Prediction Models for Solitary Pulmonary Nodules Based on Curvelet Textural Features and Clinical Parameters

Abstract

Lung cancer, one of the leading causes of cancer-related deaths, usually appears as solitary pulmonarynodules (SPNs) which are hard to diagnose using the naked eye. In this paper, curvelet-based textural featuresand clinical parameters are used with three prediction models [a multilevel model, a least absolute shrinkage andselection operator (LASSO) regression method, and a support vector machine (SVM)] to improve the diagnosisof benign and malignant SPNs. Dimensionality reduction of the original curvelet-based textural features wasachieved using principal component analysis. In addition, non-conditional logistical regression was used tofind clinical predictors among demographic parameters and morphological features. The results showed that,combined with 11 clinical predictors, the accuracy rates using 12 principal components were higher than thoseusing the original curvelet-based textural features. To evaluate the models, 10-fold cross validation and backsubstitution were applied. The results obtained, respectively, were 0.8549 and 0.9221 for the LASSO method,0.9443 and 0.9831 for SVM, and 0.8722 and 0.9722 for the multilevel model. All in all, it was found that usingcurvelet-based textural features after dimensionality reduction and using clinical predictors, the highest accuracyrate was achieved with SVM. The method may be used as an auxiliary tool to differentiate between benign andmalignant SPNs in CT images.