
浏览全部资源
扫码关注微信
1.汕头大学医学院,汕头 515041
2.广东医科大学深圳宝安临床医学院放射科,深圳 518100
3.南方科技大学第一附属医院放射科,深圳 518020
林清秀,Email:527115161@qq.com,研究方向:妇科影像诊断。
凌人男,Email:670744652@qq.com。
收稿:2025-04-27,
修回:2025-07-21,
录用:2025-08-01,
纸质出版:2026-05-28
移动端阅览
林清秀,李金贝贝,何子坤等.基于增强T1WI影像组学及临床MRI特征组学列线图术前预测子宫内膜癌侵袭性组织类型的价值[J].重庆医科大学学报,2026,51(05):623-631.
Lin Qingxiu,Li Jinbeibei,He Zikun,et al.Predictive value of a nomogram based on enhanced T1WI radiomics and clinico-MRIomics for aggressive histological types of endometrial cancer before surgery[J].Journal of Chongqing Medical University,2026,51(05):623-631.
林清秀,李金贝贝,何子坤等.基于增强T1WI影像组学及临床MRI特征组学列线图术前预测子宫内膜癌侵袭性组织类型的价值[J].重庆医科大学学报,2026,51(05):623-631. DOI: 10.13406/j.cnki.cyxb.003958.
Lin Qingxiu,Li Jinbeibei,He Zikun,et al.Predictive value of a nomogram based on enhanced T1WI radiomics and clinico-MRIomics for aggressive histological types of endometrial cancer before surgery[J].Journal of Chongqing Medical University,2026,51(05):623-631. DOI: 10.13406/j.cnki.cyxb.003958.
目的
2
建立1个基于增强T1WI(contrast enhanced T1 weighted imaging,CE-T1WI)的影像组学和临床磁共振成像(magnetic resonance imaging,MRI)特征组学的列线图,评估其术前预测子宫内膜癌侵袭性组织类型的价值。
方法
2
回顾性分析南方科技大学第一附属医院2016年1月至2024年8月经手术病理诊断为子宫内膜癌患者192例,其中侵袭性组织学类型48例,按7∶3比例随机分为训练集(
n
=134)
和测试集(
n
=58)。采用ITK-SNAP软件提取影像组学特征,采用多因素分析及最小绝对收缩和选择算子(the least absolute shrinkage and selection operator,LASSO)对影像组学特征进行筛选。同时收集临床指标和MRI特征共11个参数,包括年龄、体质指数(body mass index,BMI)、糖类抗原199(carbohydrate antigen 199,CA199)、CA125、中性粒细胞与淋巴细胞比值(neutrophil-to-lymphocyte ratio,NLR)、表观弥散系数(apparent diffusion coefficient,ADC)、肿瘤与子宫肌层T2WI信号比值(RatioT
2
),肿瘤累及子宫肌层
>
1/2(Deep),肿瘤累及子宫肌层的深度(D
tra
),肿瘤的最大直径(D
max
)和增强后肿瘤与子宫肌层信号强度比值(Ratio
CE
),采用Spearman相关分析筛选MRI特征。基于影像组学和临床MRI特征组学分别构建LR、SVM、KNN、RF、Extra Trees、XGBoost、LightGBM、MLP模型,绘制受试者工作特征(receiver operating characteristic,ROC)曲线,利用ROC曲线筛选出最佳影像组学和临床MRI特征组学模型以绘制列线图,使用校准曲线评估校准度,使用决策曲线分析(decision curve analysis,DCA)评估应用价值。
结果
2
7个影像组学特征和5个临床MRI特征指标(年龄、CA199、ADC、Ratio
CE
、Deep)进入到组学模型。影像组学和临床MRI特征组学构建的LightGBM模型预测效能最佳,训练集和测试集中的AUC分别为0.889和0.788,临床MRI特征组学模型分别为0.893和0.835。构建的列线图在测试集中的AUC、准确率、敏感性、特异性分别为0.888、0.879、0.800和0.907,经校准曲线和DCA分析,列线图的校准度较高,临床净收益较大。
结论
2
基于CE-T1WI的影像组学和临床MRI特征组学构建的列线图,可以较好的术前评估子宫内膜癌侵袭性组织类型,对指导患者个体化治疗有着重要临床意义。
Objective
2
To evaluate the preoperative predictive value of a nomogram based on contrast-enhanced T1-weighted imaging (CE-T1WI) radiomics and clinical-MRIomics for aggressive histological types of endometrial cancer.
