Objective: This study aimed to explore and validate the application value of CT radiomics in preoperative risk stratification for pediatric malignant peripheral neuroblastoma tumors (pNTs). Methods: Clinical data from 160 pediatric patients with malignant pNTs admitted to our hospital between December 2016 and December 2024 were collected and analyzed. The cohort was divided into a training group (n=112) and a testing group (n=48). Risk stratification was determined according to the Children"s Oncology Group (COG) criteria, classifying patients into high-risk and non-high-risk groups (intermediate-risk + low-risk). Imaging features were extracted from non-contrast CT scans and contrast-enhanced arterial and venous phase images. After dimensionality reduction and feature selection, a radiomics model was constructed using logistic regression (LR). Univariate and multivariate LR analyses of clinical and imaging features identified variables for establishing clinical models. Imaging and clinical features were integrated to construct combined models. Model performance was evaluated using receiver operating characteristic (ROC) curves, and clinical utility was assessed via decision curve analysis (DCA). Results: A total of 3,591 radiomics features were extracted from regions of interest (ROIs) in plain CT, enhanced arterial phase, and venous phase images. Ultimately, seven valuable radiomics features were selected. The radiomics model (R-score) constructed using LR achieved an area under the curve (AUC) of 0.843 in the training group and 0.820 in the testing group. A clinical model was constructed based on the selected clinical features. The combined model demonstrated superior discriminatory ability compared to both the clinical model and the radiomics model (training group AUC = 0.854, testing group AUC = 0.859). Decision curve analysis (DCA) indicated optimal clinical net benefit within a specific threshold range. Conclusion: The combined model integrating pediatric clinical features and radiomics demonstrates strong discriminatory ability for COG risk stratification in pediatric malignant pNTs, providing a noninvasive and effective approach for pre-treatment decision-making.