Filters








523 Hits in 8.6 sec

Sparsity and Independence: Balancing Two Objectives in Optimization for Source Separation with Application to fMRI Analysis

Zois Boukouvalas, Yuri Levin-Schwartz, Vince D. Calhoun, Tülay Adalı
2018 Journal of the Franklin Institute  
algorithm and provide guidance on how to balance these two objectives in real world applications where the ground truth is not available.  ...  Two popular types of diversity that have proven useful for many applications are statistical independence and sparsity.  ...  making it difficult to balance 90 these two different forms of diversity, independence and sparsity.  ... 
doi:10.1016/j.jfranklin.2017.07.003 fatcat:msduek56lne6bbvrh2j2kggn5a

The Dangers of Following Trends in Research: Sparsity and Other Examples of Hammers in Search of Nails

Tulay Adali, H. Joel Trussell, Lars Kai Hansen, Vince D. Calhoun
2018 Proceedings of the IEEE  
They are enjoying unprecedented attention nowadays with deep nets, in large part due to the availability of large amounts of data, coupled with significantly increased computing power.  ...  Exciting new directions in any area are, of course, positive, especially in research, as they tend to bring significantly increased activity and fuel discoveries.  ...  DISC USSION We have warned against the dangers of the hype a new tool generates and how this impacts our research landscape in multiple ways. It is highly unlikely  ... 
doi:10.1109/jproc.2018.2823428 fatcat:niv536johjdxrijrtj2hpn5vre

Connecting natural and artificial neural networks in functional brain imaging using structured sparsity [article]

Christopher R Cox, Timothy T Rogers
2018 bioRxiv   pre-print
Artificial neural network models have long proven useful for understanding healthy, disordered, and developing cognition, but this work has often proceeded with little connection to functional brain imaging  ...  Using a simple model to generate synthetic data, we show that four contemporary methods each have critical and complementary blind-spots for detecting distributed signal.  ...  Sparse Overlapping Sets lasso for multitask learning and its application to fMRI analysis. in Advances in Neural Information Processing Systems (2013). 26. Rao, N., Nowak, R., Cox, C. & Rogers, T.  ... 
doi:10.1101/390534 fatcat:oh6ksea5fbhhdmpr3ztx5uyfiy

Application of Polynomial Spline Independent Component Analysis to fMRI Data [chapter]

Atsushi Kawaguchi, Young K., Xuemei Huang
2012 Independent Component Analysis for Audio and Biosignal Applications  
A new learning algorithm for blind signal separation, Advances in Neural Information Processing Systems 8: 757-763. 207 Application of Polynomial Spline Independent Component Analysis to fMRI Data Fig  ...  Application To demonstrate the applicability of the proposed method to real data, we separate fMRI data into independent spatial components that can be used to determine three-dimensional brain maps.  ... 
doi:10.5772/50343 fatcat:u77kdmipwfe37kiv53nwcoiehm

s-SMOOTH: Sparsity and Smoothness Enhanced EEG Brain Tomography

Ying Li, Jing Qin, Yue-Loong Hsin, Stanley Osher, Wentai Liu
2016 Frontiers in Neuroscience  
TGV is able to preserve the source edge and recover the spatial distribution of the source intensity with high accuracy.  ...  In this paper, we propose a novel Sparsity and SMOOthness enhanced brain TomograpHy (s-SMOOTH) method to improve the reconstruction accuracy by integrating two recently proposed regularization techniques  ...  ACKNOWLEDGMENTS The authors would like to thank Dr. Ming Yan for the inspiring discussion on TV and TGV. This work was supported in part by the California Capital Equity LLC and the Keck foundation.  ... 
doi:10.3389/fnins.2016.00543 pmid:27965529 pmcid:PMC5125305 fatcat:wysrib6pvfhlhecn5im6o2mglm

Quantification and normalization of noise variance with sparsity regularization to enhance diffuse optical tomography

