基于医学影像的人工智能模型在预测乳腺癌腋窝非前哨淋巴结转移中的应用进展  

Advances in artificial intelligence model based on medical imaging in predicting axillary non-sentinel lymph node metastasis in breast cancer

作　　者：孙旭成 毛宁[2] 林凡 王培源 谢海柱[2] SUN Xucheng;MAO Ning;LIN Fan;WANG Peiyuan;XIE Haizhu(School of Medical Imaging,Binzhou Medical University,Yantai 264000,China)

机构地区：[1]滨州医学院医学影像学院,山东烟台264000 [2]青岛大学附属医院烟台毓璜顶医院影像科,山东烟台264001 [3]滨州医学院烟台附属医院医学影像科,山东烟台264100

出　　处：《山东医药》2025年第1期155-159,共5页Shandong Medical Journal

摘　　要：乳腺癌腋窝非前哨淋巴结(NSLN)转移的评估对于制定个体化治疗方案至关重要。近年来,随着人工智能(AI)技术不断发展,其应用场景亦不断拓展。目前,AI技术已被广泛应用于NSLN的评估,多种AI模型被设计用于分析对比增强乳腺X线摄影(CEM)、动态对比增强磁共振(DCE-MRI)及超声图像,并预测乳腺癌腋窝NSLN转移。AI模型主要分为传统机器学习模型和深度学习模型,其中,传统机器学习模型具有计算复杂度低、数据需求少、模型可解释性强的优点,而深度学习模型则具有速度快、准确率高的特点。传统机器学习模型和深度学习模型在优化NSLN转移临床诊疗策略方面发挥重要作用。Evaluation of axillary non-sentinel lymph node(NSLN)metastases in breast cancer is essential for individualized treatment.In recent years,with the continuous development of artificial intelligence(AI),it is widely used in the evaluation of NSLN.At present,AI technology has been widely used to evaluate the status of NSLN,and a variety of AI models have been designed to analyze contrast-enhanced mammography(CEM),dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI),and ultrasound images and to predict axillary NSLN metastasis of breast cancer.AI models mainly include traditional machine learning models and deep learning models.Among them,the traditional machine learning model has the advantages of low computational complexity,less data requirements,and strong model interpretability,while the deep learning model has the characteristics of fast speed and high accuracy.Traditional machine learning models and deep learning models play an important role in optimizing clinical diagnosis and treatment strategies for NSLN metastasis.

关 键 词：乳腺癌 非前哨淋巴结转移 医学影像 人工智能 传统机器学习模型 深度学习模型 

分 类 号：R737.9[医药卫生—肿瘤]