AUTHOR=Zhao Jiabi , Sun Lin , Sun Ke , Wang Tingting , Wang Bin , Yang Yang , Wu Chunyan , Sun Xiwen 

TITLE=Development and Validation of a Radiomics Nomogram for Differentiating Pulmonary Cryptococcosis and Lung Adenocarcinoma in Solitary Pulmonary Solid Nodule

JOURNAL=Frontiers in Oncology

VOLUME=Volume 11 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2021.759840

DOI=10.3389/fonc.2021.759840

ISSN=2234-943X

ABSTRACT=Objective: To develop and validate the performance of a computer tomography(CT)-based radiomics nomogram in differentiating pulmonary cryptococcosis(PC) and lung adenocarcinoma(LAC) manifesting as solitary pulmonary solid nodules(SPSNs).
Materials and Methods: We retrospectively recruited 213 patients with pulmonary cryptococcosis and 213 cases of lung adenocarcinoma (matched based on age and gender) with their clinical characteristics and radiological features. Radiomics features were extracted from each SPSNs. The minimum redundancy and maximum relevance (mRMR) and least absolute shrinkage and operator (LASSO) algorithms were applied to select useful features for building a radiomics signature. After multivariable logistics regression analysis, a radiomics nomogram model incorporating clinical characteristics , CT radiological features, and radiomics signature were constructed and evaluated its diagnostic performance by receiver operating characteristic(ROC) curve analysis and clinical effectiveness by decision curve analysis(DCA). 
Results: A total of 1130 radiomics features were extracted from the CT image and the 24 most significant features were retained and constructed the radiomics signature. Three factors — maximum diameter, lobulation, and pleural retraction were found to be independent predictors and used to build the radiomics nomogram, which showed better diagnostic ability than any single model. The area under curve (AUC) yielded was 0.91 and 0.89 in training and test cohorts, respectively. The net reclassification index (NRI) indicated the nomogram had significantly improved prediction performance than the clinical model in the training and test cohorts(all NRI p<0.05) and Decision curve analysis demonstrated that the radiomics nomogram was more clinically useful than the clinical model.
Conclusions: The radiomics nomogram we proposed can preoperatively distinguish between LAC and PC in patients with a SPSN.