| 新的ICA算法实现成组fMRI信号盲分离 |
| |
| 引用本文： | 唐焕文,张伟伟,史振威,等. 新的ICA算法实现成组fMRI信号盲分离[J]. 大连理工大学学报, 2007, 47(5): 773-776 |
| |
| 作者姓名： | 唐焕文 张伟伟 史振威 等 |
| |
| 作者单位： | 大连理工大学,计算生物学和生物信息学研究所,辽宁,大连,116024;大连理工大学,计算生物学和生物信息学研究所,辽宁,大连,116024;大连理工大学,神经信息学研究所,辽宁,大连,116024;清华大学,自动化系,北京,100084;大连理工大学,神经信息学研究所,辽宁,大连,116024 |
| |
| 摘 要： | 独立成分分析(ICA)方法已被成功地用于处理功能磁共振成像(fMRI)信号,但主要是用于处理单个被试的fMRI信号,对于多个被试的情况却很少考虑.为此利用一种扩展的ICA方法--Group ICA来处理多个被试的fMRI信号,结果表明这种方法在保证结果准确性的前提下,可以大大减少计算量,快速获得统计结果.计算中应用的是NewFP算法,统计结果表明这种算法在估计激活的时间动力学准确性上优于FastICA算法.
|
| 关 键 词： | 独立成分分析 盲源分离 功能磁共振成像 |
| 文章编号： | 1000-8608（2007）05-0773-04 |
| 修稿时间： | 2005-12-20 |
Blind source separation for group fMRI signals using a new independent component analysis algorithm |
| |
| TANG Huan-wen,ZHANG Wei-wei,SHI Zhen-wei,et al. Blind source separation for group fMRI signals using a new independent component analysis algorithm[J]. Journal of Dalian University of Technology, 2007, 47(5): 773-776 |
| |
| Authors: | TANG Huan-wen ZHANG Wei-wei SHI Zhen-wei et al |
| |
| Affiliation: | 1. Inst. of Comput. Biol. and Bioinformatics, Dalian Univ. of Technol., Dalian 116024, China; 2. Inst. of Neuroinformatics, Dallan Univ. of Technol., Dalian 116024, China; 3. Dept. of Autom., Tsinghua Univ., Beijing 100084, China |
| |
| Abstract: | Independent component analysis(ICA) has been used effectively for processing the functional magnetic resonance imaging(fMRI) data,but usually the data come from one subject.To process the signals from a group of subjects,an extended independent component analysis method,Group ICA is proposed.The results show that this method can reduce the data and receive the statistical result fast.In the processing,an independent component analysis method named new fixed-point is used,and the results show that the new method is superior to the FastICA on the accuracy of estimating the temporal dynamics of activations. |
| |
| Keywords: | independent component analysis blind source separation functional magnetic resonance imaging |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《大连理工大学学报》浏览原始摘要信息 |
|
点击此处可从《大连理工大学学报》下载全文 |
|
