网站首页            期刊简介             编委会             投稿指南             期刊订阅             下载中心             在线留言            联系我们             English
  2025年4月4日 星期五  
文章快速检索
中国生物医学工程学报  2024, Vol. 43 Issue (5): 582-595    DOI: 10.3969/j.issn.0258-8021.2024.05.007
  论著 本期目录 | 过刊浏览 | 高级检索 |
基于U-Net卷积神经网络的目标分子人机交互观测方法
张新峰1, 殷文斌1, 方金鹏1, 张新梅2*

1(扬州大学 信息工程学院,江苏 扬州 225127)
2(西安市人民医院(西安市第四医院)超声医学中心,西安 710004)
An Observation Method of Human-Computer Interaction for Identification of Target MoleculesBased on U-Net Convolutional Neural Network
Zhang Xinfeng1, Yin Wenbin1, Fang Jinpeng1, Zhang Xinmei2*
1(College of Information Engineering, Yangzhou University, Yangzhou 225127, Jiangsu, China)
2(Ultrasound Medical Center, Xi'an People's Hospital (Xi'an Fourth Hospital), Xi'an 710004, China)
全文: PDF (12269 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 对生命活动中发挥重要作用的分子进行观测是发现生命活动内在机理的重要手段。现有的生物医学影像处理方法大多集中在对特定物质的检测和识别,难以适应不断变化的科研需要。为此,本研究提出了一种基于U-Net卷积神经网络的人机交互方法以识别生物医学影像中所有同类分子,如:细胞、蛋白质等。首先利用U-Net卷积网络将待观察的分子影像转换为深度特征图,然后使用目标分子的特征在整个特征图上进行匹配,以检测出所有感兴趣的同类分子。之后利用通道和空间可靠的判别式相关滤波器构建多目标跟踪器以实现对目标分子的持续追踪。结果表明,该方法可以通过简单的人机交互快速检测出感兴趣的同类分子,获取目标分子的数量、分布以及相互作用等重要信息,Attention-based U-Net和U-Net在从细胞核、人类蛋白质图谱、细菌和血红细胞数据集中随机抽取的200张静态测试影像上的各项指标表现稳定,平均精度的平均值分别为0.912 5和0.898 1,同时对小鼠干细胞动态影像中的目标跟踪准确且保持稳定,证明了方法的有效性,可满足生命科学研究中对微观生命过程观测的需要。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
张新峰
殷文斌
方金鹏
张新梅
关键词 深度学习U-Net目标检测目标跟踪人机交互    
Abstract:The observation of molecules that play an important role in life activities is an important way to discover intrinsic mechanisms of the life activities. Most of existing biomedical image processing methods focus on the detection and identification of specific substances, however, it is difficult to adapt to changing demands of scientific research. To this end, this paper proposed a human-computer interaction method based on U-Net convolutional neural network to identify all the same molecules in biomedical images, such as nucleic cell, proteins, etc. First, the U-Net convolutional network was used to convert molecular images to deep feature maps, and then the features of the target molecules were used to match on the entire feature map to detect all the same molecules of interest. Then, the CSR-DCF (discriminative correlation filter with channel and spatial reliability) algorithm was used to build a multi-target tracker to achieve continuous tracking of the target molecules. Experimental results showed that the proposed method was able to quickly detect similar molecules of interest through simple human-computer interaction, and obtain important information on the number, distribution and interactions of target molecules. Attention-based U-Net and U-Net performed consistently on 200 static test images randomly selected from Nucleus, Human Protein Atlas, Bacteria and Blood Red Cell datasets, withaverage precision mean values of 0.912 5 and 0.898 1, respectively. At the same time, the tracking of targets in the dynamic images of mouse stem cells was accurate and stable, proving the effectiveness of the method to meet the needs of microscopic life process observation in life science research.
Key wordsdeep learning    U-Net    target detection    target tracking    human-computer interaction
收稿日期: 2022-04-24     
PACS:  R318  
基金资助:国家自然科学基金 (61801417);江苏省大学生创新创业训练计划项目(202111117056Y)
通讯作者: *E-mail: zhangxinmeinihao@126.com   
引用本文:   
张新峰, 殷文斌, 方金鹏, 张新梅. 基于U-Net卷积神经网络的目标分子人机交互观测方法[J]. 中国生物医学工程学报, 2024, 43(5): 582-595.
Zhang Xinfeng, Yin Wenbin, Fang Jinpeng, Zhang Xinmei. An Observation Method of Human-Computer Interaction for Identification of Target MoleculesBased on U-Net Convolutional Neural Network. Chinese Journal of Biomedical Engineering, 2024, 43(5): 582-595.
链接本文:  
http://cjbme.csbme.org/CN/10.3969/j.issn.0258-8021.2024.05.007     或     http://cjbme.csbme.org/CN/Y2024/V43/I5/582
版权所有 © 2015 《中国生物医学工程学报》编辑部
本系统由北京玛格泰克科技发展有限公司设计开发