AUTHOR=Cheng Meiying , Tan Shifang , Ren Tian , Zhu Zitao , Wang Kaiyu , Zhang Lingjie , Meng Lingsong , Yang Xuhong , Pan Teng , Yang Zhexuan , Zhao Xin 

TITLE=Magnetic resonance imaging radiomics to differentiate ovarian sex cord-stromal tumors and primary epithelial ovarian cancers

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2023

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.1073983

DOI=10.3389/fonc.2022.1073983

ISSN=2234-943X

ABSTRACT=Objective To evaluate the diagnostic ability of magnetic resonance imaging (MRI) based radiomics and traditional characteristics to differentiate between Ovarian sex cord-stromal tumors (SCSTs) and epithelial ovarian cancers (EOCs). 
Methods We consecutively included a total of 148 patients with 173 tumors (81 SCSTs in 73 patients and 92 EOCs in 75 patients), who were randomly divided into development and testing cohorts at a ratio of 8:2. Radiomics features were extracted from the entire tumor of each patient, we built radiomics model based on the selected radiomics features. Univariate and multivariate analysis identified the independent clinical features and conventional MR parameters predictors for SCSTs and EOCs, with nomogram visualized the ultimately predictive models. All models were constructed using the logistic regression (LR) classifier. The performance of each model was evaluated by the receiver operating characteristic (ROC) curve. Calibration and decision curves analysis (DCA) were used to evaluate the performance of nomogram.
Results Fifteen radiomics signature were finally selected to construct the radiomics model. The radiomics model exhibited superior predictive ability, with AUCs of 0.923 ± 0.021, 0.837 ± 0.094 and 0.878 (95% CI: 764-0.992) in the training, validation and testing cohorts, respectively. And when radiomics was combined with traditional parameters, the diagnostic efficacy was further improved, with AUCs of 0.939 ± 0.012， 0.869 ± 0.059 and 0.902 (95% CI: 0.804-1.000) in the training, validation and testing cohorts, respectively.
Conclusion We believe that the radiomics approach could be a more objective and accurate way to distinguish between SCSTs and EOCs, and the mixed model developed in our study could provide a comprehensive, effective method for clinicians to develop an appropriate management strategy.
Keywords Ovarian sex cord-stromal tumor, Epithelial ovarian cancer, Magnetic resonance imaging, Radiomics, Prediction model