Limits...
Spatial-temporal modelling of fMRI data through spatially regularized mixture of hidden process models.

Shen Y, Mayhew SD, Kourtzi Z, Tiňo P - Neuroimage (2013)

Bottom Line: Thus, the total number of parameters in the model does not depend on the number of voxels.The results strongly suggest that, compared with occipitotemporal regions, the frontal ones are less homogeneous, requiring two HPM prototypes per region.Spatio-temporal heterogeneity in the frontal regions seems to be associated with diverse dynamic localizations of the two hidden processes in different subregions of frontal ROIs.

View Article: PubMed Central - PubMed

Affiliation: School of Computer Science, The University of Birmingham, Birmingham, UK. Electronic address: y.shen.2@cs.bham.ac.uk.

Show MeSH
Log model evidence as function of the number of prototypes for 4 example data sets from different ROIs.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC4066951&req=5

f0045: Log model evidence as function of the number of prototypes for 4 example data sets from different ROIs.

Mentions: As discussed in the previous section, determining the number of necessary prototypes from fMRI data is computationally very expensive when model evidence is used. Therefore, approximation of model evidence, such as BIC or free energy is often used. In this work, the model evidence approach is tested with four example fMRI data sets. Each data set is derived from one of four ROIs that are considered in this work. Fig. 9 shows that two prototypes are clearly needed for SFG and MFG while for V1 and LO, a single prototype is probably sufficient.


Spatial-temporal modelling of fMRI data through spatially regularized mixture of hidden process models.

Shen Y, Mayhew SD, Kourtzi Z, Tiňo P - Neuroimage (2013)

Log model evidence as function of the number of prototypes for 4 example data sets from different ROIs.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC4066951&req=5

f0045: Log model evidence as function of the number of prototypes for 4 example data sets from different ROIs.
Mentions: As discussed in the previous section, determining the number of necessary prototypes from fMRI data is computationally very expensive when model evidence is used. Therefore, approximation of model evidence, such as BIC or free energy is often used. In this work, the model evidence approach is tested with four example fMRI data sets. Each data set is derived from one of four ROIs that are considered in this work. Fig. 9 shows that two prototypes are clearly needed for SFG and MFG while for V1 and LO, a single prototype is probably sufficient.

Bottom Line: Thus, the total number of parameters in the model does not depend on the number of voxels.The results strongly suggest that, compared with occipitotemporal regions, the frontal ones are less homogeneous, requiring two HPM prototypes per region.Spatio-temporal heterogeneity in the frontal regions seems to be associated with diverse dynamic localizations of the two hidden processes in different subregions of frontal ROIs.

View Article: PubMed Central - PubMed

Affiliation: School of Computer Science, The University of Birmingham, Birmingham, UK. Electronic address: y.shen.2@cs.bham.ac.uk.

Show MeSH