Independent Component Analysis for Brain fMRI Does Indeed Select for Maximal Independence

Calhoun, Vince D.; Potluru, Vamsi K.; Phlypo, Ronald; Silva, Rogers F.; Pearlmutter, Barak A.; Caprihan, Arvind; Plis, Sergey M.; Adalı, Tülay

doi:10.1371/journal.pone.0073309

Cited by 76 publications

(60 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, Daubechies et al (2009) claimed that two ICA algorithms, InfomaxICA and FastICA, select for sparsity rather than independence. The follow-up work by Calhoun et al (2013) showed that the claims made in Daubechies et al (2009) were not supported by experimental evidence and the ICA algorithms are indeed identifying maximally independent sources (Calhoun et al, 2013). Our study focuses on using sparse modeling for fMRI analysis.…”

Section: Discussionmentioning

confidence: 95%

A two-step super-Gaussian independent component analysis approach for fMRI data

Yao

Zhang

et al. 2015

NeuroImage

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 95%

A two-step super-Gaussian independent component analysis approach for fMRI data

Yao

Zhang

et al. 2015

NeuroImage

View full text Add to dashboard Cite

“…Spatial Independent Component Analysis (ICA) of the reconstructed fMRI data and the original fully available fMRI dataset is performed via GIFT toolbox 2 . ICA is a data driven method that has been widely used in resting state fMRI to recover the set of spatially independent brain RSNs [38,39,40,41,42].…”

Section: Qualitative Analysismentioning

confidence: 99%

Double temporal sparsity based accelerated reconstruction of compressively sensed resting-state fMRI

Aggarwal

Gupta

2017

Computers in Biology and Medicine

View full text Add to dashboard Cite

A number of reconstruction methods have been proposed recently for accelerated functional Magnetic Resonance Imaging (fMRI) data collection. However, existing methods suffer with the challenge of greater artifacts at high acceleration factors. This paper addresses the issue of accelerating fMRI collection via undersampled k -space measurements combined with the proposed Double Temporal Sparsity based Reconstruction (DTSR) method with the l 1 − l 1 norm constraint. The robustness of the proposed DTSR method has been thoroughly evaluated both at the subject level and at the group level on real fMRI data. Results are presented at various acceleration factors. Quantitative analysis in terms of Peak Signal-to-Noise Ratio (PSNR) and other metrics, and qualitative analysis in terms of reproducibility of brain Resting State Networks (RSNs) demonstrate that the proposed method is accurate and robust. In addition, the proposed DTSR method preserves brain networks that are important for studying fMRI data. Compared to the existing accelerated fMRI reconstruction methods, the DTSR method shows promising potential with an improvement of 10-12dB in PSNR with acceleration factors upto 3.5. Simulation results on real data demonstrate that DTSR method can be used to acquire accelerated fMRI with accurate detection of RSNs.

show abstract

“…It assumes that the observed signals are mixtures of statistically independent sources [5]. Therefore, it aims to decompose the mixed signals into the independent sources.…”

Section: Problem Formulationmentioning

confidence: 99%

“…In order to produce physiologically interpretable robust features, ICA has been applied to brain imaging data. Successful application of ICA on fMRI can be attributed to sparsity [6] and statistical independence between the underlying sources [5].…”

Section: Problem Formulationmentioning

confidence: 99%

Privacy-preserving source separation for distributed data using independent component analysis

Imtiaz

Silva

Baker

et al. 2016

2016 Annual Conference on Information Science and Systems (CISS)

Self Cite

View full text Add to dashboard Cite

Building good feature representations and learning hidden source models typically requires large sample sizes. In many applications, however, the size of the sample at an individual data holder may not be sufficient. One such application is neuroimaging analyses for mental health disorders -there are many individual research groups, each with a moderate number of subjects. Pooling such data can enable efficient feature learning, but privacy concerns prevent sharing the underlying data. We propose a model for private feature learning in which the data holders share differentially private views of their respective datasets to enable collaborative learning of a joint feature map. We give an example of such an algorithm for independent component analysis (ICA) -a popular blind source separation algorithm used in neuroimaging analyses. Our algorithm is a differentially private version of the recently proposed distributed joint ICA algorithm. We evaluate the performance of this method on simulated functional magnetic resonance imaging (fMRI) data.

show abstract

Independent Component Analysis for Brain fMRI Does Indeed Select for Maximal Independence

Cited by 76 publications

References 16 publications

A two-step super-Gaussian independent component analysis approach for fMRI data

A two-step super-Gaussian independent component analysis approach for fMRI data

Double temporal sparsity based accelerated reconstruction of compressively sensed resting-state fMRI

Privacy-preserving source separation for distributed data using independent component analysis

Contact Info

Product

Resources

About