卷积神经网络在早期肺癌诊疗方面的应用进展  

Advances in the application of convolutional neural network in the diagnosis and treatment of early lung cancer

作　　者：薛宇航 陈东来[2] 佘云浪 陈昶[2] 陈勇兵[1] 徐卫华[1] Xue Yuhang;Chen Donglai;She Yunlang;Chen Chang;Chen Yongbing;Xu Weihua(Department of Thoracic Surgery,Second Affiliated Hospital of Soochow University,Medical College of Soochow University,Suzhou 215004,China;Department of Thoracic Surgery,Shanghai Pulmonary Hospital,Shanghai 200433,China)

机构地区：[1]苏州大学附属第二医院胸心外科,215000 [2]同济大学附属上海市肺科医院胸外科,200433

出　　处：《中华胸心血管外科杂志》2020年第9期569-572,共4页Chinese Journal of Thoracic and Cardiovascular Surgery

基　　金：国家自然科学基金(81800406);苏州市重点学科建设(SZXK201803);苏州市科技计划项目(SS2019061);苏州大学附属第二医院科研预研基金项目(SDFEYBS1902)。

摘　　要：近年来卷积神经网络(convolutional neural network,CNN)在辅助分析肺癌方面表现出良好的应用前景和研究价值。CNN可以从图像数据中自主学习以提取特定的与临床相关的特征。本文对构建CNN通过输入CT图像进行端到端分析,从而在术前预测肺癌风险和临床治疗效果;以及通过输入肺结节组织切片图像进行术后病理学分析作一综述。In recent years,convolutional neural network(CNN)has exhibited its promising prospect and research value in assisting the analysis of lung cancer.The CNN can automatically learn from the images to extract specific clinically relevant features.This review summarized the construction of CNN by inputting CT images for end-to-end analysis to predict lung cancer risk and the efficacy of clinical treatment,as well as the application of CNN for analyzing pathological slides of resected pulmonary nodules via inputting histopathologic images.

关 键 词：早期肺癌 卷积神经网络 病理学分析 肺结节 CT图像 自主学习 辅助分析 

分 类 号：R734.2[医药卫生—肿瘤]