Methods
2
A total of 192 patients with endometrial cancer pathologically confirmed by surgery at The First Affiliated Hospital,Southern University of Science and Technology,from January 2016 to August 2024 were retrospectively analyzed,including 48 patients with aggressive histological types. The patients were randomly divided into a training set(
n
=134) and a test set(
n
=58) at a ratio of 7∶3. Radiomic features were extracted using ITK-SNAP,and feature selection was performed using multivariable analysis and least absolute shrinkage and selection operator(LASSO) regression.
Clinical and MRI features were screened using Spearman correlation analysis from 11 parameters,including age,body mass index(BMI),carbohydrate antigen 19-9(CA199),CA125,neutrophil-to-lymphocyte ratio(NLR),apparent diffusion coefficient(ADC),tumor-to-myometrium signal ratio on T2WI(RatioT
2
),myometrial invasion
>
1/2(Deep),depth of myometrial invasion(D
tra
),maximum tumor diameter(D
max
),and postcontrast tumor-to-myometrium signal intensity ratio(Ratio
CE
). Based on radiomics and clinical-MRIomics features,eight models were constructed,including logistic regression(LR),support vector machine(SVM),k-nearest neighbor(KNN),random forest(RF),Extra Trees,XGBoost,LightGBM,and multilayer perceptron(MLP). Receiver operating characteristic(ROC) curves were used to select the optimal radiomics and clinical-MRIomics models for constructing the nomogram. Calibration curves were used to evaluate calibration,and decision curve analysis(DCA) was used to assess clinical utility.
Results
2
Seven radiomic features and five clinical-MRI features,namely age,CA199,ADC,RatioCE,and Deep,were incorporated into the models. Among the radiomics and clinical-MRIomics models,the LightGBM model showed the best predictive performance. The AUCs of the radiomics model were 0.889 and 0.788 in the training and test sets,respectively,while those of the clinical-MRIomics model were 0.893 and 0.835,respectively. In the test set,the AUC,accuracy,sensitivity,and specificity of the nomogram were 0.888,0.879,0.800,and 0.907,respectively. Calibration curve and DCA showed that the nomogram had good calibration and a relatively high clinical net benefit.
Conclusion
2
The nomogram based on CE-T1WI radiomics and clinical-MRIomics can effectively predict aggressive histological types of endometrial cancer before surgery,providing guidance for individualized treatment.
宋艳 , 刘爱军 . 子宫内膜癌FIGO 2023分期的更新及其临床病理意义 [J ] . 中华妇产科杂志 , 2024 , 59 ( 3 ): 179 - 183 .
Song Y , Liu AJ . Updates of the FIGO 2023 staging for endometrial cancer and their clinicopathological significance [J ] . Chin J Obstet Gynecol , 2024 , 59 ( 3 ): 179 - 183 .
WHO Classification of Tumours Editorial Board . WHO classification of tumours:female genital tumours [M ] . 5th ed . Lyon : IARC Publications , 2020 : 252 - 267 .
Crosbie EJ , Kitson SJ , McAlpine JN , et al . Endometrial cancer [J ] . Lancet , 2022 , 399 ( 10333 ): 1412 - 1428 .
Berek JS , Matias-Guiu X , Creutzberg C , et al . FIGO staging of endometrial cancer:2023 [J ] . Int J Gynaecol Obstet , 2023 , 162 ( 2 ): 383 - 394 .
谢玲玲 , 林仲秋 . 2024 NCCN子宫肿瘤临床实践指南(第1版)解读 [J ] . 中国实用妇科与产科杂志 , 2023 , 39 ( 11 ): 1122 - 1127 .
Xie LL , Lin ZQ . Interpretation of the 2024 NCCN clinical practice guidelines in oncology for uterine neoplasms(Version 1) [J ] . Chin J Pract Gynecol Obste , 2023 , 39 ( 11 ): 1122 - 1127 .
王 娜 , 郭云峰 , 宋立芹 , 等 . HE4、CA125及CA199在鉴别子宫内膜癌肌层浸润深度及病理分期中的临床应用价值 [J ] . 重庆医科大学学报 , 2022 , 47 ( 10 ): 1220 - 1225 .
Wang N , Guo YF , Song LQ , et al . Clinical value of serum HE4,CA125 and CA199 in identifying myometrial invasion depth and pathological stage of endometrial carcinoma [J ] . J Chongqing Med Univ , 2022 , 47 ( 10 ): 1220 - 1225 .
