基于CT影像组学的机器学习模型预测局部进展期胃癌新辅助化疗效果  

Machine learning models based on CT radiomics for predicting the outcome of neoadjuvant chemotherapy in locally advanced gastric cancer

作　　者：韩枫[1] 王燕燕 杜妍 程家铭 王二娟 宋瑞瑞 Han Feng;Wang Yanyan;Du Yan;Cheng Jiaming;Wang Erjuan;Song Ruirui(Department of Information,Shanxi Province Cancer Hospital,Shanxi Hospital Affiliated to Cancer Hospital,Chinese Academy of Medical Sciences,Cancer Hospital Affiliated to Shanxi Medical University,Taiyuan 030013,China;Department of Medical Imaging,Shanxi Medical University,Taiyuan 030001,China;Department of Medical Sciences,Shanxi Medical University,Taiyuan 030001,China;the First Clinical Medical College,Shanxi Medical University,Taiyuan 030001,China;Department of Medical Imaging,Shanxi Province Cancer Hospital,Shanxi Hospital Affiliated to Cancer Hospital,Chinese Academy of Medical Sciences,Cancer Hospital Affiliated to Shanxi Medical University,Taiyuan 030013,China)

机构地区：[1]山西省肿瘤医院、中国医学科学院肿瘤医院山西医院、山西医科大学附属肿瘤医院信息科,太原030013 [2]山西医科大学医学影像学院,太原030001 [3]山西医科大学医学科学院,太原030001 [4]山西医科大学第一临床医学院,太原030001 [5]山西省肿瘤医院、中国医学科学院肿瘤医院山西医院、山西医科大学附属肿瘤医院医学影像科,太原030013

出　　处：《肿瘤研究与临床》2025年第1期1-7,共7页Cancer Research and Clinic

基　　金：山西省卫生健康委科研课题(2023113);山西省医学重点科研项目(2023XM014);山西省教育厅研究生教育创新计划(2024SJ158)。

摘　　要：目的探讨基于CT影像组学的机器学习模型预测局部进展期胃癌(LAGC)患者新辅助化疗(NAC)效果的价值。方法回顾性病例系列研究。纳入2018年1月至2020年11月在山西省肿瘤医院术前行NAC的LAGC患者279例,按7∶3的比例随机分为训练集(196例)和验证集(83例)。依据肿瘤退缩分级(TRG)分为NAC反应良好(GR)组(TRG 0~1级,55例)、NAC反应不良(PR)组(TRG 2~3级,224例)。收集患者的临床病理资料,如年龄、性别、分化程度、临床T和N分期,以及癌胚抗原(CEA)、糖类抗原199(CA199)水平。从增强门静脉期CT图像中提取影像组学特征,并对特征进行三步降维筛选,然后采用逻辑回归(LR)、朴素贝叶斯(NB)、随机森林(RF)、支持向量机(SVM)和极限梯度提升(XGB)5种机器学习算法构建影像组学的预测模型。绘制受试者工作特征曲线(ROC)和决策分析(DCA)曲线,评估各模型对LAGC患者NAC效果的预测效能及其临床效益。结果196例训练集患者中,GR组39例,PR组157例;83例验证集患者中GR组16例,PR组67例。训练集与验证集间以及训练集与验证集中GR组与PR组间患者的临床病理资料比较,差异均无统计学意义(均P>0.05)。基于门静脉期CT图像感兴趣区域共提取102个影像组学特征,最终筛选出6个关键特征,分别为original_firstorder_10Percentile、original_firstorder_RoubustMeanAbsoluteDeviation、originalglcmIdmn、originalglcmMCC、original_ngtdmBusyness、original_ngtdmContrast;6个特征在GR和PR组间差异有统计学意义(均P<0.05)。采用LR、NB、RF、SVM和XGB机器学习算法构建5种基于CT影像组学的预测模型,训练集中预测NAC的ROC曲线下面积分别为0.553、0.709、0.668、0.772和0.790,验证集中分别为0.662、0.622、0.683、0.752和0.784,基于XGB构建的模型综合效能最佳,其准确度、灵敏度和特异度分别为0.771、0.562和0.821;5种机器学习模型在训练集的DCA中,基于XGB构建的模型提供了更高的净效益。结�ObjectiveTo investigate the value of machine learning models based on CT radiomics for predicting the outcome of neoadjuvant chemotherapy(NAC)in patients with locally advanced gastric cancer(LAGC).MethodsA retrospective case series study was conducted.A total of 279 LAGC patients receiving NAC before surgery in Shanxi Province Cancer Hospital from January 2018 to November 2020 were included.According to a ratio of 7∶3,all patients were randomly divided into the training set(196 cases)and the validation set(83 cases).According to the tumor regression grade(TRG),the pathological grade was divided into the good response of NAC(GR)group(TRG 0-1,55 cases)and the poor response of NAC(PR)group(TRG 2-3,224 cases).The clinicopathological data of patients were collected,such as age,gender,differentiation degree,clinical T and N staging,carcinoembryonic antigen(CEA),and carbohydrate antigen 199(CA199)level.Radiomics features were extracted from the enhanced CT images in the vein phase,and the features were screened by 3-step dimensionality reduction.And then 5 machine learning algorithms including logistic regression(LR),naive bayes(NB),random forest(RF),support vector machine(SVM)and extreme gradient boosting(XGB)were applied to build prediction models based on the CT radiomics.The receiver operating characteristic(ROC)curve and the decision analysis(DCA)curve were drawn to evaluate the predictive performance and clinical benefit of each model on the outcome of NAC in patients with LAGC.ResultsAmong 196 patients in the training set,there were 39 cases in GR group and 157 cases in PR group;among 83 patients in the validation set,there were 16 cases in GR group and 67 cases in PR group.There were no statistically significant differences in clinicopathological data of patients between the training and validation sets,or between GR and PR groups in the training and validation sets(all P>0.05).A total of 102 radiomics features were extracted from region of interest of CT images in the vein phase,and 6 key features were finall

关 键 词：胃肿瘤 新辅助化疗 机器学习 体层摄影术 X线计算机 

分 类 号：R735.2[医药卫生—肿瘤]