摘要
在神经科学和计算机科学等领域,研究人员通过统计模型和深度学习等方法探索不同状态间功能性脑网络工作机制的差别;但现有的功能性脑网络研究工具多用于寻找支持某种假设的证据或传达科学发现,存在功能单一的缺点;针对上述问题,文章设计并实现了一个用于功能性核磁共振数据的交互式可视分析系统BrainDVis,帮助研究人员寻找不同状态间功能性脑网络的多方面差异;BrainDVis将功能性脑网络差异分析功能、网络特征参数分析功能、模块化结构分析功能、功能性连接分析功能相关联,提供多视图协同交互的方法帮助研究人员自主探索,寻找差异;最后使用公开数据集进行实验,验证了系统的可行性和有效性。
In the fields of neuroscience and computer science,researchers generally utilize statistical models and deep learning to explore differences in the internal working mechanism of the functional brain networks between different states.However,existing functional brain networks research methods are mainly used to find evidence that supports a certain hypothesis and convey scientific discoveries,these methods have single functions.In response to the above problems,we designed BrainDVis,an interactive visual analysis tool for functional MRI data,which can be used to help researchers explore the many differences between functional brain networks in different states.BrainDVis correlates functional brain networks difference comparison function,brain network parameter analysis function,modular structure analysis function,and brain functional connection analysis function,provides Multi-View Collaborative Interaction to help researchers explore independently and find differences.Finally,using public data set to conduct experiments,the feasibility and effectiveness of the system are verified.
作者
张振兴
吴亚东
廖竞
王娇
Zhang Zhenxing;Wu Yadong;Liao Jing;Wang jiao(School of Computer Science and Technology,Southwest University of Science and Technology,Mianyang,Sichuan 621000,China;School of Computer,Sichuan University of Science&Engineering,Zigong,Sichuan 643000,China;School of Information Engineering,Southwest University of Science and Technology,Mianyang,Sichuan 621000,China)
出处
《计算机测量与控制》
2021年第1期220-226,共7页
Computer Measurement &Control
基金
国家自然科学基金(61802320,61872304)
国家重点研发计划资助项目(2016QY04W0801)
国防基础科研计划(JCKY 2018404C001)。
关键词
可视分析
功能性脑网络
功能性核磁共振(fMRI)
多视图协同交互
Visual Analysis
functional brain networks
Functional Magnetic Resonance Imaging(fMRI)
Multi-View Collaborative Interaction