Prediction Of Histopathological Grade And Ki-67 Index Expression Of Myxoid Soft Tissue Sarcoma Based On MRI Radiomics

Objective: To study the value of MRI imaging features in predicting histopathological grade and Ki-67 index expression of myxoid soft tissue sarcoma,and to explore the relationship between MRI imaging features and myxoid soft tissue sarcoma grade and Ki-67 index expression.Methods : Collect 40 cases of myxoid soft tissue sarcoma confirmed by pathology in our hospital from January 2013 to August 2020,Grade I 12 cases,Grade II 14,Grade III 14,GE 1.5T MR scanner was used in all cases to obtain T1 WI enhanced scanning images,manually segment the lesions on the enhanced scanning images layer by layer T1 WI and extract imaging features,Based on histopathological grading dimensionality reduction,9 imaging labels were screened,and 8 features were screened based on Ki-67 expression dimensionality reduction.The patients were randomly assigned to the training set and the verification set,with a ratio of 2:1.The diagnostic model of soft tissue sarcoma grading / Ki-67 expression was established in the training set by logistic regression analysis,and verified in the verification set.The classification model of soft tissue sarcoma and the expression model of Ki-67 index were drawn ROC and the area under the ROC(AUC)was calculated to compare the diagnostic efficacy of each model.Results:According to the image label,the diagnostic efficiency of the model is better and the fitting degree of the model is better.Conclusion: The MRI imaging model has great value in predicting histopathological grade and Ki-67 index expression of myxoid soft tissue sarcoma.