摘要
目的探索基于对比增强CT(CE-CT)的影像组学方法在将术前cT4分期胃癌患者区分为pT4b分期与no-pT4b分期中的应用价值。方法回顾性搜集术前行CE-CT检查并分期为cT4的691例胃癌患者的临床及影像资料,随机分为训练组(n=460)和验证组(n=231)。根据患者病理分期分为pT4b分期与no-pT4b分期,分析训练组患者的临床特征及基于静脉期CE-CT图像的组学特征,构建临床模型、影像组学模型、临床结合影像组学列线图。在构建影像组学模型中,最小绝对收缩和选择算子(LASSO)算法、最小冗余最大相关算法(mRMR)用于特征选择和降维;支持向量机算法(SVM)作为机器学习分类器。通过受试者工作特征曲线下面积(AUC)、决策曲线分析法(DCA)及校正曲线评估模型性能。结果691例胃癌患者,病理分期T4b分期者332例,非T4b分期者359例。在验证组中,临床模型(Borrmann分型+cT分期)AUC为0.853 95%置信区间(95%CI:0.821~0.923),其中影像医师主观评估的cT分期AUC为0.805(95%CI:0.746~0.862);组学模型AUC为0.781(95%CI:0.711~0.846);临床结合影像组学列线图是最佳预测模型,其AUC为0.876(95%CI:0.824~0.931)。在训练和验证组中,列线图的校正曲线均表现出真实与预测结果之间良好的一致性,DCA曲线也均显示了较高的正向净收益,表现出其良好的临床应用价值。结论基于静脉期CE-CT图像的影像组学模型性能未优于临床模型与主观评估的cT分期,说明本研究的影像组学方法在区分cT4分期患者中具有一定的局限性。然而,临床结合影像组学列线图取得了最佳的预测效能与临床收益,说明临床结合影像组学方法可提高模型性能,并对cT4分期胃癌患者选择治疗方式存在一定参考价值。
Objective Preoperative accurate staging was crucial for cT4 stage gastric cancer patients.The purpose of this study was to explore the value of the radiomics method based on contrast-enhanced computed tomography(CE-CT)in differentiating gastric cancer patients with preoperative cT4 stage into pT4b stage and no-pT4b stage.Methods The clinical information and CE-CT image date of 691 gastric cancer patients(460 in the training set,231 in the validation set)who underwent preoperative CECT examination and were staged as cT4 were retrospectively collected.According to the pathological stage,the patients were divided into pT4b stage and no-pT4b stage.In the training set,the clinical characteristics and the radiomics features extracted from the venous-phase CE-CT images were analyzed to construct the clinical model,radiomics model and clinical combined radiomics nomogram.Two kinds of radiomics feature selection methods were used to dimensionality reduction:The least absolute shrinkage and selection operator(LASSO)algorithm,and the minimum redundancy maximum relevance(mRMR)algorithm,Support vector machine(SVM)algorithm as machine learning classifiers.The model performance was evaluated by receiver operating characteristic(ROC)area under the curve(AUC),Decision Curve Analysis(DCA)curve and calibration curve.Results A total of 691 gastric cancer patients were included in this study:332 patients with pT4b stage and 359 patients with no-pT4b stage.In the validation set,the AUC of the clinical model was 0.853(95%confidence interval(95%CI):0.821-0.923),and the AUC of cT staging assessed by the radiologist's subjective assessment was 0.805(95%CI:0.746-0.862);The AUC of the radiomics model was 0.781(95%CI:0.711-0.846);Nomogram was the best predictive model with an AUC of 0.876(95%CI:0.824-0.931).In the training and validation sets,the calibration curves of the nomogram showed satisfactory agreement between prediction and actual result,the DCA curves also showed the large clinical positive net benefit of nomogram.Conclusion The performance of the radiomics model based on venous phase CE-CT images was not superior to clinical models and subjective assessment of cT stage,indicating the radiomics method of this study has certain limitations in distinguishing patients with cT4 staging.However,the clinical combined radiomics nomogram achieved the best predictive performance and clinical benefit,indicating that the clinical combined radiomics method can improve the model performance,which has a certain reference value for the selection of treatment methods for patients with cT4 stage gastric cancer.
作者
刘波
张登云
王鹤翔
张大玮
王赫
张群
周冠知
张坚
LIU Bo;ZHANG Dengyun;WANG Hexiang(Department of Gastrointestinal Surgery,Pingdu District,The Affiliated Hospital of Qingdao University,Qingdao,Shandong Province 266700,P.R.China)
出处
《临床放射学杂志》
北大核心
2023年第1期77-85,共9页
Journal of Clinical Radiology
基金
国家自然科学基金面上项目基金资助项目(编号:81770631)。
关键词
胃癌
增强CT
影像组学
T分期
Gastric cancer
Contrast-enhanced CT
Radiomics
T staging
