乳腺癌ABVS超声影像组学特征联合ABVS超声及血清学特征预测HER2状态  

The Prediction Value of Ultrasound Radiomics Features Based on ABVS Combined with ABVS Ultrasound and Serological Features for Preoperative HER2 Status in Breast Cancer

作　　者：王惠[1] 陈飞[1] 李金桥 李瑞霞 张炜阳[1] 黄赞赟 郭顺林[3] Wang Hui;Chen Fei;Li Jinqiao;Li Ruixia;Zhang Weiyang;Huang Zanyun;Guo Shunlin(Department of Ultrasound,The First Hospital of Lanzhou University,Lanzhou 730000,China;The First Clinical Medicine College of Lanzhou University,Lanzhou 730000,China;Department of Radiology,The First Hospital of Lanzhou University,Lanzhou730000,China)

机构地区：[1]兰州大学第一医院超声科,兰州市730000 [2]兰州大学第一临床医学院,兰州市730000 [3]兰州大学第一医院放射科,兰州市730000

出　　处：《中国超声医学杂志》2024年第5期516-520,共5页Chinese Journal of Ultrasound in Medicine

基　　金：甘肃省科技计划资助(No.23JRRA1589);甘肃省卫生健康行业科研项目(No.GSWSKY2022-54);兰州大学第一医院院内基金(No.ldyyyn2019-65)。

摘　　要：目的探讨基于自动乳腺容积扫查(ABVS)超声影像组学、ABVS超声和血清学特征的加权组合模型对乳腺癌人表皮生长因子受体2(HER2)状态的预测性能。方法回顾性纳入271例浸润性乳腺癌患者,其中174例随机分为训练集和验证集,97例作为测试集。每位患者均进行病理检测及术前ABVS检查。基于ABVS瘤内、瘤周区域提取的超声影像组学特征、ABVS超声和血清学特征构建模型。采用模型加权组合方法构建最优模型。使用受试者工作特征(ROC)曲线评估模型的预测性能。结果在验证集中,最优加权组合模型的曲线下面积(AUC)、灵敏度和特异度分别为0.83(95%CI:0.69~0.96)、100%和62.5%。在测试集中,最优加权组合模型的AUC、灵敏度和特异度分别为0.69(95%CI:0.58~0.80)、69.44%和51.61%。结论基于ABVS的瘤内及瘤周超声影像组学特征、ABVS超声及血清学特征的加权组合模型能够无创预测乳腺癌HER2状态。Objective To investigate the prediction efficacy of the weighted combination model based on automated breast volume scanner(ABVS)ultrasound radiomics features,ABVS ultrasound and serological features for human epidermal growth factor receptor 2(HER2)status in breast cancer.Methods A total of 271 patients with invasive breast cancer were included in the retrospective study,of which 174 patients were randomized into the training and validation sets,and 97 patients were the test set.pathology testing and preoperative ABVS examination were performed in each patient.Ultrasound radiomics features extracted from the ABVS-based tumor,peritumoral region,ABVS ultrasound and serological features were used to construct the model.The model-weighted combination method was used to construct the optimal model.The receiver operating characteristic(ROC)curves were utilized to evaluate the predictive performance of the models.Results For the validation set,the optimized weighted combination model achieved the area under the curve(AUC)of 0.83(95%CI:0.69-0.96),with the sensitivity and specificity were 100%,62.5%,respectively;For the test set,the optimized weighted combination model attained the AUC of 0.69(95%CI:0.58-0.80),with the sensitivity and specificity were 69.44%,51.61%,respectively.Conclusions The weighted combination model composed of ABVS-based intratumoral and peritumoral ultrasound radiomics features,ABVS ultrasound and serological features is enable to noninvasively predict of HER2 status in breast cancer.

关 键 词：乳腺癌 人表皮生长因子受体2 自动乳腺容积扫查 超声影像组学 

分 类 号：R73[医药卫生—肿瘤]