The Feasibility Study Of CT Radiomics Signature Of Pathological Grading And Early Recurrence In Hepatocellular Carcinoma Patients

Part Ⅰ The Feasibility Study of Predicting the Grade of Hepatocellular Carcinoma Based on CT Radiomics SignaturePurpose:To investigate the value of CT-based radiomics signatures to preoperative predict the grade of hepatocellular carcinoma(HCC).Materials and Methods:Materials and Methods:There were 429 patients who confirmed as HCC by surgical pathology were divided into training set(n=329)and validation set(n=100).Radiomics features based on arterial-and portal venous-phase CT images were extracted,and radiomics signatures were generated using the least absolute shrinkage and selection operator(LASSO)logistic regression model.The prediction performances of pathological grade in HCC using radiomics signature,clinical factors,and the combined models were assessed by AUCs.Results:Radiomics signatures could successfully categorize high-grade and low-grade HCC(p<0.001)in both training and test data-sets.Regarding the performance of radiomics signatures,clinical factors,and the combined model(from the combined arterial-and portal venous-phase)for HCC grading prediction,the areas under the curve(AUCs)were 0.815,0.691,and 0.833 in the training data-sets,respectively,and 0.694,0.626,and 0.703 in the test data-sets,respectively.In clinical model,patient’s age,tumor size and AFP level were independent predictors.Both the AFP level and radiomics signature were independent predictors of HCC grade(p<0.05).Conclusions:The radiomics signature is an important factor for discriminating high-grade and low-grade HCC.The combination of the radiomics signature with clinical factors performed the best among the three predictive models for preoperative HCC grading.Part Ⅱ The Feasibility Study of CT radiomics in prediction of early recurrence in hepatocellular carcinomaPurpose:To appraise the ability of the CT radiomics signature for prediction of early recurrence in patients with hepatocellar carcinoma(HCC).Materials and Methods:A set of 325 HCC patients were enrolled in this retrospective study and the whole dataset was divided into 2 cohorts,including:"training set"(225 patients)and "test set"(100 patients).All patients who underwent partial hepatectomy were followed up at least within 1 year.656 radiomics features were extracted from arterial-and portal venous-phase CT images.Lasso regression model was used for data dimension reduction,feature selection,and radiomics signature building.Univariate analysis was used to identify clinical and radiomics significant features.Models(radiomics signature,clinical model,and combined model)were evaluated by area under the curve(AUC)of receiver operating characteristic(ROC)curve.The models’performances for prediction of early recurrence were assessed.Results:The radiomics signature was built by 14 selected radiomics features and was significantly associated with early recurrence(P<0.001),the AUCs of the "train set"and the "test set" were 0.818(95%CI:0.760-0.865)and 0.719(95%CI:0.621-0.805),respectively.The tumor size,tumor capsule and gamma-glutamyl transferase(GGT)were significantly associated with early recurrence in the clinical model(P<0.05).The combined model showed incremental prognostic value,with the AUCs of"training dataset" and "test dataset" were 0.846(95%CI:0.792-0.890)and 0.737(95%CI:0.640-0.820),respectively.The radiomics signature,tumor size and the level of gamma-glutamyl transferase were independent predictors of early recurrence(P<0.05).Conclusions:The CT radiomics signature can be conveniently used to predict the early recurrence in patient with HCC.The combined model performed better for prediction of early recurrence than radiomics signature or clinical model.