Results 1  10
of
45
Bayesian fMRI time series analysis with spatial priors
 NeuroImage
, 2005
"... We describe a Bayesian estimation and inference procedure for fMRI time series based on the use of General Linear Models (GLMs). Importantly, we use a spatial prior on regression coefficients which embodies our prior knowledge that evoked responses are spatially contiguous and locally homogeneous. F ..."
Abstract

Cited by 43 (13 self)
 Add to MetaCart
(Show Context)
We describe a Bayesian estimation and inference procedure for fMRI time series based on the use of General Linear Models (GLMs). Importantly, we use a spatial prior on regression coefficients which embodies our prior knowledge that evoked responses are spatially contiguous and locally homogeneous. Further, using a computationally efficient Variational Bayes framework, we are able to let the data determine the optimal amount of smoothing. We assume an arbitrary order AutoRegressive (AR) model for the errors. Our model generalizes earlier work on voxelwise estimation of GLMAR models and inference in GLMs using Posterior Probability Maps (PPMs). Results are shown on simulated data and on data from an eventrelated fMRI experiment.
Multilevel linear modelling for FMRI group analysis using Bayesian inference
 Neuroimage
, 2004
"... Functional magnetic resonance imaging studies often involve the acquisition of data from multiple sessions and/or multiple subjects. A hierarchical approach can be taken to modelling such data with a general linear model (GLM) at each level of the hierarchy introducing different random effects varia ..."
Abstract

Cited by 41 (6 self)
 Add to MetaCart
Functional magnetic resonance imaging studies often involve the acquisition of data from multiple sessions and/or multiple subjects. A hierarchical approach can be taken to modelling such data with a general linear model (GLM) at each level of the hierarchy introducing different random effects variance components. Inferring on these models is nontrivial with frequentist solutions being unavailable. A solution is to use a Bayesian framework. One important ingredient in this is the choice of prior on the variance components and toplevel regression parameters. Due to the typically small numbers of sessions or subjects in neuroimaging, the choice of prior is critical. To alleviate this problem, we introduce to neuroimage modelling the approach of reference priors, which drives the choice of prior such that it is noninformative in an informationtheoretic sense. We propose two inference techniques at the top level for multilevel hierarchies (a fast approach and a slower more accurate approach). We also demonstrate that we can infer on the top level of multilevel hierarchies by inferring on the levels of the hierarchy separately and passing summary statistics of a noncentral multivariate t distribution between them.
Hybrid Dirichlet mixture models for functional data
"... Summary. In functional data analysis, curves or surfaces are observed, up to measurement error, at a finite set of locations, for, say, a sample of n individuals. Often, the curves are homogeneous, except perhaps for individualspecific regions that provide heterogeneous behaviour (e.g. ‘damaged ’ a ..."
Abstract

Cited by 10 (0 self)
 Add to MetaCart
(Show Context)
Summary. In functional data analysis, curves or surfaces are observed, up to measurement error, at a finite set of locations, for, say, a sample of n individuals. Often, the curves are homogeneous, except perhaps for individualspecific regions that provide heterogeneous behaviour (e.g. ‘damaged ’ areas of irregular shape on an otherwise smooth surface). Motivated by applications with functional data of this nature, we propose a Bayesian mixture model, with the aim of dimension reduction, by representing the sample of n curves through a smaller set of canonical curves. We propose a novel prior on the space of probability measures for a random curve which extends the popular Dirichlet priors by allowing local clustering: nonhomogeneous portions of a curve can be allocated to different clusters and the n individual curves can be represented as recombinations (hybrids) of a few canonical curves. More precisely, the prior proposed envisions a conceptual hidden factor with klevels that acts locally on each curve. We discuss several models incorporating this prior and illustrate its performance with simulated and real data sets. We examine theoretical properties of the proposed finite hybrid Dirichlet mixtures, specifically, their behaviour as the number of the mixture components goes to 1 and their connection with Dirichlet process mixtures.
Correlations and multiple comparisons in functional imaging – a statistical perspective
 Perspectives in Psychological Science
, 2009
"... Vul et al. claim in their paper that the correlations reported in fMRI studies are commonly overstated because researchers tend to report only the highest correlations, or only those correlations that exceed some threshold. Their paper has in a short time given rise to a spirited debate about key st ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
(Show Context)
Vul et al. claim in their paper that the correlations reported in fMRI studies are commonly overstated because researchers tend to report only the highest correlations, or only those correlations that exceed some threshold. Their paper has in a short time given rise to a spirited debate about key statistical issues at the heart of most functional neuroimaging studies. The debate provides a useful opportunity to discuss core statistical issues in neuroimaging and ultimately provides a chance for the field to grow and move forward. This commentary approaches the debate from a fundamentally statistical perspective. We begin by summarizing several of the key points under discussion, followed by our own commentary on these issues from a statistical point of view. We conclude our discussion by contemplating whether it may be time to move beyond the correlation and multiple comparisons framework, which is causing so much confusion, and instead represent all relevant research questions as parameters in one coherent multilevel model. 2
Comparing methods of analyzing fmri statistical parametric maps
 NeuroImage
, 2004
"... Approaches for the analysis of statistical parametric maps (SPMs) can be crudely grouped into three main categories in which different philosophies are applied to delineate activated regions. These being type I error control thresholding, false discovery rate (FDR) control thresholding and posterior ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
Approaches for the analysis of statistical parametric maps (SPMs) can be crudely grouped into three main categories in which different philosophies are applied to delineate activated regions. These being type I error control thresholding, false discovery rate (FDR) control thresholding and posterior probability thresholding. To better understand the properties of these main approaches, we carried out a simulation study to compare the approaches as they would be used on real data sets. Using default settings, we find that posterior probability thresholding is the most powerful approach, and type I error control thresholding provides the lowest levels of type I error. False discovery rate control thresholding performs in between the other approaches for both these criteria, although for some parameter settings this approach can approximate the performance of posterior probability thresholding. Based on these results, we discuss the relative merits of the three approaches in an attempt to decide upon an optimal approach. We conclude that viewing the problem of delineating areas of activation as a classification problem provides a highly interpretable framework for comparing the methods. Within this framework, we highlight the role of the loss function, which explicitly penalizes the types of errors that may occur in a given analysis.
Trialbytrial data analysis using computational models
, 2009
"... In numerous and highprofile studies, researchers have recently begun to integrate computational models ..."
Abstract

