Computed Tomography Radiomics for the Preoperative Prediction of Cervical Lymph Node Metastasis in Papillary Thyroid Carcinoma: Development and External Validation
Zhi-Jiang Han,Peiying Wei,Zhongxiang Ding,Dingcun Luo,Liping Qian,Jian Zhou,Haibin Wang,Wenli Cai,Ming Zhang
DOI: https://doi.org/10.21203/rs.3.rs-26914/v1
2020-05-14
Abstract:Abstract Background Cervical lymph node (LN) status is a critical factor related to the treatment and prognosis of papillary thyroid carcinoma (PTC). The aim of this study was to investigate the preoperative predictions of cervical LN metastasis in PTC using computed tomography (CT) radiomics. Methods A total of 134 PTC patients who underwent CT examinations were enrolled in the study at two institutes between January 2018 and January 2020. Of these patients, 289 cervical LNs (institute 1: 206 LNs from 88 patients; institute 2: 83 LNs from 46 patents) were selected. All the cases had been confirmed by surgery and pathology. Each LN was segmented and 1408 radiomic features were calculated radiomic features in noncontrast and contrast-enhanced CT images. Features were selected using the Boruta algorithm followed by an iterative culling-out algorithm. We compared four machine learning classifiers, including random forest (RF), support vector machine (SVM), neural network (NN), and naïve bayes (NB) for the classification of LN metastasis. The models were first trained and validated by 10-fold cross-validation using data from institute 1 and then tested using independent data from institute 2. The performance of the models was compared using the area under the receiver operating characteristic curves (AUC). Results Seven radiomic features were selected for building the models − 3 histogram statistical textures, 1 gray level co-occurrence matrix texture, and 3 gray level zone size matrix textures. The AUCs of the radiomic models with 10-fold cross-validation were 0.941 (95% confidence interval [CI]: 0.93–0.95), 0.943 (95% CI: 0.93–0.95), 0.914 (95% CI: 0.90–0.95), and 0.905 (95% CI: 0.88–0.91) for RF, SVM, NN, and NB, respectively. The AUCs for the testing data were 0.926 (95% CI: 0.86–0.98), 0.932 (95% CI: 0.88–0.98), 0.925 (95% CI: 0.86–0.97), and 0.912 (95% CI: 0.83–0.98) for RF, SVM, NN, and NB, respectively. Conclusions CT radiomic model demonstrated robustness in preoperative classification of LN metastases for patients with PTC, which may provide significant support for clinical decision making and prognosis evaluation.