黄雅洁 , 赵文学 , 刘珂 , 等 . 术前指标对子宫内膜癌淋巴脉管间隙浸润风险的影响 [J ] . 重庆医科大学学报 , 2023 , 48 ( 6 ): 703 - 707 .
Huang YJ , Zhao WX , Liu K , et al . Impact of preoperative indicators on the risk of lymphovascular space invasion in endometrial cancer [J ] . J Chongqing Med Univ , 2023 , 48 ( 6 ): 703 - 707 .
Yan B , Zhao TT , Deng Y , et al . Preoperative prediction of lymph node metastasis in endometrial cancer patients via an intratumoral and peritumoral multiparameter MRI radiomics nomogram [J ] . Front Oncol , 2024 , 14 : 1472892 .
韩瑛 , 刘崇东 . 子宫浆液性癌的起源及临床研究进展 [J ] . 中国实用妇科与产科杂志 , 2020 , 36 ( 8 ): 779 - 781 .
Han Y , Liu CD . Origin and clinical research progress of uterine serous carcinoma [J ] . Chin J Pract Gynecol Obstet , 2020 , 36 ( 8 ): 779 - 781 .
刘从容 , 刘 岩 . 子宫内膜癌传统组织病理与分子分型整合诊断的临床意义 [J ] . 中国实用妇科与产科杂志 , 2023 , 39 ( 11 ): 1073 - 1080 .
Liu CR , Liu Y . Clinical significance of integrated diagnosis based on traditional histopathology and molecular classification of endometrial carcinoma [J ] . Chin J Pract Gynecol Obstet , 2023 , 39 ( 11 ): 1073 - 1080 .
黄成香 , 李 翠 , 张海堂 , 等 . 子宫内膜癌肿瘤体积与子宫体积比值联合血清AFP、CA199、HE4的表达对病理分级和预后的评估价值 [J ] . 中华内分泌外科杂志(中英文) , 2025 , 19 ( 4 ): 589 - 594 .
Huang CX , Li C , Zhang HT , et al . Value of tumor volume to uterine volume ratio combined with serum AFP,CA199,HE4 expression in evaluating pathological grade and prognosis of endometrial carcinoma [J ] . Chin J Endocr Surg , 2025 , 19 ( 4 ): 589 - 594 .
林玲玲 , 程杰军 . 放射影像学检查在子宫内膜癌诊断中的价值 [J ] . 中国实用妇科与产科杂志 , 2023 , 39 ( 11 ): 1065 - 1070 .
Lin LL , Cheng JJ . Value of imaging examination in the diagnosis and treatment of endometrial cancer [J ] . Chin J Pract Gynecol Obstet , 2023 , 39 ( 11 ): 1065 - 1070 .
Kishimoto K , Tajima S , Maeda I , et al . Endometrial cancer:correlation of apparent diffusion coefficient(ADC) with tumor cellularity and tumor grade [J ] . Acta Radiol , 2016 , 57 ( 8 ): 1021 - 1028 .
Ling RN , Jin HT , Zhang H . The qualitative and quantitative characteristics of serous endometrial carcinoma on MRI:applying a novel nomogram for predicting an aggressive histological type [J ] . Front Oncol , 2025 , 15 : 1472250 .
Yue XN , He XY , He SJ , et al . Multiparametric magnetic resonance imaging-based radiomics nomogram for predicting tumor grade in endometrial cancer [J ] . Front Oncol , 2023 , 13 : 1081134 .
Moro F , Albanese M , Boldrini L , et al . Developing and validating ultrasound-based radiomics models for predicting high-risk endometrial cancer [J ] . Ultrasound Obstet Gynecol , 2022 , 60 ( 2 ): 256 - 268 .
Wang YD , Liu W , Lu YY , et al . Fully automated identification of lymph node metastases and lymphovascular invasion in endometrial cancer from multi-parametric MRI by deep learning [J ] . J Magn Reson Imaging , 2024 , 60 ( 6 ): 2730 - 2742 .
Jiang XT , Song JC , Zhang AN , et al . Preoperative assessment of MRI-invisible early-stage endometrial cancer with MRI-based radiomics analysis [J ] . J Magn Reson Imaging , 2023 , 58 ( 1 ): 247 - 255 .
Huang ML , Ren J , Jin ZY , et al . Application of magnetic resonance imaging radiomics in endometrial cancer:a systematic review and meta-analysis [J ] . Radiol Med , 2024 , 129 ( 3 ): 439 - 456 .
0
浏览量
10
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621