首页|期刊导航|武警医学|早期子宫内膜癌风险分层列线图预测模型的构建及验证

早期子宫内膜癌风险分层列线图预测模型的构建及验证OA

Construction and validation of a nomogram prediction model for risk stratification of ear-ly-stage endometrial cancer

中文摘要英文摘要

目的 探讨基于临床指标联合多序列 MRI 征象构建的列线图模型在预测早期子宫内膜癌(EC)风险分层的临床价值.方法 回顾性分析 2014-07 至 2025-07 于武警四川总队医院确诊为早期 EC 的 150 例患者的临床及 MRI 资料.按7∶3 的比例随机分为训练集和测试集,比较训练集中低风险组与非低风险组两组间临床指标及 MRI 征象的差异.对训练集的各变量进行 logistic 回归分析,确定预测因子,构建列线图模型,使用受试者操作特征(ROC)曲线下面积(AUC)、校准曲线及决策曲线评估列线图的预测效能.结果 训练集纳入 105 例,测试集 45 例.训练集中肿瘤位置、前后径、宫角受累、肿瘤相对肌层 ADC 值比、动脉期内膜与肌层间强化带、动脉期病灶边缘强化为预测非低风险组早期 EC 的预测因子.由以上 6 个因素构建的列线图模型在训练集和测试集 ROC 曲线 AUC(95%CI)分别为 0.979(0.957,1.000)、0.934(0.867,1.000).训练集和测试集校准曲线拟合度较好,决策曲线分析显示列线图具有较好的临床实用性.结论 临床指标和 MRI 征象构建的列线图在预测早期子宫内膜癌风险分层中具有较高的准确性,对辅助临床制定治疗方案具有重要价值.

Objective To investigate the clinical value of a nomogram model based on clinical indicators and multi-sequence MRI features in predicting the risk stratification of early-stage endometrial cancer(EC).Methods A retrospective analysis was con-ducted on the clinical and MRI data of 150 patients diagnosed with early-stage EC in Sichuan Provincial Corps Hospital of Chinese Peo-ple's Armed Police Force from July 2014 to July 2025.The patients were randomly divided into a training set and a test one at a 7∶3 ratio.The differences in clinical indicators and MRI features between the low-risk and non-low-risk groups in the training set were illus-trated.Logistic regression analysis was employed on the variables in the training set to identify the predictors,which were then used to construct a nomogram model.The predictive performance of the nomogram was evaluated using the area under the receiver operating characteristic curve(AUC),calibration curves,and decision curve analysis.Results A total of 105 patients were assigned to the training set,and 45 to the test one.Tumor location,anteroposterior diameter,uterine corner involvement,relative ADC value ratio of tumor to myometrium,arterial phase enhancement zone between the intima and muscularis,and arterial phase lesion edge enhancement were identified as predictors for non-low-risk early-stage EC in the training set.The area under the ROC curve(AUC)for the nomo-gram model constructed from these six factors was 0.979(0.957,1.000)for the training set and 0.934(0.867,1.000)for the test one respectively.The calibration curves demonstrated good fit in both sets,and the decision curve analysis indicated favorable clinical utility of the nomogram.Conclusions The nomogram constructed using clinical indicators and MRI features exhibits high accuracy in predicting the risk stratification of early-stage EC and holds significant value in guiding clinical treatment decisions.

朱源义;方永昆;关鉴;徐丽;郑昊宇;向科;张巍

614000 乐山,武警四川总队医院放射科614000 乐山,武警四川总队医院放射科614000 乐山,武警四川总队医院放射科614000 乐山,武警四川总队医院放射科614000 乐山,武警四川总队医院放射科614000 乐山,武警四川总队医院放射科614000 乐山,武警四川总队医院放射科

医药卫生

子宫内膜癌磁共振成像列线图预测风险分层

endometrial carcinomamagnetic resonance imagingnomogrampredictionlevel of risk

《武警医学》 2026 (8)

655-663,9

10.3969/j.issn.1004-3594.2026.08.002

评论