Development and Validation a Radiomic Nomogram for the Preoperative Prediction of Indeterminate Small (≤10 mm) Pulmonary Nodules

Abstract

Background: The incidence of indeterminate small pulmonary nodules (ISPN) has been increasing worldwide, largely due to more sensitive computed tomography (CT) screening. However, CT is unsatisfactory in the discrimination of ISPN (≤10 mm). Methods: In total, 210 pathologically confirmed ISPN were divided into a development set (n = 105) and a validation set (n = 105). Radiomic features were extracted from the cancerous volumes of interest (VOIs) on contrast-enhanced CT images. LASSO regression model was used for data dimension reduction, feature selection, and radiomic signature building. Based on the multivariable logistic regression analysis, a radiomic nomogram was developed incorporating the radiomic signature and the useful CT features reported by radiologists. Discrimination, calibration and clinical usefulness of the nomogram were evaluated. Findings: Consisting of selected 6 features, the radiomic signature showed a favorable discrimination in the development dataset with an AUC of 0·877, which was confirmed in the validation dataset with an AUC of 0·811. The CT signature including two CT features attained an AUC of 0·825 and 0·797 in the development and validation datasets, respectively. The radiomic nomogram, incorporating the radiomic signature and two useful CT predictors, showed good calibration and discrimination in the development and validation datasets (AUC, 0·891 and 0·863, respectively). The decision curve analysis indicated that the nomogram was clinically useful. Interpretation: The combined radiomic nomogram has the potential to be used as a non-invasive tool for individualized preoperative prediction of ISPN. It would facilitate decision-making and improve the management of ISPN in clinical setting. Funding Statement: This work was funded by the National Health and Family Planning Commission of the People‘s Republic of China (201402013). Declaration of Interests: The authors have no conflicts of interest to declare. Ethics Approval Statement: This retrospective study was approved by our institutional review board and the requirement for informed consent was waived.