目的 探讨基于动态对比增强磁共振成像(dynamic contrast-enhanced magnetic resonance imaging,DCE-MRI)和扩散加权成像(diffusion-weighted imaging,DWI)的瘤内及瘤周影像组学预测乳腺癌人表皮生长因子受体2(human epidermal growth factor receptor-2,HER-2)状态的价值.材料与方法 回顾性分析246例经术后病理证实的乳腺癌患者的临床及影像学资料,按7∶3比例随机分为训练组和验证组.采用ITK-SNAP软件手动勾画病灶瘤内感兴趣区,使用PHIgo-AK软件进行瘤周的扩展并提取瘤内及瘤周的影像组学特征.采用最小冗余最大相关(max-relevance and min-redundancy,mRMR)算法等选择DCE-MRI、DWI瘤内及瘤周的最优特征数.分别建立单序列及联合序列的影像组学模型,采用受试者工作特征(receiver operating characteristic,ROC)曲线对各模型的预测效能进行分析,并计算曲线下面积(area under the curve,AUC),选出预测效能最高的模型,在训练组中从临床及常规影像学特征中通过单因素logistic回归筛选出预测HER-2状态的独立危险因素,结合预测效能最高模型的影像组学标签评分(radiomic score,rad-score)建立融合模型,并以诺模图(nomogram)展示,采用AUC值,决策曲线分析(decision curve analysis,DCA)评估模型的效能及临床价值.结果 基于DCE-MRI和DWI瘤内及瘤周的影像组学联合模型预测HER-2状态的AUC值在训练组和验证组分别为0.953和0.948,效能最高.肿瘤最大径是区分乳腺癌HER-2状态的独立危险因素,最终结合rad-score和肿瘤最大径建立的融合模型对乳腺癌HER-2状态有良好的预测效能,在训练组的AUC值为0.961,验证组为0.958.结论 基于DCE-MRI和DWI瘤内及瘤周的影像组学方法对乳腺癌HER-2状态的预测具有良好的价值.
Value of intratumoral and peritumoral radiomics based on DCE-MRI and DWI in predicting HER-2 status in breast cancer
Objective:To explore the value of dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)and diffusion-weighted imaging(DWI)based intratumoral and peritumoral radiomics methods in predicting the status of human epidermal growth factor receptor 2(HER-2)in breast cancer.Materials and Methods:Clinical and imaging data of 246 patients with pathologically proven breast cancer were retrospectively analyzed and randomly divided into training group and verification group according to a ratio of 7∶3.ITK-SNAP software was used to manually outline the intratumoral areas of interest,and PHIgo-AK software was used to expand the peritumoral and extract the intratumoral and peritumoral radiomics features.The optimal number of intratumor and peritumor features of DCE-MRI and DWI were selected by max-relevance and min-redundancy(mRMR)algorithm.Radiomics models of single sequence and combined sequence were established respectively,and the prediction efficiency of each model was analyzed by receiver operating characteristic(ROC)curve.The area under the curve(AUC)was calculated to select the model with the highest predictive efficiency.Independent risk factors for predicting HER-2 status were screened from clinical and routine imaging features in the training group through single logistic regression.A fusion model was established by combining the radiomic score(rad-score)of the model with the highest predictive power,and then presented by nomogram.AUC value,decision curve analysis and DCA were used to evaluate the efficacy and clinical value of the model.Results:The combined intratumoral and peritumoral imaging model based on DCE-MRI and DWI predicted the AUC value of HER-2 status in the training group and the verification group,which were 0.953 and 0.948,respectively,with the highest efficiency.Tumor maximum diameter is an independent risk factor for distinguishing breast cancer HER-2 status.Finally,the fusion model established by combining rad-score and tumor maximum diameter has good predictive efficacy for breast cancer HER-2 status,with the AUC value of 0.961 in the training group and 0.958 in the verification group.Conclusions:The intratumoral and peritumoral radiomic methods based on DCE-MRI and DWI have good value in the prediction of breast cancer HER-2 status.
breast cancerhuman epidermal growth factor receptor 2radiomicsperitumormagnetic resonance imagingdynamic contrast-enhanced magnetic resonance imagingdiffusion-weighted imaging
万方数据
维普
