A spatiotemporal nonparametric Bayesian model of multi-subject fMRI data

Published on Jun 1, 2016in The Annals of Applied Statistics2.083
· DOI :10.1214/16-AOAS926
Linlin Zhang5
Estimated H-index: 5
(Rice University),
Michele Guindani21
Estimated H-index: 21
(University of Texas MD Anderson Cancer Center)
+ 2 AuthorsMarina Vannucci38
Estimated H-index: 38
(Rice University)
In this paper we propose a unified, probabilistically coherent framework for the analysis of task-related brain activity in multi-subject fMRI experiments. This is distinct from two-stage “group analysis” approaches traditionally considered in the fMRI literature, which separate the inference on the individual fMRI time courses from the inference at the population level. In our modeling approach we consider a spatiotemporal linear regression model and specifically account for the between-subjects heterogeneity in neuronal activity via a spatially informed multi-subject nonparametric variable selection prior. For posterior inference, in addition to Markov chain Monte Carlo sampling algorithms, we develop suitable variational Bayes algorithms. We show on simulated data that variational Bayes inference achieves satisfactory results at more reduced computational costs than using MCMC, allowing scalability of our methods. In an application to data collected to assess brain responses to emotional stimuli our method correctly detects activation in visual areas when visual stimuli are presented.
📖 Papers frequently viewed together
4 Authors (Linlin Zhang, ..., Marina Vannucci)
#1Wenguang Sun (SC: University of Southern California)H-Index: 14
#2Brian J. Reich (NCSU: North Carolina State University)H-Index: 33
Last. Armin Schwartzman (NCSU: North Carolina State University)H-Index: 20
view all 5 authors...
type="main" xml:id="rssb12064-abs-0001"> The paper develops a unified theoretical and computational framework for false discovery control in multiple testing of spatial signals. We consider both pointwise and clusterwise spatial analyses, and derive oracle procedures which optimally control the false discovery rate, false discovery exceedance and false cluster rate. A data-driven finite approximation strategy is developed to mimic the oracle procedures on a continuous spatial domain. Our multipl...
#1Linlin Zhang (Rice University)H-Index: 5
#2Michele Guindani (University of Texas MD Anderson Cancer Center)H-Index: 21
Last. Marina Vannucci (Rice University)H-Index: 38
view all 4 authors...
Abstract In this paper we present a novel wavelet-based Bayesian nonparametric regression model for the analysis of functional magnetic resonance imaging (fMRI) data. Our goal is to provide a joint analytical framework that allows to detect regions of the brain which exhibit neuronal activity in response to a stimulus and, simultaneously, infer the association, or clustering, of spatially remote voxels that exhibit fMRI time series with similar characteristics. We start by modeling the data with...
#1Timothy D. Johnson (UM: University of Michigan)H-Index: 62
#2Zhuqing Liu (UM: University of Michigan)H-Index: 2
Last. Thomas E. Nichols (Warw.: University of Warwick)H-Index: 91
view all 4 authors...
The Potts model has enjoyed much success as a prior model for image segmentation. Given the individual classes in the model, the data are typically modeled as Gaussian random variates or as random variates from some other parametric distribution. In this manuscript we present a non-parametric Potts model and apply it to an FMRI study for the pre-surgical assessment of peritumoral brain activation. In our model we assume that the Z-score image from a patient can be segmented into activated, deact...
#1Jaesik Jeong (IUPUI: Indiana University – Purdue University Indianapolis)H-Index: 8
#2Marina Vannucci (Rice University)H-Index: 38
Last. Kyungduk Ko (BSU: Boise State University)H-Index: 7
view all 3 authors...
Summary This article considers linear regression models with long memory errors. These models have been proven useful for application in many areas, such as medical imaging, signal processing, and econometrics. Wavelets, being self-similar, have a strong connection to long memory data. Here we employ discrete wavelet transforms as whitening filters to simplify the dense variance–covariance matrix of the data. We then adopt a Bayesian approach for the estimation of the model parameters. Our infer...
#1Nilotpal Sanyal (MU: University of Missouri)H-Index: 7
#2Marco A. R. Ferreira (MU: University of Missouri)H-Index: 14
Abstract We develop a methodology for Bayesian hierarchical multi-subject multiscale analysis of functional Magnetic Resonance Imaging (fMRI) data. We begin by modeling the brain images temporally with a standard general linear model. After that, we transform the resulting estimated standardized regression coefficient maps through a discrete wavelet transformation to obtain a sparse representation in the wavelet space. Subsequently, we assign to the wavelet coefficients a prior that is a mixture...
