycoplasma pneumoniae pneumonia(MPP).Methods:A retrospective analysis was conducted on a total of 129 patients who underwent chest X-ray examinations at our institution between April 2023 and September 2023.Regions of interest(ROI)were delineated on the images.Radiomic features were extracted from the images.The samples were randomly divided into a training group(n=103)and a test group(n=26)at an 8:2 ratio.The least absolute shrinkage and selection operator(LASSO)was used to reduce data dimensionality and select features.Logistic regression(LR)was employed to establish clinical models,radiomic models,and a combined model for predicting MPP.Combining clinical independent predictors and Radscore to construct the combined model and drawing its nomogram.The performance of the composite models was evaluated using metrics such as the area under the receiver operating characteristic(ROC)curve(AUC),accuracy,specificity,sensitivity,positive predictive value,and negative predictive value.DeLong test was used to compare the predictive performance among the models.Calibration curves for the models were plotted,and Hosmer-Lemeshow test was applied to assess model fit.Decision curve analysis(DCA)was used to evaluate the clinical utility of each model.Results:In the training group,the AUC for the clinical model,radiomic model,and combined model were 0.731[95%Confidence Interval(CI):0.631~0.832]、0.804[95%CI:0.718~0.890]、0.808[95%CI:0.721~0.896],respectively.In the test group,the AUCs were 0.839[95%CI:0.679~0.999]、0.881[95%CI:0.754~1.000]、0.899[95%CI:0.777~1.000],respectively.DeLong test results indicated a statistically significant difference between the combined model and the clinical model(P<0.05),while there was no statistically significant difference between the combined model and the radiomic model(P>0.05).The Hosmer-Lemeshow test indicated that the models were well calibrated(P=0.705,0.705,0.336).The results of DCA demon

关 键 词：影像组学 列线图 肺炎 支原体 

分 类 号：R195.1[医药卫生—卫生统计学] R563.1[医药卫生—卫生事业管理]