Cited by 6 (2 self)
 Add to MetaCart
In numerous and highprofile studies, researchers have recently begun to integrate computational models
Wavelengthdependent modulation of brain responses to a working memory task by daytime light exposure. Cereb. Cortex 17:2788–2795
 Darsaud A, Sterpenich V, Albouy G, Dijk DJ, Maquet P
, 2007
"... In addition to classical visual effects, light elicits nonvisual brain responses, which profoundly influence physiology and behavior. These effects are mediated in part by melanopsinexpressing lightsensitive ganglion cells that, in contrast to the classical photopic system that is maximally sensiti ..."
Abstract

Cited by 4 (0 self)
 Add to MetaCart
In addition to classical visual effects, light elicits nonvisual brain responses, which profoundly influence physiology and behavior. These effects are mediated in part by melanopsinexpressing lightsensitive ganglion cells that, in contrast to the classical photopic system that is maximally sensitive to green light (550 nm), is very sensitive to blue light (470480 nm). At present, there is no evidence that blue light exposure is effective in modulating nonvisual brain activity related to complex cognitive tasks. Using functional magnetic resonance imaging, we show that, while participants perform an auditory working memory task, a short (18 min) daytime exposure to blue (470 nm) or green (550 nm) monochromatic light (3 3 10 13 photons/cm 2 /s) differentially modulates regional brain responses. Blue light typically enhanced brain responses or at least prevented the decline otherwise observed following green light exposure in frontal and parietal cortices implicated in working memory, and in the thalamus involved in the modulation of cognition by arousal. Our results imply that monochromatic light can affect cognitive functions almost instantaneously and suggest that these effects are mediated by a melanopsinbased photoreceptor system.
A Topographic Latent Source Model for fMRI Data
"... We describe and evaluate a new statistical generative model of functional magnetic resonance imaging (fMRI) data. The model, topographic latent source analysis (TLSA), assumes that fMRI images are generated by a covariatedependent superposition of latent sources. These sources are defined in terms ..."
Abstract

Cited by 4 (1 self)
 Add to MetaCart
We describe and evaluate a new statistical generative model of functional magnetic resonance imaging (fMRI) data. The model, topographic latent source analysis (TLSA), assumes that fMRI images are generated by a covariatedependent superposition of latent sources. These sources are defined in terms of basis functions over space. Importantly, the number of parameters in the model does not depend on the number of voxels, enabling a parsimonious description of activity patterns that avoids many of the pitfalls of traditional voxelbased approaches. Our spatial model leads naturally to a multisubject extension in which latent sources at the subjectlevel are treated as perturbations of a grouplevel template. We evaluate TLSA according to prediction, reconstruction and reproducibility metrics and show that it compares favorably to a Naive Bayes model while using fewer parameters. We also describe a hypothesistesting framework that can be used to identify significant latent sources. Keywords:
A Hybrid Approach to Brain Extraction from
 Premature Infant MRI, Scandinavian Conference on Image Analysis (SCIA 2011), LNCS
, 2011
"... Abstract. This paper describes a novel automatic skullstripping method for premature infant data. A skullstripping approach involves the removal of nonbrain tissue from medical brain images. The new method reduces the image artefacts, generates binary masks and multiple thresholds, and extracts t ..."
Abstract

Cited by 2 (1 self)
 Add to MetaCart
(Show Context)
Abstract. This paper describes a novel automatic skullstripping method for premature infant data. A skullstripping approach involves the removal of nonbrain tissue from medical brain images. The new method reduces the image artefacts, generates binary masks and multiple thresholds, and extracts the region of interest. To define the outer boundary of the brain tissue, a binary mask is generated using morphological operators, followed by region growing and edge detection. For a better accuracy, a threshold for each slice in the volume is calculated using kmeans clustering. The segmentation of the brain tissue is achieved by applying a region growing and finalized with a local edge refinement. This technique has been tested and compared to manually segmented data and to four wellestablished state of the art brain extraction methods.