CT影像组学列线图模型预测胃肠道间质瘤Ki-67表达的临床研究  

Clinical Study on Predicting the Expression of Ki-67 in Gastrointestinal Stromal Tumors by CT Radiomics Nomogram Model

作　　者：贾济波 刘原庆 冯飞文 胡粟[1,3] 胡春洪[1] JIA Jibo;LIU Yuanqing;FENG Feiwen(Department of Radiology,the First Affiliated Hospital of Soochow University,Suzhou,Jiangsu Province 215006,P.R.China)

机构地区：[1]苏州大学附属第一医院放射科,215006 [2]昆山市第三人民医院放射科,215300 [3]苏州大学影像医学研究所,215006

出　　处：《临床放射学杂志》2023年第5期783-788,共6页Journal of Clinical Radiology

摘　　要：目的 建立临床-CT影像组学列线图模型并验证其在术前预测胃肠道间质瘤(GIST)Ki-67表达的应用价值。方法 回顾性搜集145例经病理证实的GIST患者,并将其分为Ki-67高表达组和低表达组(8%为临界值)。提取10个临床及CT图像特征(年龄、性别、肿瘤部位、大小、形态、边界、囊变或坏死、钙化、表面溃疡、强化方式)用于构建临床模型。对病例的CT图像分别特征提取,并分为平扫期(N)、动脉期(A)、静脉期(V)及影像三期(N+A+V)4组,对上述4组通过Select Percentile和最小绝对收缩与选择算子(LASSO)算法降维、筛选组学特征后,分别使用5种分类器建立各期组学模型。采用受试者工作特征曲线下面积(AUC)进行量化。使用DeLong检验比较各模型间AUC值差异,得到最佳影像组学模型。然后组合临床模型和最佳影像组学模型建立临床-CT影像组学列线图模型。结果 经比较基于XGBoost分类器构建的临床-CT影像三期(N+A+V)的联合模型最佳,具有较高的预测效能,其在训练组和测试组的AUC分别达0.99(95%CI:0.97~1.00)、0.82(95%CI:0.66~0.95)。结论 临床-CT影像组学列线图模型在术前预测GIST的Ki-67表达上具有较高价值,从而为GIST的诊疗及预后判断提供参考。Objective To establish a clinical-CT radiomics nomogram model and verify its application in predicting the expression of Ki-67 in gastrointestinal stromal tumors(GIST)preoperatively.Methods 145 patients with GIST confirmed by pathology were retrospectively analyzed.and were divided into Ki-67 high expression group(>8%)and low expression group(≤8%).Ten clinical and CT image features(age,gender,tumor location,size,contour,margin,cystic degeneration or necrosis,calcification,surface ulcer and enhancement mode)were included for the construction of clinical models.The CT images-based radiomics features were extracted and divided into four groups:plain scan phase(N),arterial phase(A),venous phase(V)and image phase II(N+A+V).After dimensionality reduction and radiomics feature selection by select percentile and least absolute shrinkage and selection operator(LASSO)algorithm,five classifiers were used to establish the radiomics model of each period.The area under curve(AUC)of receiver operating characteristic(ROC)was used for quantification.DeLong test was used to compare the differences of AUC values between models to obtain the best model.Then the clinical model and the best radiomics model were combined to establish the nomogram model.Results After comparison,the joint model of clinical-three phases(N+A+V)based on XGBoost classifier is the best and has high prediction efficiency,with the AUC of 0.99,(95%CI:0.97-1.00)in training group and 0.82(95%CI:0.66-0.95)in the test group.Conclusion The nomogram model is a valuable method for predicting the expression of Ki-67 in GIST,so as to provide reference for diagnosis,treatment and prognosis of GIST.

关 键 词：胃肠道间质瘤 KI-67 体层摄影术 X线计算机 影像组学 列线图 

分 类 号：R735[医药卫生—肿瘤] R730.44[医药卫生—临床医学]