摘要
介绍了使用率较高的深度学习基础网络的结构模型及常用改进方法,阐述了使用深度学习神经网络对MRI及CT图像中的脑部、肺部、肝脏、胰腺、前列腺等器官进行自动分割的研究进展。指出了未来应更多关注3D图像分割和少样本训练,开发更适用的网络结构,以提高器官分割准确性和分割效率。
The structure model and common improvement methods of deep learning neural network were introduced and the research progress of automatic segmentation of brain,lung,liver,pancreas,prostate and other organs in MRI and CT images by using deep learning neural network were described.It's pointed out more attention had to be paid to 3D image segmentation and less sample training in the future to develop applicable network structure and improve the accuracy and efficiency of organ segmentation.
作者
郭雯
鞠忠建
吴青南
全红
戴相昆
GUO Wen;JU Zhong-jian;WU Qing-nan;QUAN Hong;DAI Xiang-kun(School of Physics Science and Technology,Wuhan University,Wuhan 430072,China;Department of Radiation Oncology,the First Medical Center,Chinese PLA General Hospital,Beijing 100853,China;Peking University International Hospital,Beijing 102206,China)
出处
《医疗卫生装备》
CAS
2020年第1期85-94,共10页
Chinese Medical Equipment Journal
基金
国家自然科学基金(61671204)
解放军总医院临床科研扶持基金(2017FC-TSYS-3027)
关键词
深度学习
医学图像
自动分割
卷积神经网络
危及器官
deep learning
medical image
automatic segmentation
convolution neural network
organs at risk
