Differential diagnosis of Xpert MTB/RIF-negative pulmonary tuberculosis and non-tuberculous mycobacteria pulmonary disease based on CT radiomics
10.3969/j.issn.1002-1671.2025.05.009
- VernacularTitle:基于CT影像组学鉴别Xpert MTB/RIF阴性肺结核及非结核分枝杆菌肺病的研究
- Author:
Shengwei LU
1
;
Feng LI
;
Qi DAI
;
Jingfeng ZHANG
;
Jianjun ZHENG
Author Information
1. 浙江中医药大学宁波研究生联合培养基地,浙江 宁波 315000;宁波市第二医院放射科,浙江 宁波 315000
- Publication Type:Journal Article
- Keywords:
radiomics;
Xpert MTB/RIF;
pulmonary tuber-culosis;
non-tuberculous mycobacteria pulmonary disease
- From:
Journal of Practical Radiology
2025;41(5):757-761
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the value of a CT radiomics model in differentiating Xpert MTB/RIF-negative pulmonary tuber-culosis(PTB)from non-tuberculous mycobacteria pulmonary disease(NTM-PD).Methods A retrospective analysis was performed on 90 patients with Xpert MTB/RIF-negative PTB and 127 patients with NTM-PD.All patients were randomly divided into training set and testing set at the ratio of 7∶3.Radiomics features were extracted from chest CT images.Feature dimensionality reduction and selection were sequentially performed using the maximum relevance and minimum redundancy(mRMR)algorithm and the least absolute shrinkage and selection operator(LASSO)algorithm.Clinical,radiomics,and combined models were constructed by multi-variable logistic regression.The area under the curve(AUC)of receiver operating characteristic(ROC)curve was utilized to assess the model diagnostic performance.Calibration curves were used to evaluate model stability,and the decision curve analysis(DCA)was used to evaluate the clinical utility.Results The combined model had the highest diagnostic performance in both training and testing sets,with AUC of 0.90 and 0.86,respectively,which were higher than clinical and radiomics models.The calibration curve showed that the combined model had a good consistency between the predicted and the actual observations,and DCA revealed the highest clinical benefit.Conclusion The clinical-radiomics combined model has excellent predictive ability in differentiating Xpert MTB/RIF-negative PTB from NTM-PD,which can provide robust support for clinical diagnosis.