#1Karl J. Friston (Wellcome Trust Centre for Neuroimaging)H-Index: 241
Abstract Over the past 20 years, neuroimaging has become a predominant technique in systems neuroscience. One might envisage that over the next 20 years the neuroimaging of distributed processing and connectivity will play a major role in disclosing the brain's functional architecture and operational principles. The inception of this journal has been foreshadowed by an ever-increasing number of publications on functional connectivity, causal modeling, connectomics, and multivariate analyses of d...
Jun 14, 2011 in AISTATS (International Conference on Artificial Intelligence and Statistics)
#1Chong Wang (Princeton University)H-Index: 44
The hierarchical Dirichlet process (HDP) is a Bayesian nonparametric model that can be used to model mixed-membership data with a potentially infinite number of components. It has been applied widely in probabilistic topic modeling, where the data are documents and the components are distributions of terms that reflect recurring patterns (or “topics”) in the collection. Given a document collection, posterior inference is used to determine the number of topics needed and to characterize their dis...
#1Lee M. Harrison (Ebor: University of York)H-Index: 23
#2Gary G. R. Green (Ebor: University of York)H-Index: 45
Abstract Functional MRI provides a unique perspective of neuronal organization; however, these data include many complex sources of spatiotemporal variability, which require spatial preprocessing and statistical analysis. For the latter, Bayesian models provide a promising alternative to classical inference, which uses results from Gaussian random field theory to assess the significance of spatially correlated statistic images. A Bayesian approach generalizes the application of these ideas in th...
#1Alicia Quirós (URJC: King Juan Carlos University)H-Index: 10
#2Raquel Montes Diez (URJC: King Juan Carlos University)H-Index: 3
Last. Dani Gamerman (UFRJ: Federal University of Rio de Janeiro)H-Index: 22
view all 3 authors...
Abstract This research describes a new Bayesian spatiotemporal model to analyse block-design BOLD fMRI studies. In the temporal dimension, we parameterise the hemodynamic response function's (HRF) shape with a potential increase of signal and a subsequent exponential decay. In the spatial dimension, we use Gaussian Markov random fields (GMRF) priors on activation characteristics parameters (location and magnitude) that embody our prior knowledge that evoked responses are spatially contiguous and...
Dec 7, 2009 in NeurIPS (Neural Information Processing Systems)
#1Feng Yan (Purdue University)H-Index: 9
#2Ningyi Xu (Microsoft)H-Index: 17
Last. Yuan Qi (Purdue University)H-Index: 13
view all 3 authors...
The recent emergence of Graphics Processing Units (GPUs) as general-purpose parallel computing devices provides us with new opportunities to develop scalable learning methods for massive data. In this work, we consider the problem of parallelizing two inference methods on GPUs for latent Dirichlet Allocation (LDA) models, collapsed Gibbs sampling (CGS) and collapsed variational Bayesian (CVB). To address limited memory constraints on GPUs, we propose a novel data partitioning scheme that effecti...
Cited By39
This article introduces an R package to perform statistical analysis for task-based fMRI data at both individual and group levels. The analysis to detect brain activation at the individual level is based on modeling the fMRI signal using Matrix-Variate Dynamic Linear Models (MDLM). Therefore, the analysis for the group stage is based on posterior distributions of the state parameter obtained from the modeling at the individual level. In this way, this package offers several R functions with diff...
#1Johnatan Cardona Jiménez (USP: University of São Paulo)H-Index: 1
#2Carlos Eduardo Pereira (USP: University of São Paulo)H-Index: 70
Abstract null null A modeling procedure for task-based functional magnetic resonance imaging (fMRI) data analysis using a Bayesian matrix-variate dynamic linear model (MVDLM) is presented. With this type of model, less complex than the more traditional temporal-spatial models, it is possible to take into account the temporal and, at least locally, the spatial structures that are usually present in this type of data. Thus, every voxel in the brain image is jointly modeled with its nearest neighbo...
#1Gang Chen (NIH: National Institutes of Health)H-Index: 58
#2Paul A. Taylor (NIH: National Institutes of Health)H-Index: 22
Last. Luiz Pessoa (UMD: University of Maryland, College Park)H-Index: 67
view all 6 authors...