Jixing Yao, Fenghua Tian, Yothin Rakvongthai, Soontorn Oraintara, Hanli Liu
2015 Biomedical Optics Express  
Specifically, we have implemented this quantification of noise variance to normalize the measurement signals from all source-detector channels along with sparsity regularization to provide high-quality  ...  The reconstructed images demonstrate that quantification and normalization of noise variance with sparsity regularization (QNNVSR) is an effective reconstruction approach to greatly enhance the spatial  ...  Figure 9 illustrates clearly that two targeted objects are well separated with SR and QNNVSR, while Tikhonov and GLS could not resolve two objects at all with much blurry cut-off edges.  ... 
doi:10.1364/boe.6.002961 pmid:26309760 pmcid:PMC4541524 fatcat:hcqrtd3aija3dehf27ds23ge2i

SISSY: An efficient and automatic algorithm for the analysis of EEG sources based on structured sparsity

H. Becker, L. Albera, P. Comon, J.-C. Nunes, R. Gribonval, J. Fleureau, P. Guillotel, I. Merlet
2017 NeuroImage  
To overcome these problems, we propose to include an additional regulariza- * tion term that imposes sparsity in the original source domain and to solve the resulting optimization problem using the alternating  ...  However, this technique suffers from several problems: it leads to amplitude-biased source estimates, it has difficulties in separating close sources, and it has a high computational complexity due to  ...  Therefore, SISSY proves to be a promising method for the reconstruction of extended brain sources, in particular for application in epilepsy.  ... 
doi:10.1016/j.neuroimage.2017.05.046 pmid:28576413 fatcat:wfa37aj2vrbsvexnurmvpjmrli

Deep neural network with weight sparsity control and pre-training extracts hierarchical features and enhances classification performance: Evidence from whole-brain resting-state functional connectivity patterns of schizophrenia

Junghoe Kim, Vince D. Calhoun, Eunsoo Shim, Jong-Hwan Lee
2016 NeuroImage  
weight sparsity in each hidden layer via L 1 -norm regularization.  ...  The objective of this study was to adopt the DNN for whole-brain restingstate FC pattern classification of schizophrenia (SZ) patients vs. healthy controls (HCs) and identification of aberrant FC patterns  ...  These sponsors had no involvement in the study design, data collection, analysis or interpretation of data, manuscript preparation, or the decision to submit for publication.  ... 
doi:10.1016/j.neuroimage.2015.05.018 pmid:25987366 pmcid:PMC4644699 fatcat:qttutmjktzgc7gdk7r5ev7k5ve

Temporally-independent functional modes of spontaneous brain activity

S. M. Smith, K. L. Miller, S. Moeller, J. Xu, E. J. Auerbach, M. W. Woolrich, C. F. Beckmann, M. Jenkinson, J. Andersson, M. F. Glasser, D. C. Van Essen, D. A. Feinberg (+2 others)
2012 Proceedings of the National Academy of Sciences of the United States of America  
Soon after spatial ICA was first proposed for task FMRI (3), temporal ICA (optimizing for temporal independence between components) was also applied to task FMRI (4).  ...  For these reasons, nearly all applications of ICA to FMRI (including resting-state FMRI) have to date used spatial ICA.  ...  Biswal BB, Ulmer JL (1999) Blind source separation of multiple signal sources of fMRI data sets using independent component analysis. J Comput Assist Tomogr 23: 265-271. 5.  ... 
doi:10.1073/pnas.1121329109 pmid:22323591 pmcid:PMC3286957 fatcat:vrjvkoqunrbd7fxmivzhllsc4e

Development of ICA and IVA Algorithms with Application to Medical Image Analysis [article]

Zois Boukouvalas
2018 arXiv   pre-print
Independent component analysis (ICA) is a widely used BSS method that can uniquely achieve source recovery, subject to only scaling and permutation ambiguities, through the assumption of statistical independence  ...  Independent vector analysis (IVA) extends the applicability of ICA by jointly decomposing multiple datasets through the exploitation of the dependencies across datasets.  ...  well as fMRI-like data; -Explore the trade-offs between independence and sparsity in the ICA optimization framework and provide a guidance on how to balance these two objectives in real world applications  ... 
arXiv:1801.08600v1 fatcat:nu2xlytexrcnnblws7zieo77be

