The Clinical Value Of Enhanced CT Radiomics And 3D Texture Analysis In FIGO Staging And BRCA Gene Mutation Of Epithelial Ovarian Cancer

The clinical value of enhanced CT imaging in the staging of serous ovarian cancerObjectiveTo investigate the preoperative diagnostic value of abdominal and pelvic enhanced CT imaging in serous ovarian FIGO staging.MethodsPreoperative CT data of 137 patients with ovarian cancer in Anhui Provincial Hospital from June 2018 to November 2019 were retrospectively analyzed.Serous ovarian cancer was confirmed by surgery and pathology,including 68 patients with FIGO stage I-II and 69 patients with FIGO stage III-IV.The Region Of Intrest（ROI）was delineated for preoperative CT images Of each patient（including arterial and venous phases）.The main drawing software used in this study was ITK-SNAP software,which adjusted the images to the same window width and window position before drawing.A.K.software was used to automatically extract image omics features,and mann-Whitney U test + single-factor logistic regression analysis+m RMR（Maximum Relevance Minimum Redundancy）was used to remove redundant features and retain the features with the highest Relevance.Furthermore,multivariate logistic regression analysis was used to find independent predictors and establish prediction models,and Receiver Operating Characteristic Curve（ROC）was used to evaluate the diagnostic performance of the models.The arterial and venous data were divided into training sets and validation sets with a ratio of 7:3.The stability of the model established in this study mainly depended on100 repeated cross-validation.ResultsThe prediction efficiency of the model is good.The accuracy,specificity and sensitivity of preoperative FIGO stage diagnosis for serous ovarian cancer were 0.85,0.91,0.80 in arterial stage and 0.88,0.77,0.99 in venous stage,respectively.The diagnostic efficiency of the model was mainly measured by Area Under the ROC Curve（AUC）,of which,arterial phase: 0.91（95%CI:0.86-0.96）,venous phase: 0.93（95%CI:0.89-0.97）,indicating that image-based staging had a strong coincidence with clinical FIGO staging and high diagnostic value.ConclusionEnhanced CT imaging omics has an important clinical reference value for preoperative FIGO staging evaluation of serous ovarian cancer,which can effectively guide the preoperative evaluation and provide important clinical information for the multidisciplinary surgical team that needs to participate in the operation.Conventional CT imaging features combined with 3D texture analysis predict BRCA mutation status in advanced epithelial ovarian cancerObjectiveTo evaluate the predictive value of routine CT features combined with 3D texture analysis for prediction of BRCA gene mutation status in advanced epithelial ovarian cancer.MethodsRetrospective analysis was performed on patients with masses occupying the pelvic space confifirmed by pathology and complete preoperative images in our hospital,including 37 and 58 cases with mutant type and wild type BRCA,respectively（total: 95 cases）.The enrolled patients’ routine CT features were analyzed by two radiologists.Then,ROIs were jointly determined through negotiation,and the ITK-SNAP software package was used for3 D outlining of the third-stage images of the primary tumor lesions and obtaining texture features.For routine CT features and texture features,Mann Whitney U tests,single-factor logistic regression analysis,minimum redundancy,and maximum correlation were used for feature screening,and the performance of individual features was evaluated by ROC curves.Multivariate logistic regression analysis was used to further screen features,fifind independent predictors,and establish the prediction model.The established model’s diagnostic effificiency was evaluated by ROC curve analysis,and the histogram was obtained to conduct visual analysis of the prediction model.ResultsAmong the routine CT features,the type of peritoneal metastasis,mesenteric involvement,and supradiaphragmatic lymph node enlargement were correlated with BRCA gene mutation（P < 0.05）,whereas the location of the peritoneal metastasis（in the astrohepatic ligament）was not signifificantly correlated with BRCA gene mutation（P > 0.05）.Multivariate logistic regression analysis retained six features,including one routine CT feature and fifive texture features.Among them,the type of peritoneal metastasis was used as an independent predictor（P < 0.05）,which had the highest diagnostic effificiency.Its AUC,accuracy,specifificity,and sensitivity were 0.74,0.79,0.90,and 0.62,respectively.The prediction model based on the combination of routine CT features and texture features had an AUC of 0.86（95% CI: 0.79–0.94）and accuracy,specifificity,and sensitivity of 0.80,0.76,and 0.81,respectively,indicating a better performance than that of any single feature.ConclusionBoth routine CT features and texture features had value for predicting the mutation state of the BRCA gene,but their predictive effificiency was low.When the two types of features were combined to establish a predictive model,the model’s predictive effificiency was signifificantly higher than that of independent features.