The application value of CT-based radiomics in differentiating pneumonia-type mucinous adenocarcinoma from organizing pneumonia
10.3969/j.issn.1002-1671.2025.08.011
- VernacularTitle:基于CT影像组学鉴别肺炎型黏液腺癌与机化性肺炎的应用价值
- Author:
Xiaoqing LI
1
;
Kexin XIE
;
Rong LIU
;
Can CUI
;
Shuai REN
;
Hai XU
;
Liang ZENG
Author Information
1. 南京中医药大学附属医院放射科,江苏 南京 210029
- Publication Type:Journal Article
- Keywords:
radiomics;
pneumonia-type mucinous adenocarcinoma;
organizing pneumonia;
nomogram;
computed tomography
- From:
Journal of Practical Radiology
2025;41(8):1304-1309
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the application value of CT-based radiomics in differentiating pneumonia-type mucinous adenocarcinoma(PTMA)from organizing pneumonia(OP).Methods A total of 52 PTMA patients and 102 OP patients were retrospectively included and randomly divided into training set(n=124)and test set(n=30)in an 8∶2 ratio.Eight PTMA patients and 22 OP patients from another hospital during the same period were included as external validation set(n=30).Clinical characteristics and CT signs of the patients were selected to construct the clinical model.Radiomics features were extracted and dimensionality reduction was performed through the least absolute shrinkage and selection operator(LASSO)algorithm.A radiomics model was constructed and the Radiomics score(Radscore)was calculated.The Radscore was combined with clinical factors to establish the combined model and a nomogram was illustrated.The models' fitting degree was analyzed by the calibration curve,while their efficacy was evaluated by the receiver operating characteristic(ROC)curve and decision curve analysis(DCA).Results The clinical model,established based on the border,cystic space and bronchial leafless tree sign,achieved area under the curve(AUC)of 0.850,0.782,and 0.759 in the training set,test set,and external validation set,respectively.Thirteen features were obtained to construct the radiomics model,with AUC of 0.925,0.865,and 0.830,respectively.The AUC of the combined model were 0.970,0.905,and 0.864,respectively,in which the calibration curve demonstrated good model fitting.DCA indicated that the combined model had the greatest clinical net benefit.Conclusion The combined model based on CT radiomics can effectively distinguish PTMA from OP.
