超声特征联合血清学指标构建子宫内膜癌淋巴结转移预测模型
Construction of a lymph node metastasis prediction model for endometrial cancer using ultrasound features combined with serological indicators
张萌 1姜伊娜 2王春光 2刘禹彤1
作者信息
- 1. 山东第二医科大学医学影像学院 山东 潍坊 261053
- 2. 山东省潍坊市人民医院超声科 山东 潍坊 261041
- 折叠
摘要
目的 探讨子宫内膜癌超声特征联合血清学指标及其他临床特征,探寻子宫内膜癌患者发生淋巴结转移的危险因素,构建列线图并预测其发生.方法 选取我院经病理诊断为子宫内膜癌患者 499 例.按 7:3 的比例分为训练集组(n=356)与验证集组(n=143),采用单因素与多因素Logistic回归分析筛选子宫内膜癌淋巴结转移的危险因素,绘制列线图模型,通过受试者工作特征曲线(ROC)、校准曲线、临床决策曲线(DCA)对该模型进行评估.结果 单因素分析显示超声特征中的病变形状、边界、彩色多普勒血流分级,血清学指标(术前)中的糖类抗原 125(CA125)、组织学分级、组织学分型与子宫内膜癌淋巴结转移相关,通过多因素分析,彩色多普勒血流分级、CA125、组织学分级是子宫内膜癌淋巴结转移的独立危险因素,OR值分别为 1.75(95%CI 1.20～2.56)、4.61(95%CI 2.12～10.01)、1.93(95%CI 1.07～3.49),据此建立列线图,训练集组的曲线下面积(AUC)为 0.796(95%CI 0.718～0.874),验证集组AUC为 0.715(95%CI 0.538～0.892),区分度良好,通过校准曲线、DCA进行验证,该模型的准确性及临床应用前景良好.结论 结合超声检查特征与血清学指标成功地构建子宫内膜癌淋巴结状态的评估模型,该模型对评估淋巴结状态有较高的准确性,也为子宫内膜癌的诊断和治疗提供影像学依据.
Abstract
Objective To explore the risk factors for lymph node metastasis in patients with endometrial cancer,to construct a nomogram and to predict its occurrence by combining with ultrasonic features of endometrial cancer,serological indicators and other clinical features.Methods Four hundred and ninety-nine patients of endometrial cancer were divided into a training set(n=356)and a validation set(n=143)according to a ratio of 7:3.Univariate and multivariate logistic regression analysis was used to screen the risk factors for lymph node metastasis of endometrial cancer,and a nomogram model was drawn.The model was evaluated by receiver operating characteristic curve(ROC),calibration curve,and clinical decision curve(DCA).Results The shape,boundary,CDFI of the lesion in ultrasonic characteristics,CA125 in serologic indicators(preoperative),histological grade,and histological type were correlated with lymph node status of endometrial cancer.The OR values were 1.75(95%CI 1.2～2.56),4.61(95%CI 2.12～10.01),1.93(95%CI 1.07～3.49).The AUC of training set was 0.796(95%CI 0.718～0.874),the AUC of validation set was0.715(95%CI 0.538～0.892).The model was well distinguished and verified by calibration curve and clinical decision curve(DCA).The accuracy and clinical application prospect of the model were good.Conclusion An evaluation model of lymph node status of endometrial cancer is successfully constructed by combining ultrasonic characteris-tics and serological indexes.The model has high accuracy in evaluating lymph node status and provides more diagnosis and treat-ment ideas for endometrial cancer.
关键词
子宫内膜癌/淋巴结/预测模型/超声检查/糖类抗原125Key words
Endometrial cancer/Lymph node/Predicationmodel/Ultrasound/CA-125Antigen引用本文复制引用
出版年
2024
NETL
NSTL
万方数据