MRI Features and Diagnostic Value of Diffusion-Weighted Imaging in Pediatric Ovarian Tumors: A Multicenter Retrospective Study

Background Ovarian tumors in children and adolescents represent a diagnostically challenging and heterogeneous group of diseases. Accurate preoperative differentiation between benign and malignant lesions is essential for optimal treatment planning and preservation of reproductive potential. Magnetic resonance imaging (MRI) plays a key role in the evaluation of indeterminate adnexal masses; however, conventional morphological MRI features often overlap between benign and malignant tumors. Diffusion-weighted imaging (DWI) may provide additional diagnostic information by reflecting tumor cellularity. Methods This multicenter retrospective observational study included pediatric patients with suspected ovarian tumors examined at two tertiary referral centers between 2020 and 2025. A total of 152 patients were evaluated, of whom 96 underwent pelvic MRI. Histopathological confirmation of diagnosis was available for all patients, while diffusion-weighted imaging with apparent diffusion coefficient (ADC) mapping was performed in 32 cases. MRI assessment included analysis of tumor morphology, presence of solid components, contour characteristics, extracapsular extension, ascites, lymphadenopathy, and peritoneal deposits. Diagnostic performance of MRI and DWI was assessed using receiver operating characteristic (ROC) analysis. Results Several morphological MRI features, including solid components, irregular margins, extracapsular growth, ascites, lymphadenopathy, and peritoneal deposits, were significantly more frequent in malignant tumors (p < 0.001). Tumor size and cystic–solid architecture alone were not reliable indicators of malignancy. ADC values of the solid tumor component were significantly lower in malignant compared with benign lesions (0.8 ± 0.09 ×10⁻³ mm²/s vs. 1.2 ± 0.13 ×10⁻³ mm²/s, p = 0.019). ROC analysis demonstrated high diagnostic accuracy of MRI based on morphological features (AUC = 0.94). The combined use of morphological MRI and DWI improved diagnostic performance, with a sensitivity of 91.0%, specificity of 88.0%, and overall accuracy of 89.0%. Conclusions MRI is a highly informative modality for the evaluation of ovarian tumors in children and adolescents; however, conventional morphological criteria alone may be insufficient, particularly in early-stage malignant epithelial tumors. Diffusion-weighted imaging with quantitative ADC assessment provides significant additional diagnostic value and improves differentiation between benign and malignant ovarian tumors. A comprehensive MRI approach incorporating both morphological and functional parameters may enhance diagnostic confidence and support optimal clinical decision-making in pediatric patients.