Radiomic-based machine learning model for predicting the surgical risk in children with abdominal neuroblastoma.

Abstract

Background: Preoperative imaging assessment of surgical risk is very important for the prognosis of these children. To develop and validate a radiomics-based machine learning model based on the analysis of radiomics features to predict surgical risk in children with abdominal neuroblastoma (NB).

Methods: A retrospective study was conducted from April 2019 to March 2021 among 74 children with abdominal NB. A total of 1874 radiomic features in MR images were extracted from each patient. Support vector machines (SVMs) were used to establish the model. Eighty percent of the data were used as the training set to optimize the model, and 20% of the data were used to validate its accuracy, sensitivity, specificity and area under the curve (AUC) to verify its effectiveness.

Results: Among the 74 children with abdominal NB, 55 (65%) had surgical risk and 19 (35%) had no surgical risk. A t test and Lasso identified that 28 radiomic features were associated with surgical risk. After developing an SVM-based model using these features, predictions were made about whether children with abdominal NB had surgical risk. The model achieved an AUC of 0.94 (a sensitivity of 0.83 and a specificity of 0.80) with 0.890 accuracy in the training set and an AUC of 0.81 (a sensitivity of 0.73 and a specificity of 0.82) with 0.838 accuracy in the test set.

Conclusions: Radiomics and machine learning can be used to predict the surgical risk in children with abdominal NB. The model based on 28 radiomic features established by SVM showed good diagnostic efficiency.