The Study Of Effectivness And Safety Of High-intensity Focused Ultrasound In Treatment Of Uterine Fibroids Based On Data Mining

Background:Uterine fibroids are the most common disease of women of childbearing age,accounting for about 52%of gynecological benign tumors.The traditional methods of treating uterine fibroids mainly include drug treatment and surgical treatment.In recent years,high-intensity focused ultrasound(HIFU)ablation as a non-invasive treatment method has achieved satisfactory results in the clinical treatment of uterine fibroids.A large number of clinical trials have confirmed its safety and effectiveness.The rapid development of Internet-related technologies has laid the foundation for data mining and analysis processing.Methods based on big data collection and analysis have gradually become routine and effective research methods in the medical industry.Objectie:Use data mining methods to further explore the impact of various factors in clinical application on the safety and effectiveness of HIFU treatment of uterine fibroids to improve clinical training and diagnosis and treatment.Methods:Analyze the data of uterine fibroids in Ronghai/Chongqing Medical University's"Twelfth Five-Year Plan"Uterine Fibroids Support Project.Screening was based on the exclusion criteria for single uterine fibroids.A total of 907 cases were included in the validity part.The regression analysis method in data mining technology was used to build a regression prediction model of the ablation rate using a Python program.The accuracy of each model was compared to find the best regression prediction model.A total of 552 cases were included in the safety part.The classification prediction method in data mining technology was used to build a classification prediction model of postoperative complications on Python software.The accuracy of each model was compared to find the best classification prediction model.Results:1.The evaluation score of the validity part support vector machine regression(SVR)model is better than the other four regression algorithms:its Explained Variance Score,Mean Absolute Error(MAE),and Mean Squared Error,MSE),the coefficient of determination(r~2 Score)were 0.77,0.08,0.01,0.71,and the average ablation rate of 907 myomas was(83.14±15.94)%.2.A total of 452 complications occurred in 911cases of safety.The main manifestations were lower abdominal pain,sacrococcygeal pain,and vaginal drainage.Research indicates,the logistic regression model had the best classification prediction effect on lower abdominal pain.The classification prediction of tail pain is better,and the secondary discriminant analysis model has the highest accuracy rate for vaginal drainage classification prediction.Conclusions:HIFU is safe and effective in treating uterine fibroids.Data mining technology can guide HIFU clinical training and treatment.It can effectively eliminate secondary angiography and reduce the probability of complications,and bring convenience to clinicians and patients.