基于常规超声和三维能量多普勒的列线图预测宫颈癌前病变的临床价值  

Clinical value of nomogram based on conventional ultrasound and three-dimensional power Doppler in predicting cervical precancerous lesions

作　　者：温道浩 赵鹏 王乐 周瑾 王映人 WEN Daohao;ZHAO Peng;WANG Le;ZHOU Jin;WANG Yingren(Department of Ultrasound,the Affiliated Hospital of Xuzhou Medical University,Huai’an 223000,China;Department of Imaging,Huai’an NO.4 People’s Hospital,Huai’an 223000,China)

机构地区：[1]徐州医科大学附属淮安医院超声科,江苏淮安223000 [2]江苏省淮安市第四人民医院影像科,江苏淮安223000

出　　处：《临床超声医学杂志》2025年第3期215-220,共6页Journal of Clinical Ultrasound in Medicine

基　　金：淮安市科技发展计划项目(CSWS2022-23)。

摘　　要：目的基于常规超声和三维能量多普勒构建预测宫颈癌前病变的列线图模型,探讨其临床应用价值。方法选取于我院就诊的77例宫颈上皮内瘤变患者(研究组)和同期健康体检者80例(对照组),均行常规超声和三维能量多普勒检查,比较两组一般资料和超声检查结果的差异。采用多因素Logistic回归分析筛选预测宫颈癌前病变的独立影响因素,并构建列线图模型。绘制受试者工作特征(ROC)曲线、校准曲线分别评估列线图模型预测宫颈癌前病变的诊断效能和校准度。结果两组妊娠次数、流产次数、宫颈癌家族史、宫颈形态异常占比、宫颈管黏膜异常占比、宫颈血管形状异常占比,以及血流速度、血管指数(VI)、血流指数(FI)、宫颈血管血流指数(VFI)比较差异均有统计学意义(均P<0.001)。多因素Logistic回归分析显示流产次数,以及宫颈形态、宫颈血管形状、血流速度、VI、FI、VFI均为预测宫颈癌前病变的独立影响因素(均P<0.05),由此构建列线图模型。ROC曲线分析显示,列线图模型预测宫颈癌前病变的曲线下面积为0.987,高于各独立影响因素单独应用,差异均有统计学意义(均P<0.05)。校准曲线分析显示,列线图模型预测宫颈癌前病变的校准度较好。结论基于常规超声和三维能量多普勒构建的列线图模型在预测宫颈癌前病变中具有较高的临床应用价值,有助于临床早期筛查高风险人群。Objective To construct nomogram model based on conventional ultrasound and three-dimensional power Doppler in predicting cervical precancerous lesions,and to explore its clinical application value.Methods A total of 77 patients with cervical intraepithelial neoplasia(study group)and 80 healthy subjects(control group)in our hospital were enrolled.All participants underwent conventional ultrasound and three-dimensional power Doppler examinations.The differences in clinical characteristics and ultrasound examination results between the two groups were compared.Independent influencing factors were identified by multivariate Logistic regression analysis,and a nomogram model was constructed.The diagnostic performance and calibration of the model were assessed by receiver operating characteristic(ROC)curve,calibration curve,respectively.Results There were statistically significant differences in gravidity,abortion history,family history of cervical cancer,proportion of abnormal cervical morphology,abnormal cervical mucosa,abnormal vascular shape,blood flow velocity,vascular index(VI),flow index(FI),and vascular flow index(VFI)between the two groups(all P<0.001).Multivariate Logistic regression analysis showed that abortion times,as well as cervical morphology,vascular shape,blood flow velocity,VI,FI,and VFI were all independent influencing factors for predicting cervical precancerous lesions(all P<0.05).A nomogram model was constructed based on these influencing factors.ROC curve analysis showed that the area under the curve(AUC)of the nomogram model for predicting cervical precancerous lesions was 0.987,which was higher than that of each independent influencing factor alone,with statistically significant differences(all P<0.05).Calibration curve analysis showed that the nomogram model had good calibration for predicting cervical precancerous lesions.Conclusion The nomogram model based on conventional ultrasound and three-dimensional power Doppler parameters has high clinical application value in predicting cervical pre

关 键 词：超声检查 能量多普勒 三维 宫颈癌前病变 列线图模型 

分 类 号：R445.1[医药卫生—影像医学与核医学] R737.33[医药卫生—诊断学]