“A full spatiotemporal model of the data is generally not considered feasible and a number of short cuts are taken throughout the course of the analysis.” (p. 440)

“While all the preprocessing steps outlined above are essential for the standard model assumptions required for statistical analysis to hold, there needs to be a clear understanding of the effects they have on both the spatial and temporal correlation structure. More generally, it is necessary to study the interactions among the individual preprocessing steps.” (p.451)

“There has been explosive interest in the use of fMRI in recent years. The rapid pace of development and the interdisciplinary nature of the neurosciences present an enormous challenge to researchers. Moving the field forward requires a collaborative team with expertise in psychology, neuroanatomy, neurophysiology, physics, biomedical engineering, statistics, signal processing and a variety of other disciplines depending on the research question. True interdisciplinary collaboration is exceedingly challenging, because team members must know enough about the other disciplines to be able to talk intelligently with experts in each field. Due to the importance that statistics plays in this research, it is important that more statisticians get involved in these research teams for the methodology to reach its full potential. ” (p. 461)

PS: Information about the talk I refer to:

PradoRaquel Prado (Department of Statistics at the University of California Santa Cruz)

Title: Bayesian models for complex-valued fMRI

Abstract: Detecting which voxels/regions are actived by an external stimulus is one of the main goals in the analysis of functional magnetic resonance imaging (fMRI) data. Voxel time series in fMRI are complex-valued signals consisting of magnitude and phase components, however, most studies discard the phase and only use the magnitude data. We present a Bayesian variable selection approach for detecting activation at the voxel level from complex-valued fMRI (f(c)MRI) recorded during task experiments. We show that this approach leads to fast and improved detection of activation when compared to alternative magnitude-only approaches. We discuss and illustrate modeling extensions that incorporate additional spatial structure via kernel convolution for more flexible analysis of f(c)MRI. The complex-valued spatial model encourages voxels to be activated in clusters, which is appropriate in applied settings, as the execution of complex cognitive tasks, and therefore brain activation, usually involve populations of neurons spanning across many voxels rather than isolated voxels. Finally, we present models that can handle multi-subject data and allow us to infer connectivity at the region-specific level in addition to voxel-specific activation. Model performance is evaluated through extensive and physically realistic simulation studies and in the analysis of human f(c)MRI.

]]>BTW, I have a link to a paper by Lindquist (2008) regarding the statistical analysis of fMRI data.

https://projecteuclid.org/download/pdfview_1/euclid.ss/1242049389

]]>