Using T2-Weighted MR Imaging-Derived Radiomics to Classify Cervical Lymphadenopathy in Children

Abstract

Background: Cervical lymphadenopathy is common in children with diverse causes varying from benign to malignant, their similar manifestations make differential diagnosis difficult. This study intended to investigate whether radiomic models using conventional magnetic resonance imaging (MRI) could classify pediatric cervical lymphadenopathy. Methods: A total of 419 cervical lymph nodes (LNs) from 146 patients, belonging to four common causes (kikuchi disease, reactive hyperplasia, suppurative lymphadenitis and malignancy), were randomly divided into training and testing sets at a ratio of 7:3. For each LN, 1218 features were extracted from T2-weighted images. Then the least absolute shrinkage and selection operator (LASSO) model were used to select the most relevant ones. Two models were built using support vector machine classifier, one was to classify benign and malignant LNs and the other further distinguished four different diseases. The performance was assessed by receiver operating characteristic curves and decision curve analysis. Results: By LASSO, 20 features were selected to construct a model to distinguish benign and malignant LNs, which achieved an area under the curve (AUC) of 0.89 and 0.80 in the training and testing set respectively. Sixteen features were selected to construct a model to distinguish four different cervical lymphadenopathies. For each etiology (kikuchi disease, reactive hyperplasia, suppurative lymphadenitis and malignancy respectively), an AUC of 0.97, 0.91, 0.88 and 0.87 was achieved in the training set, and an AUC of 0.96, 0.80, 0.82 and 0.82 was achieved in the testing set. Conclusion: MRI-derived radiomic analysis provides a promising noninvasive approach for distinguishing cervical lymphadenopathy in children.