瘤内联合瘤周超声组学列线图预测乳腺癌腋窝淋巴结转移的临床价值  

Clinical Value of Intratumoral Combined with Peritumoral Ultrasonic Radiomics Nomogram in Predicting Axillary Lymph Node Metastasis in Breast Cancer

作　　者：杨茹 许雨晴 慈红非 李阳 Yang Ru;Xu Yuqing;Ci Hongfei;Li Yang(Department of Ultrasound,The First Affiliated Hospital of Bengbu MedicalUniversity,Bengbu,Anhui 233000,China;Department of Pathology,The First Affiliated Hospital of Bengbu MedicalUniversity,Bengbu,Anhui 233000,China)

机构地区：[1]蚌埠医科大学第一附属医院超声科,安徽省蚌埠市233000 [2]蚌埠医科大学第一附属医院病理科,安徽省蚌埠市233000

出　　处：《中国超声医学杂志》2025年第3期284-287,共4页Chinese Journal of Ultrasound in Medicine

基　　金：蚌埠医科大学校级课题(No.Byycx23094)。

摘　　要：目的 探讨基于治疗前乳腺癌原发灶超声图像瘤内及瘤周组学的列线图模型预测乳腺癌腋窝淋巴结(ALN)转移的价值。方法 回顾性选择女性乳腺癌患者180例,基于二维超声图像获取瘤内、瘤周及瘤内+瘤周的组学特征,并联合临床特征构建列线图模型。通过受试者工作特征曲线评估模型的预测效能。结果 在临床特征中,超声-ALN及ALN触诊是独立危险因素,在组学模型中,瘤内+瘤周模型预测效能较高,二者联合构建列线图模型。列线图模型在训练集及验证集中诊断性能均最佳,曲线下面积分别为0.914、0.843。结论 瘤内+瘤周超声影像组学模型可更好地预测ALN转移,联合临床特征的列线图模型可进一步提高模型的预测性能。Objective To explore the value of a nomogram model based on intratumoral and peritumoral ra-diomics features from pre-treatment ultrasound images in predicting axillary lymph node(ALN)metastasis in breast cancer.Methods A retrospective analysis was conducted on 180 female breast cancer patients.Radiomics features were extracted from intratumoral,peritumoral,and combined intratumoral+peritumoral regions using two-dimensional ultrasound images.These features were integrated with clinical characteristics to construct a nomogram model.The predictive performance of the model was evaluated through receiver operating characteristic curve analy-sis.Results In clinical features,ultrasound-ALN and ALN palpation were identified as independent risk factors.Among the omics models,the combined intratumoral and peritumoral model showed higher predictive power,the no-mogram model constructed by combining radiomics and clinical features showed the best diagnostic performance in both the training and validation cohorts,with area under the curve of o.914 and o.843 respectively.Conclusions The intratumoral and peritumoral ultrasound radiomics model can predict ALN metastasis better,incorporating clinical fea-tures into a nomogram model further improves its predictive performance.

关 键 词：乳腺癌 腋窝淋巴结 瘤周 超声组学 列线图 

分 类 号：R73[医药卫生—肿瘤]