用于脑功能磁共振成像的独立成分分析并未选择独立性。
Independent component analysis for brain fMRI does not select for independence.
作者信息
Daubechies I, Roussos E, Takerkart S, Benharrosh M, Golden C, D'Ardenne K, Richter W, Cohen J D, Haxby J
机构信息
Center for the Study of Brain, Mind and Behavior, Princeton University, Princeton, NJ 08544, USA.
出版信息
Proc Natl Acad Sci U S A. 2009 Jun 30;106(26):10415-22. doi: 10.1073/pnas.0903525106. Epub 2009 Jun 25.
Abstract
InfoMax and FastICA are the independent component analysis algorithms most used and apparently most effective for brain fMRI. We show that this is linked to their ability to handle effectively sparse components rather than independent components as such. The mathematical design of better analysis tools for brain fMRI should thus emphasize other mathematical characteristics than independence.
摘要
信息最大化算法（InfoMax）和快速独立成分分析算法（FastICA）是脑功能磁共振成像（fMRI）中最常用且显然最有效的独立成分分析算法。我们表明，这与它们有效处理稀疏成分而非独立成分本身的能力有关。因此，用于脑fMRI的更好分析工具的数学设计应强调除独立性之外的其他数学特征。
相似文献
1
Independent component analysis for brain fMRI does not select for independence.
Proc Natl Acad Sci U S A. 2009 Jun 30;106(26):10415-22. doi: 10.1073/pnas.0903525106. Epub 2009 Jun 25.
2
Improved FastICA algorithm in fMRI data analysis using the sparsity property of the sources.
J Neurosci Methods. 2016 Apr 1;263:103-14. doi: 10.1016/j.jneumeth.2016.02.010. Epub 2016 Feb 12.
3
SACICA: a sparse approximation coefficient-based ICA model for functional magnetic resonance imaging data analysis.
J Neurosci Methods. 2013 May 30;216(1):49-61. doi: 10.1016/j.jneumeth.2013.03.014. Epub 2013 Apr 4.
4
Unsupervised learning and mapping of active brain functional MRI signals based on hidden semi-Markov event sequence models.
IEEE Trans Med Imaging. 2005 Feb;24(2):263-76. doi: 10.1109/tmi.2004.841225.
5
A variant of logistic transfer function in Infomax and a postprocessing procedure for independent component analysis applied to fMRI data.
Magn Reson Imaging. 2007 Jun;25(5):703-11. doi: 10.1016/j.mri.2006.09.038. Epub 2007 Mar 26.
6
WASICA: An effective wavelet-shrinkage based ICA model for brain fMRI data analysis.
J Neurosci Methods. 2015 May 15;246:75-96. doi: 10.1016/j.jneumeth.2015.03.011. Epub 2015 Mar 16.
7
Comparison of two exploratory data analysis methods for fMRI: unsupervised clustering versus independent component analysis.
IEEE Trans Inf Technol Biomed. 2004 Sep;8(3):387-98. doi: 10.1109/titb.2004.834406.
8
Independent component analysis for brain FMRI does indeed select for maximal independence.
PLoS One. 2013 Aug 29;8(8):e73309. doi: 10.1371/journal.pone.0073309. eCollection 2013.
9
Evaluation of PCA and ICA of simulated ERPs: Promax vs. Infomax rotations.
Hum Brain Mapp. 2007 Aug;28(8):742-63. doi: 10.1002/hbm.20304.
10
Analysis of fMRI data by blind separation into independent spatial components.
Hum Brain Mapp. 1998;6(3):160-88. doi: 10.1002/(SICI)1097-0193(1998)6:3<160::AID-HBM5>3.0.CO;2-1.
引用本文的文献
1
Simultaneously capturing excessive variations and smooth dynamics of the underlying neural activity using spatiotemporal basis expansion and multisubject fMRI data.
Sci Rep. 2025 Apr 20;15(1):13638. doi: 10.1038/s41598-025-97651-7.
2
Sparse Independent Component Analysis with an Application to Cortical Surface fMRI Data in Autism.
