摘要
钼靶检查是当前临床诊断乳腺肿瘤的常规手段,患者痛苦相对较小、简便易行、分辨率高、可重复性好。为了提高诊断效率,减小误诊风险,针对乳腺钼靶图像开发基于人工智能的计算机辅助诊断系统(computer-aided diagnosis,CAD)显得尤为重要。传统的分类方法需要使用大量的手工特征,而深度学习能够自动从数据中学习特征,避免了传统算法中人工设计、提取特征的复杂性和局限性。我们从感兴趣区域和全图两个方面对近年来基于深度学习的乳腺钼靶图像分类方法研究进展予以综述和展望。调研发现深度学习在乳腺钼靶图像分类方面展示了不错的效果,其中基于深度卷积神经网络的分类方法已经成为当下的热门技术。
In the clinical diagnosis of breast cancer,mammography has become one of the routine examination methods because of its advantages of relatively small pain,easy operation,high resolution and good repeatability.In order to improve the diagnostic efficiency and reduce the risk of misdiagnosis,it is particularly important to develop a computer-aided diagnosis(CAD)system for mammogram.Traditional classification methods require a large number of manual features,while deep learning can automatically learn features from data,avoiding the complexity and limitations of manual design and extraction of features in traditional algorithms.We review the research progress on classification methods for mammogram based on deep learning in recent years from the region of interest and the whole image two aspects.According to the survey,we find that deep learning has shown good results in the classification of mammogram,and the classification method based on deep convolutional neural network has become a popular technology at present.
作者
包昌宇
彭俊川
胡楚婷
简文静
王先明
刘维湘
BAO Changyu;PENG Junchuan;HU Chuting;JIAN Wenjing;WANG Xianming;LIU Weixiang(School of Biomedical Engineering, Shenzhen University Health Science Center, Shenzhen 518055, China;Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging, Shenzhen 518060;National-Regional Key Technology Engineering Laboratory for Medical Ultrasound , Shenzhen 518060;Department of Breast and Thyroid, Shenzhen Second People's Hospital of , The First Affiliated Hospital of Shenzhen University Health Science Center,Shenzhen 518035)
出处
《生物医学工程研究》
2020年第2期208-213,共6页
Journal Of Biomedical Engineering Research
基金
深圳市科技应用示范项目(KJYY20170724100440556)。
关键词
乳腺肿瘤
钼靶
计算机辅助诊断系统
图像分类
深度学习
卷积神经网络
Breast cancer
Mammogram
Computer-aided diagnosis system
Image classification
Deep learning
Convolutional neural network