基于增强CT影像组学-临床综合模型术前鉴别卵巢囊腺瘤的研究  

Application value of combined model based on enhanced CT radiomics-clinical features for the preoperative distinguishing ovarian serous cystadenoma and mucinous cystadenoma

作　　者：李娇娇[1] 刘迪[1] 张丽 杨帆 陈静[1] 叶兆祥 LI Jiao-jiao;LIU Di;ZHANG Li(Department of Medical Imaging,the First Affiliated Hospital of Hebei North University,Hebei 075000,China)

机构地区：[1]河北北方学院附属第一医院医学影像部,河北张家口075000 [2]天津医科大学肿瘤医院(国家肿瘤临床医学研究中心/天津市肿瘤防治重点实验室/天津市恶性肿瘤临床医学研究中心)放射科,天津300060

出　　处：《放射学实践》2025年第3期377-382,共6页Radiologic Practice

基　　金：河北省医学科学研究课题计划资助(20241189)。

摘　　要：目的:建立增强CT影像组学-临床综合模型并评估其在术前鉴别卵巢浆液性囊腺瘤和粘液性囊腺瘤的应用价值。方法:回顾性收集222例经病理证实的卵巢浆液性囊腺瘤或粘液性囊腺瘤患者的临床和CT资料。PyRadiomics被用于提取影像组学特征,对特征降维,计算影像组学标签(RADscore),构建影像组学模型。分析患者的临床、CT特征资料,建立临床模型。将RADscore与临床、CT特征组合建立综合模型。采用受试者操作特征(ROC)曲线下面积(AUC)评估模型的鉴别诊断性能。利用决策曲线及校正曲线分析模型的临床应用价值。结果:在训练集和测试集中,临床、影像组学及综合模型的AUC分别为0.772、0.766、0.867和0.704、0.751和0.826。校准和决策曲线提示综合模型具有较好的临床应用价值。结论:综合模型在鉴别卵巢浆液性囊腺瘤和粘液性囊腺瘤方面具有潜在的价值。Objective:To develop and evaluate a combined model integrating enhanced CT radiomics and clinical features for preoperative distinction between ovarian serous and mucinous cystadenoma.Methods:A total of 222 patients with ovarian serous cystadenoma or mucinous cystadenoma confirmed by pathology were retrospectively included.The clinical,CT data and portal venous phase CT images were collected.The radiomics features were extracted by PyRadiomics.Selected radiomics features were used to calculate radiomics score and develop radiomics model.The crucial clinical data and CT features were used to develop clinical model.The RADscore was combined with significant clinical and CT features to establish the combined model.The area under the receiver operating characteristic(ROC)curve(AUC)was applied to evaluate the performance of the models.The decision and calibration curves were used to estimate the clinical application value of the models.Results:In the training cohort,the AUCs for the clinical model,radiomics model and combined model were 0.772,0.766 and 0.867,respectively.In the test cohort,the AUCs were 0.704,0.751 and 0.826,respectively.The calibration curve and decision curve confirmed that the combined model held good clinical application value.Conclusion:The combined model had potential value in differentiating ovarian serous cystadenoma from mucinous cystadenoma.

关 键 词：卵巢囊腺瘤 影像组学 列线图 体层摄影术 X线计算机 

分 类 号：R814.42[医药卫生—影像医学与核医学] R737.31[医药卫生—放射医学] R-05[医药卫生—临床医学]