【作者】 张坤；

【导师】 刘明辉；

【作者基本信息】 中南大学 ， 临床医学（专业学位）， 2024， 硕士

【摘要】 目的:本研究旨在构建预测附件肿块良恶性的列线图(Nomogram)模型,并比较其与RMI4模型及O-RADS模型的诊断效能。方法:收集2017年1月至2020年10月在我院接受附件肿块切除手术的患者700例,包括750例附件肿块。采用分层随机抽样划分训练集与验证集。记录所有患者的临床信息、附件肿块的超声图像以及超声特征。采用单因素及多因素logistic回归筛选独立危险因素并构建Nomogram模型,并将该模型与RMI4及O-RADS进行比较。结果:超声得分、绝经状态、肿块最大径、CA125升高、HE4升高是预测附件肿块良恶性的独立危险因素。在训练集中,Nomogram模型的ROC曲线下面积(AUC)为0.918(0.889-0.947),灵敏度和特异度分别为85.6%、85.8%。在验证集中,Nomogram模型的AUC为0.922(0.881-0.963),灵敏度和特异度分别为78.0%、89.2%。此外,Nomogram模型在训练集和验证集中均显示出良好的校准效率。DCA曲线表明Nomogram模型具有临床实用性。与RMI4和O-RADS相比,Nomogram模型具有最高的AUC和净效益。结论:Nomogram模型应用简单,诊断效能明显优于RMI4模型,与0-RADS模型诊断效能相当,能够弥补超声医师诊断经验不足,提高诊断准确性,并有望成为更加易于推广应用的附件肿瘤良恶性预测模型。图7幅,表7个,参考文献58篇更多还原

【Abstract】 Objective:The aim of this study is to establish a Nomogram model for predicting the benign and malignant nature of adnexal masses and to compare its diagnostic efficacy with the RMI4 and the O-RADS.Method:We collected 700 cases of patients who underwent appendage mass resection surgery in our hospital from January 2017 to October 2020,including 750 appendage masses.A stratified random sampling method was used to divide the training set and the validation set.The clinical information,ultrasound images,and ultrasound characteristics of all patients were recorded.Univariate and multivariate logistic regression were used to screen independent risk factors and construct a nomogram model.The model was compared with RMI4 and O-RADS.Results:Ultrasound score,menopausal status,maximum diameter of the mass,elevated CA125,and elevated HE4 were independent risk factors for predicting the benign or malignant nature of adnexal masses.In the training set,the area under the ROC curve(AUC)of the Nomogram model was 0.918(0.889-0.947),with sensitivity and specificity of 85.6%and 85.8%,respectively.In the validation set,the AUC of the Nomogram model was 0.922(0.881-0.963),with sensitivity and specificity of 78.0%and 89.2%,respectively.In addition,the Nomogram model showed good calibration efficiency in both the training set and the validation set.The DCA curve showed that the nomogram model had clinical utility.Compared with RMI4 and O-RADS,the Nomogram model had the highest AUC and net benefit.Conclusion:The Nomogram model is simple to apply and its diagnostic efficiency is significantly superior to the RMI4 model,while being comparable to the O-RADS model.It can compensate for the lack of diagnostic experience among ultrasound physicians,improve diagnostic accuracy,and has the potential to become a more easily applicable and widely promoted model for predicting the benign and malignant nature of adnexal tumors.更多还原