Objective: This article aimed to differentiate noncalcified hamartoma from pulmonary carcinoid preoperatively using computed tomography (CT) radiomics approaches. Materials and Methods:The unenhanced CT (UECT) and contrast-enhanced CT (CECT) data of noncalcified hamartoma (n = 73) and pulmonary carcinoid (n = 54; typical/atypical carcinoid = 13/41) were retrospectively analyzed. The patients were randomly divided into the training and validation sets. A total of 396 radiomics features were extracted from UECTand CECT, respectively. The features were selected by using the minimum redundancy maximum relevance and the least absolute shrinkage and selection operator to construct a radiomics model. Clinical factors and radiomics features were integrated to build a nomogram model. The performance of clinical factors, radiomics, and nomogram models on the differential diagnosis between noncalcified hamartoma and carcinoid were investigated. Diagnostic performance of radiologists was also explored. Result:In regard to distinguishing noncalcified hamartoma from carcinoid, the areas under the receiver operating characteristic curves of the clinical, radiomics, and nomogram models were 0.88, 0.94, and 0.96 in the training set UECT, and were 0.85, 0.92, and 0.96 in the training set CECT, respectively. The areas under the curve of the 3 models were 0.89, 0.96, and 0.96 in the validation set UECT, and were 0.79, 0.90, and 0.94 in the validation set CECT, respectively. The nomogram model exhibited good calibration and was clinically useful by decision curve analysis. Nomogram did not show significant improvement compared with radiomics, neither for UECT nor for CECT. Diagnostic performance of radiologists was lower than both radiomics and nomogram model. Conclusions:Radiomics approaches may be useful in distinguishing peripheral pulmonary noncalcified hamartoma from carcinoid. Radiomics features extracted from CECT provided no significant benefit when compared with UECT.

ObjectiveTo explore the potential of CT radiomics in detecting acquired T790M mutation and predicting prognosis in patients with advanced lung adenocarcinoma with progression after first- or second-generation epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI) therapy.Materials and MethodsContrast-enhanced thoracic CT was collected from 250 lung adenocarcinoma patients (with acquired T790M mutation, n = 146, without mutation, n = 104) after progression on first- or second-generation TKIs. Radiomic features were extracted from each volume of interest. The maximum relevance minimum redundancy and the least absolute shrinkage and selection operator (LASSO) regression method were used to select the optimized features in detecting acquired T790M mutation. Univariate Cox regression and LASSO Cox regression were used to establish the radiomics model to predict the progression-free survival of osimertinib treatment. Finally, nomograms (which) combined clinical factors with radscore to predict the acquired T790M mutation and prognosis were built separately. In addition, the two nomograms were validated by the concordance index (C-index), decision curve analysis (DCA), and calibration curve analysis where appropriate.ResultsClinical factors including the progression-free survival of first-line EGFR TKIs, EGFR mutation, and N stage and 12 radiomic features were useful in predicting the acquired T790M mutation. The area under the receiver operating characteristic curves (AUC) of clinical, radiomics, and nomogram models were 0.70, 0.74, and 0.78 in the training set and 0.71, 0.71, and 0.76 in the validation set, respectively. The DCA and calibration curve analysis demonstrated a good performance of the nomogram model. Clinical factors including age and first-generation EGFR TKIs and 12 radiomic features were useful in patients’ outcome prediction. The C-index of the combined nomogram was 0.686 in the training set and 0.630 in the validation set, respectively. Calibration curves demonstrated a relatively poor performance of the nomogram model.ConclusionNomogram combined clinical factors with radiomic features might be helpful to detect whether patients developed acquired T790M mutation or not after progression on first- or second-generation EGFR TKIs. Nomogram prognostic model combined clinical factors with radiomic features might have a limited value in predicting the survival of patients harboring acquired T790M mutation treated with osimertinib.