生命科学仪器2023，Vol.21Issue(3)：12-18,7.DOI:10.11967/2023210602
使用深度学习方法的脑肿瘤图像分割综述
A review of brain tumor image segmentation using depth learning
刘鹏 1刘伟峰 1唐晓英1
作者信息
- 1. 北京理工大学生命学院,北京 100081
- 折叠
摘要
Abstract
The accurate segmentation of brain tumor image can determine its appearance and location information,which is of great significance for early diagnosis and treatment.However,at present,manual methods are mainly used in clinical annotation,which leads to low efficiency and different annotation results due to different people.At present,deep learning method has been widely used in brain tumor image segmentation,and excellent results have been achieved by designing neural network structure and other methods.This paper will review and summarize the latest progress and challenges in recent years,and summarize the five modules of image preprocessing,image fea-ture extraction,image feature fusion,parameter update and prediction,and post-processing to summarize the breakthrough key technologies,so that the location of innovation points is more clear,which is conducive to induc-tion and learning.Finally,the potential research directions are prospected.关键词
脑肿瘤图像分割/深度学习/早期诊断Key words
Brain tumor image segmentation/Deep learning/Early diagnosis分类
医药卫生引用本文复制引用
刘鹏,刘伟峰,唐晓英..使用深度学习方法的脑肿瘤图像分割综述[J].生命科学仪器,2023,21(3):12-18,7.
