计算机工程与应用2025，Vol.61Issue(15)：14-35,22.DOI:10.3778/j.issn.1002-8331.2411-0136
基于深度学习的肺结节CT图像分割与分类研究综述
Review of Deep Learning-Based Segmentation and Classification of CT Images of Lung Nodules
摘要
Abstract
Lung cancer is one of the deadliest forms of cancer,and lung nodules,as early symptoms of the disease,pose a serious threat to people's lives and health.The segmentation and classification of lung nodule CT images based on deep learning can help doctors quickly and accurately diagnose early-stage nodules,which has significant clinical value for the treatment of lung cancer.To explore the segmentation and classification techniques for lung nodule CT images,firstly,common datasets and evaluation indicators are introduced.Secondly,the deep learning models for lung nodule CT image segmentation and classification are reviewed from two perspectives:single-network models and multi-network models based on U-Net,feature fusion and texture feature classification methods using convolutional neural networks.The inno-vative points of domestic and foreign research over the past 5 years are summarized through specific experiments,and the advantages and disadvantages of various models have also been outlined.Finally,the development direction in this field is discussed,offering theoretical guidance and reference for future research in this area.关键词
肺结节/深度学习/计算机辅助诊断/医学图像/卷积神经网络/U-Net模型Key words
lung nodules/deep learning/computer-aided diagnosis/medical images/convolutional neural networks/U-Net model分类
信息技术与安全科学引用本文复制引用
古力米热·阿吾旦,叶俊翔,玛依拉·阿不都克力木,王梦飞,哈里旦木·阿布都克里木,阿布都克力木·阿布力孜..基于深度学习的肺结节CT图像分割与分类研究综述[J].计算机工程与应用,2025,61(15):14-35,22.基金项目
国家自然科学基金(62366050). （62366050）
