Two-way principal component analysis for matrix-variate data, with an application to functional magnetic resonance imaging data.

Many modern neuroimaging studies acquire large spatial images of the brain observed sequentially over time. Such data are often stored in the forms of matrices. To model these matrix-variate data we introduce a class of separable processes using explicit latent process modeling. To account for the size and two-way structure of the data, we extend principal component analysis to achieve dimensionality reduction at the individual level. We introduce necessary identifiability conditions for each model and develop scalable estimation procedures. The method is motivated by and applied to a functional magnetic resonance imaging study designed to analyze the relationship between pain and brain activity.

[1]  Sheng Luo,et al.  Population Value Decomposition, a Framework for the Analysis of Image Populations , 2011, Journal of the American Statistical Association.

[2]  Yul-Wan Sung,et al.  Functional magnetic resonance imaging , 2004, Scholarpedia.

[3]  Brian Caffo,et al.  Longitudinal functional principal component analysis. , 2010, Electronic journal of statistics.

[4]  M. Lindquist.  The Statistical Analysis of fMRI Data. , 2008, 0906.3662.

[5]  D. Bosq.  Linear Processes in Function Spaces: Theory And Applications , 2000 .

[6]  H. Müller,et al.  Functional Data Analysis for Sparse Longitudinal Data , 2005 .

[7]  H. Akaike.  A new look at the statistical model identification , 1974 .

[8]  Martin A. Lindquist,et al.  Brain mediators of the effects of noxious heat on pain , 2014, PAIN®.

[9]  C. Loan,et al.  Approximation with Kronecker Products , 1992 .

[10]  Erkki Oja,et al.  Independent component analysis: algorithms and applications , 2000, Neural Networks.

[11]  Genevera I. Allen,et al.  Journal of the American Statistical Association a Generalized Least-square Matrix Decomposition a Generalized Least-square Matrix Decomposition , 2022 .

[12]  Roberto Viviani,et al.  Functional principal component analysis of fMRI data , 2005, Human brain mapping.

[13]  Luo Xiao,et al.  Fast bivariate P‐splines: the sandwich smoother , 2013 .

[14]  Joan G. Staniswalis,et al.  Nonparametric Regression Analysis of Longitudinal Data , 1998 .

[15]  Ciprian M. Crainiceanu,et al.  Two-stage decompositions for the analysis of functional connectivity for fMRI with application to Alzheimer's disease risk , 2010, NeuroImage.

[16]  J. Pekar,et al.  A method for making group inferences from functional MRI data using independent component analysis , 2001, Human brain mapping.

[17]  Ciprian M Crainiceanu,et al.  Multilevel sparse functional principal component analysis , 2014, Stat.

[18]  Peter D Hoff,et al.  SEPARABLE FACTOR ANALYSIS WITH APPLICATIONS TO MORTALITY DATA. , 2012, The annals of applied statistics.

[19]  D. Zimmerman,et al.  The likelihood ratio test for a separable covariance matrix , 2005 .

[20]  Su-Yun Huang,et al.  On multilinear principal component analysis of order-two tensors , 2011, 1104.5281.

[21]  Brian S. Caffo,et al.  Multilevel functional principal component analysis , 2009 .

[22]  Martin A. Lindquist,et al.  Functional Causal Mediation Analysis With an Application to Brain Connectivity , 2012, Journal of the American Statistical Association.

[23]  B. Silverman,et al.  Estimating the mean and covariance structure nonparametrically when the data are curves , 1991 .

[24]  I. Jolliffe.  Principal Component Analysis , 2002 .

[25]  P. Petrovic,et al.  Placebo and Opioid Analgesia-- Imaging a Shared Neuronal Network , 2002, Science.

[26]  Ciprian M Crainiceanu,et al.  Structured functional principal component analysis , 2013, Biometrics.

[27]  N. Bolger,et al.  Brain Mediators of Predictive Cue Effects on Perceived Pain , 2010, The Journal of Neuroscience.

[28]  Daoqiang Zhang,et al.  (2D)2PCA: Two-directional two-dimensional PCA for efficient face representation and recognition , 2005, Neurocomputing.

[29]  Age K. Smilde,et al.  ANOVA-simultaneous component analysis (ASCA): a new tool for analyzing designed metabolomics data , 2005, Bioinform..

[30]  E. Donchin,et al.  Spatiotemporal analysis of the late ERP responses to deviant stimuli. , 2001, Psychophysiology.

[31]  E. Donchin,et al.  Localization of the event-related potential novelty response as defined by principal components analysis. , 2003, Brain research. Cognitive brain research.