LOCUS: A Novel Decomposition Method for Brain Network Connectivity Matrices using Low-rank Structure with Uniform Sparsity [article]

Yikai Wang, Ying Guo
2020 arXiv   pre-print
In this paper, we propose a novel blind source separation method with low-rank structure and uniform sparsity (LOCUS) as a fully data-driven decomposition method for network measures.  ...  Compared with the existing method that vectorizes connectivity matrices ignoring brain network topology, LOCUS achieves more efficient and accurate source separation for connectivity matrices using low-rank  ...  Philadelphia Neurodevelopmental Cohort: Support for the collection of the data sets was provided by grant RC2MH089983 awarded to Raquel Gur and RC2MH089924 awarded to Hakon Hakorson.  ... 
arXiv:2008.08915v1 fatcat:4x2crrii45aktmjsetsjw3lgv4

Rethinking Bayesian Learning for Data Analysis: The Art of Prior and Inference in Sparsity-Aware Modeling [article]

Lei Cheng, Feng Yin, Sergios Theodoridis, Sotirios Chatzis, Tsung-Hui Chang
2022 arXiv   pre-print
To implement sparsity-aware learning, the crucial point lies in the choice of the function regularizer for discriminative methods and the choice of the prior distribution for Bayesian learning.  ...  Sparse modeling for signal processing and machine learning has been at the focus of scientific research for over two decades.  ...  network analysis [43] , "blind source separation" in electroencephalogram (EEG) and functional magnetic resonance imaging (fMRI) data analysis [44] , [45] , and "blind signal estimation" in radar/sonar  ... 
arXiv:2205.14283v1 fatcat:tbh4z5qasvgvrp5vhwi4azq64y

Revisiting sparse ICA from a synthesis point of view: Blind Source Separation for over and underdetermined mixtures

Fangchen Feng, Matthieu Kowalski
2018 Signal Processing  
Recently, it was claimed in [19] that two of the most used ICA methods for fMRI (Infomax and FastICA, see e.g. [6]) separate sparse sources rather than independent sources, leading to the conclusion that  ...  This conclusion is balanced in [20] where the authors show that these two algorithms are indeed relevant to the recovery of independent fMRI sources. Contributions and outline.  ...  ICA in the transform domain has been proposed for image processing [24] with wavelets, audio source separation with the Short Time Fourier Transform (STFT) [11] and fMRI [25] where a dictionary learning  ... 
doi:10.1016/j.sigpro.2018.05.017 fatcat:ghnpsyqlv5h3pjpyegzm6m6ioq

Constructing Nonlinear Discriminants from Multiple Data Views [chapter]

Tom Diethe, David Roi Hardoon, John Shawe-Taylor
2010 Lecture Notes in Computer Science  
There are many situations in which we have more than one view of a single data source, or in which we have multiple sources of data that are aligned.  ...  We show that our formulations are justified from both probabilistic and learning theory perspectives. We then extend the optimisation problem to account for directions unique to each view (PMFDA).  ...  It is known that Fisher Discriminant Analysis (FDA) is Bayes optimal for two Gaussian distributions with equal covariance in the input space.  ... 
doi:10.1007/978-3-642-15880-3_27 fatcat:7kgiskqtvnadxdftt2dxzdfokq

Multimodal fusion of structural and functional brain imaging in depression using linked independent component analysis

Luigi A. Maglanoc, Tobias Kaufmann, Rune Jonassen, Eva Hilland, Dani Beck, Nils Inge Landrø, Lars T. Westlye
2019 Human Brain Mapping  
We used linked independent component analysis to fuse cortical macrostructure (thickness, area, gray matter density), white matter diffusion properties and resting-state functional magnetic resonance imaging  ...  In contrast, there were no significant associations with case-control status, nor symptom loads for depression and anxiety with the brain components, nor any interaction effects with age and sex.  ...  The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. All rights reserved.  ... 
doi:10.1002/hbm.24802 pmid:31571370 fatcat:g7y765irerdihhx4mx64nehug4
« Previous Showing results 1 — 15 out of 523 results