J Am Stat Assoc. 2024;119(548):2508-2520. doi: 10.1080/01621459.2024.2370593. Epub 2024 Jul 29.
3
Three layered sparse dictionary learning algorithm for enhancing the subject wise segregation of brain networks.
Sci Rep. 2024 Aug 17;14(1):19070. doi: 10.1038/s41598-024-69647-2.
4
A novel subject-wise dictionary learning approach using multi-subject fMRI spatial and temporal components.
Sci Rep. 2023 Nov 18;13(1):20201. doi: 10.1038/s41598-023-47420-1.
5
Computing personalized brain functional networks from fMRI using self-supervised deep learning.
Med Image Anal. 2023 Apr;85:102756. doi: 10.1016/j.media.2023.102756. Epub 2023 Jan 21.
6
Magnetoencephalography can reveal deep brain network activities linked to memory processes.
Hum Brain Mapp. 2022 Oct 15;43(15):4733-4749. doi: 10.1002/hbm.25987. Epub 2022 Jun 29.
7
Discriminating mild traumatic brain injury using sparse dictionary learning of functional network dynamics.
Brain Behav. 2021 Dec;11(12):e2414. doi: 10.1002/brb3.2414. Epub 2021 Nov 13.
8
Low-Rank Tucker-2 Model for Multi-Subject fMRI Data Decomposition With Spatial Sparsity Constraint.
IEEE Trans Med Imaging. 2022 Mar;41(3):667-679. doi: 10.1109/TMI.2021.3122226. Epub 2022 Mar 2.
9
Brain Connectivity Studies on Structure-Function Relationships: A Short Survey with an Emphasis on Machine Learning.
Comput Intell Neurosci. 2021 May 27;2021:5573740. doi: 10.1155/2021/5573740. eCollection 2021.
10
Sparse representation and dictionary learning model incorporating group sparsity and incoherence to extract abnormal brain regions associated with schizophrenia.
IEEE Access. 2020;8:104396-104406. doi: 10.1109/access.2020.2999513. Epub 2020 Jun 3.
本文引用的文献
1
Fast and robust fixed-point algorithms for independent component analysis.
IEEE Trans Neural Netw. 1999;10(3):626-34. doi: 10.1109/72.761722.
2
Advances in functional and structural MR image analysis and implementation as FSL.
Neuroimage. 2004;23 Suppl 1:S208-19. doi: 10.1016/j.neuroimage.2004.07.051.
3
The chronoarchitecture of the human brain--natural viewing conditions reveal a time-based anatomy of the brain.
Neuroimage. 2004 May;22(1):419-33. doi: 10.1016/j.neuroimage.2004.01.007.
4
Probabilistic independent component analysis for functional magnetic resonance imaging.
IEEE Trans Med Imaging. 2004 Feb;23(2):137-52. doi: 10.1109/TMI.2003.822821.
5
Latency (in)sensitive ICA. Group independent component analysis of fMRI data in the temporal frequency domain.
Neuroimage. 2003 Nov;20(3):1661-9. doi: 10.1016/s1053-8119(03)00411-7.
6
Spatial independent component analysis of functional MRI time-series: to what extent do results depend on the algorithm used?
Hum Brain Mapp. 2002 Jul;16(3):146-57. doi: 10.1002/hbm.10034.
7
Distributed and overlapping representations of faces and objects in ventral temporal cortex.
Science. 2001 Sep 28;293(5539):2425-30. doi: 10.1126/science.1063736.
8
Can recent innovations in harmonic analysis 'explain' key findings in natural image statistics?
Network. 2001 Aug;12(3):371-93.
9
Spatial and temporal independent component analysis of functional MRI data containing a pair of task-related waveforms.
Hum Brain Mapp. 2001 May;13(1):43-53. doi: 10.1002/hbm.1024.
10
Generalizable patterns in neuroimaging: how many principal components?
Neuroimage. 1999 May;9(5):534-44. doi: 10.1006/nimg.1998.0425.
Zotero 插件