Abstract null Neuroimaging relies on separate statistical inferences at tens of thousands of spatial locations. Such massively univariate analysis typically requires an adjustment for multiple testing in an attempt to maintain the family-wise error rate at a nominal level of 5%. First, we examine three sources of substantial information loss that are associated with the common practice under the massively univariate framework: (a) the hierarchical data structures (spatial units and trials) are n...
#2Jeffrey S. MorrisH-Index: 51
In this paper, we introduce a new Bayesian approach for analyzing task fMRI data that simultaneously detects activation signatures and background connectivity. Our modeling involves a new hybrid tensor spatial-temporal basis strategy that enables scalable computing yet captures nearby and distant intervoxel correlation and long-memory temporal correlation. The spatial basis involves a composite hybrid transform with two levels: the first accounts for within-ROI correlation, and second between-RO...
#1Azam Saffar (Shahid Beheshti University of Medical Sciences and Health Services)
#2Vahid Malekian (UCL Institute of Neurology)H-Index: 3
Last. Yadollah Mehrabi (Shahid Beheshti University of Medical Sciences and Health Services)H-Index: 29
view all 0 authors...
Abstract null null Objective null Accuracy and precision of the statistical analysis methods used for brain activation maps are essential. Adjusting models to consider spatiotemporal correlation embedded in fMRI data may increase their accuracy, but it also introduces a high computational cost. The present study aimed to apply and assess the spatiotemporal Gaussian process (STGP) model to improve accuracy and reduce cost. null null null Methods null We applied the spatiotemporal Gaussian process...
#1Francesca Gasperoni (University of Cambridge)H-Index: 5
#2Alessandra Luati (UNIBO: University of Bologna)H-Index: 9
Last. Enzo D'Innocenzo (UNIBO: University of Bologna)H-Index: 1
view all 4 authors...
A simultaneous autoregressive score-driven model with autoregressive disturbances is developed for spatio-temporal data that may exhibit heavy tails. The model specification rests on a signal plus ...
#1Daniel SpencerH-Index: 1
#2Yu YueH-Index: 1
Last. Amanda F. MejiaH-Index: 12
view all 5 authors...
The general linear model (GLM) is a popular and convenient tool for estimating the functional brain response and identifying areas of significant activation during a task or stimulus. However, the classical GLM is based on a massive univariate approach that does not leverage the similarity of activation patterns among neighboring brain locations. As a result, it tends to produce noisy estimates and be underpowered to detect significant activations, particularly in individual subjects and small g...
#1Guanjie Chen (NIH: National Institutes of Health)H-Index: 35
#2Paul A. Taylor (NIH: National Institutes of Health)H-Index: 22
Last. Luiz Pessoa (UMD: University of Maryland, College Park)H-Index: 67
view all 6 authors...
Neuroimaging relies on separate statistical inferences at tens of thousands of spatial locations. Such massively univariate analysis typically requires adjustment for multiple testing in an attempt to maintain the family-wise error rate at a nominal level of 5%. We discuss how this approach is associated with substantial information loss because of an implicit but questionable assumption about the effect distribution across spatial units. To improve inference efficiency, predictive accuracy, and...
This paper presents a general framework for modeling dependence in multivariate time series. Its fundamental approach relies on decomposing each signal in a system into various frequency components and then studying the dependence properties through these oscillatory activities.The unifying theme across the paper is to explore the strength of dependence and possible lead-lag dynamics through filtering. The proposed framework is capable of representing both linear and non-linear dependencies that...
#1Tingting Zhang (University of Pittsburgh)H-Index: 9
#2Minh Pham (SFSU: San Francisco State University)H-Index: 2
Last. James A. Coan (UVA: University of Virginia)H-Index: 38
view all 6 authors...
Abstract Functional magnetic resonance imaging (fMRI) is one of the most popular neuroimaging technologies used in human brain studies. However, fMRI data analysis faces several challenges, including intensive computation due to the massive data size and large estimation errors due to a low signal-to-noise ratio of the data. A new statistical model and a computational algorithm are proposed to address these challenges. Specifically, a new multi-subject general linear model is built for stimulus-...
This website uses cookies.
We use cookies to improve your online experience. By continuing to use our website we assume you agree to the placement of these cookies.
To learn more, you can find in our Privacy Policy.