MRI Features And Radiomic Based On Multi-parameter MRI For Predicting The Ablation Rate Of Uterine Fibroids

PART Ⅰ ANALYSIS OF THE INFLUENCING FACTORS ON THE ABLATION RATE OF UTERINE FIBROID BY USING HIGH INTENSITY FOCUSED ULTRASOUNDObjective: To explore the influencing factors of high intensity focused ultrasound(HIFU)in the treatment of uterine fibroid.Methods: Collected clinical data of 130 patients with uterine fibroid admitted to the HIFU Center of the First Affiliated Hospital of Chongqing Medical University from January 2013 to December 2018,including the patient’s age,volume of fibroids and mean meridian before treatment,fibroids location,fibroids type,MRI T2 WI signal intensity and uniformity,T1 WI enhancement intensity and uniformity,and uterine location,etc.The preoperative and postoperative MRI images were used to calculate the preoperative volume of the fibroids and the non-perfused area after the HIFU treatment to obtain the ablation rate.The indicators above were analyzed to find out the factors that affect the efficacy of HIFU ablation of uterine fibroids.Results: Of the 130 fibroids,70(53.8%)fibroids were sufficient ablated.Univariate analysis results showed that volume of fibroids,mean meridian of fibroids,T2 WI signal intensity and uniformity,TIWI enhancement degree and uterine position had a statistically significant effect on ablation(P <0.05).Multivariate logistic regression analysis showed that T2 WI signal intensity(P=0.035)and TIWI enhancement(P=0.038)of fibroids were more likely to achieve satisfactory ablation results.Conclusion: The volume of fibroids,the average meridian of fibroids,the intensity and uniformity of T2 WI signal,the degree of T1 WI enhancement,and the position of the uterus all have a certain effect on the efficacy of HIFU in treating uterine fibroids.This study has a certain predictive effect on predicting the efficacy of HIFU ablation of uterine fibroids.PART Ⅱ RADIOMIC BASED ON MULTI-PARAMETER MRI FOR PREDICTING THE ABLATION RATE OF UTERINE FIBROIDSObjective: To establish a radiomic model based on multi-parameter MRI,and to observe the value of the model for predicting the ablation rate of uterine fibroids.Methods: Data of MR imaging of 130 patients with uterine fibroids admitted to the HIFU center of the first affiliated hospital of Chongqing medical university from January 2013 to December 2018,were retrospectively analyzed.ITK-SNAP software was used by 2 radiologists to delineate ROI of the tumors,and radiomic parameters were extracted.Then multiple logistic regression was applied to identify optimal radiomic features and construct the prediction model.The prediction performance of the classification models was evaluated by the quantitative metrics derived from the receiver operating characteristic curve(ROC).Results: A total of 972 image radiomics parameters were extracted,and 18 feature parameters were obtained after dimensionality reduction.The above 18 feature parameters were used to construct several prediction models,the performance of SVM model is the best.The AUC,ACC,sensitivity,and specificity of the SVM model in the training set to predict the ablation rate of uterine fibroids are 0.91,0.884,0.857,and 0.921,respectively.Tested in the validation set,the AUC,ACC,sensitivity,and specificity obtained are 0.891,0.849,0.814 and 0.896,respectively.Conclusion: Radiomic model based on multi-parameter MRI can be used for predicting ablation rate of uterine fibroids.