MRI影像组学在局部进展期直肠癌治疗后无疾病进展的预测作用  

The role of radiomics in predicting no disease progression after treatment of locally advanced rectal cancer

作　　者：王泽坤 耿懿康 于韬 WANG Zekun;GENG Yikang;YU Tao(Department of Medical Imaging,Liaoning Cancer Hospital&Institute,China Medical University Cancer Hospital,Dalian University of Technology Affiliated Cancer Hospital,Shenyang 110042,China;School of Intelligent Medicine of China Medical University,Shenyang 110042,China)

机构地区：[1]大连理工大学附属肿瘤医院中国医科大学肿瘤医院辽宁省肿瘤医院医学影像科,沈阳110042 [2]中国医科大学智能医学学院,沈阳110042

出　　处：《磁共振成像》2025年第1期61-67,共7页Chinese Journal of Magnetic Resonance Imaging

摘　　要：目的探讨MRI影像组学在局部进展期直肠癌新辅助放化疗(neoadjuant chemoradiotherapy,nCRT)后术后3年无疾病进展(disease-free survival,DFS)预测的能力。材料与方法回顾性分析100例局部进展期直肠癌(locally advanced rectal cancer,LARC)患者的临床信息及影像资料,其中有疾病进展50例,无疾病进展50例,按4∶1随机分配训练集和测试集。提取术前MRI的T2WI轴位快速自旋回波(fast spin echo,FSE)、T2WI矢状位影像组学特征,然后使用最小冗余最大相关滤波器进行降维。采用logistic回归方法构建临床-影像诺模图。绘制受试者工作特征(receiver operating characteristi,ROC)、决策曲线分析(decision curve analysis,DCA)和校准曲线以评价模型预测效能。结果临床信息中,糖类抗原19-9(carbohydrate antigen 19-9,CA19-9)水平在训练集和测试集中差异具有统计学意义(P<0.05)。各模型在训练集和验证集上均展现出高预测准确性,其中临床-影像诺模图的ROC曲线下面积(area under the curve,AUC)在训练集和验证集上分别达到0.933[95%置信区间(confidence interval,CI):79.7%~100.0%]和0.980(95%CI:79.7%~100.0%)。结论本研究建立的影像组学模型能够有效预测LARC患者nCRT后的DFS,可为临床决策提供重要参考。Objective:To investigate the ability of MRI imaging to predict disease-free survival(DFS)at 3 years after neoadjuant chemoradiotherapy(nCRT)for locally advanced rectal cancer(LARC).Materials and Methods:Clinical informations and imaging data of 100 patients with LARC were retrospectively analyzed,including 50 patients with disease progression and 50 patients without disease progression.The training set and test set were randomly allocated according to 4∶1.The imaging features of T2WI axis fast spin echo(FSE)and T2WI sagittal were extracted from preoperative MRI,and then dimensionality was reduced using minimum redundancy maximum correlation filter.Logistic regression was used to construct a nomogram containing the clinical parameter carbohydrate antigen 19-9(CA19-9).Receiver operating characteristi(ROC),decision curve analysis(DCA)and calibration curves were drawn to evaluate the nomogram prediction effect.Results:In clinical information,CA19-9 level was statistically significant in training set and test set(P<0.05).Among the key imaging features,the features of T2WI sagittal and T2WI FSE sequences contributed the most to the prediction of DFS.Our model demonstrated high predictive accuracy on both the training and validation sets,with the area under the ROC curve(AUC)reached 0.933[95%confidence interval(CI):79.7%to 100.0%]and 0.980(CI:79.7%to 100.0%)on the training set and validation set,respectively.Conclusions:The radiomics model established in this study can effectively predict DFS after nCRT in LARC patients,which can provide an important reference for clinical decision-making.

关 键 词：局部进展期直肠癌 磁共振成像 影像组学 预测 无病生存期 

分 类 号：R445.2[医药卫生—影像医学与核医学] R735.37[医药卫生—诊断学]