Limits...
Detecting subject-specific activations using fuzzy clustering.

Seghier ML, Friston KJ, Price CJ - Neuroimage (2007)

Bottom Line: The output identifies subjects who drive activation in specific cortical regions (local effects) or in voxels distributed across neural systems (global effects).In these cases, atypical activations may not be detected by standard tests, under parametric assumptions.The advantage of using FCP is that it searches all voxels systematically and can identify atypical activation patterns in a quantitative and unsupervised manner.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Centre for Neuroimaging, Institute of Neurology, 12 Queen Square, London WC1N 3BG, UK. m.seghier@fil.ion.ucl.ac.uk

ABSTRACT
Inter-subject variability in evoked brain responses is attracting attention because it may reflect important variability in structure-function relationships over subjects. This variability could be a signature of degenerate (many-to-one) structure-function mappings in normal subjects or reflect changes that are disclosed by brain damage. In this paper, we describe a non-iterative fuzzy clustering algorithm (FCP: fuzzy clustering with fixed prototypes) for characterizing inter-subject variability in between-subject or second-level analyses of fMRI data. The approach identifies the contribution of each subject to response profiles in voxels surviving a classical F-statistic criterion. The output identifies subjects who drive activation in specific cortical regions (local effects) or in voxels distributed across neural systems (global effects). The sensitivity of the approach was assessed in 38 normal subjects performing an overt naming task. FCP revealed that several subjects had either abnormally high or abnormally low responses. FCP may be particularly useful for characterizing outlier responses in rare patients or heterogeneous populations. In these cases, atypical activations may not be detected by standard tests, under parametric assumptions. The advantage of using FCP is that it searches all voxels systematically and can identify atypical activation patterns in a quantitative and unsupervised manner.

Show MeSH

Related in: MedlinePlus

A—(left) group activation for the main effect of object naming relative to the non-object baseline; (right) group activation for the effect of non-object relative to object naming. B—(left) percent of overlap maps that measure how many subjects are activating each voxel at p < 0.01 (uncorrected). Voxels observed in one to up twenty subjects are projected as a color-coded map. (Right) Histogram of activated voxels over a given number of subjects. Voxels observed in few subjects (e.g., one or two subjects) represent the majority.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2724061&req=5

fig7: A—(left) group activation for the main effect of object naming relative to the non-object baseline; (right) group activation for the effect of non-object relative to object naming. B—(left) percent of overlap maps that measure how many subjects are activating each voxel at p < 0.01 (uncorrected). Voxels observed in one to up twenty subjects are projected as a color-coded map. (Right) Histogram of activated voxels over a given number of subjects. Voxels observed in few subjects (e.g., one or two subjects) represent the majority.

Mentions: Group activation for the main effect of object naming, relative to the non-object baseline, is shown in Fig. 7A. Positive activations were observed in bilateral fusiform, inferior occipital gyri, cerebellum and SMA, with left lateralized effects in pre-central, inferior frontal, and middle temporal gyri. Negative activations were observed in bilateral inferior and superior parietal regions, precuneus, posterior cingulate and superior frontal gyri, with right lateralized effects in inferior temporal and pre-central gyri. These regions have been observed in previous studies with object naming tasks; see Price et al., 2005 for review. We also illustrate the percentage overlap between thresholded individual maps (Fig. 7B). Basically, these maps represent how often each voxel has been observed as “activated” across subjects at a given individual threshold (p < 0.01, uncorrected). The common voxels between subjects are not surprisingly less frequent than voxels that have been observed in only one or two subjects, suggesting that, across subjects, activated regions are variable in size, localization and statistical significance.


Detecting subject-specific activations using fuzzy clustering.

Seghier ML, Friston KJ, Price CJ - Neuroimage (2007)

A—(left) group activation for the main effect of object naming relative to the non-object baseline; (right) group activation for the effect of non-object relative to object naming. B—(left) percent of overlap maps that measure how many subjects are activating each voxel at p < 0.01 (uncorrected). Voxels observed in one to up twenty subjects are projected as a color-coded map. (Right) Histogram of activated voxels over a given number of subjects. Voxels observed in few subjects (e.g., one or two subjects) represent the majority.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2724061&req=5

fig7: A—(left) group activation for the main effect of object naming relative to the non-object baseline; (right) group activation for the effect of non-object relative to object naming. B—(left) percent of overlap maps that measure how many subjects are activating each voxel at p < 0.01 (uncorrected). Voxels observed in one to up twenty subjects are projected as a color-coded map. (Right) Histogram of activated voxels over a given number of subjects. Voxels observed in few subjects (e.g., one or two subjects) represent the majority.
Mentions: Group activation for the main effect of object naming, relative to the non-object baseline, is shown in Fig. 7A. Positive activations were observed in bilateral fusiform, inferior occipital gyri, cerebellum and SMA, with left lateralized effects in pre-central, inferior frontal, and middle temporal gyri. Negative activations were observed in bilateral inferior and superior parietal regions, precuneus, posterior cingulate and superior frontal gyri, with right lateralized effects in inferior temporal and pre-central gyri. These regions have been observed in previous studies with object naming tasks; see Price et al., 2005 for review. We also illustrate the percentage overlap between thresholded individual maps (Fig. 7B). Basically, these maps represent how often each voxel has been observed as “activated” across subjects at a given individual threshold (p < 0.01, uncorrected). The common voxels between subjects are not surprisingly less frequent than voxels that have been observed in only one or two subjects, suggesting that, across subjects, activated regions are variable in size, localization and statistical significance.

Bottom Line: The output identifies subjects who drive activation in specific cortical regions (local effects) or in voxels distributed across neural systems (global effects).In these cases, atypical activations may not be detected by standard tests, under parametric assumptions.The advantage of using FCP is that it searches all voxels systematically and can identify atypical activation patterns in a quantitative and unsupervised manner.

View Article: PubMed Central - PubMed

Affiliation: Wellcome Trust Centre for Neuroimaging, Institute of Neurology, 12 Queen Square, London WC1N 3BG, UK. m.seghier@fil.ion.ucl.ac.uk

ABSTRACT
Inter-subject variability in evoked brain responses is attracting attention because it may reflect important variability in structure-function relationships over subjects. This variability could be a signature of degenerate (many-to-one) structure-function mappings in normal subjects or reflect changes that are disclosed by brain damage. In this paper, we describe a non-iterative fuzzy clustering algorithm (FCP: fuzzy clustering with fixed prototypes) for characterizing inter-subject variability in between-subject or second-level analyses of fMRI data. The approach identifies the contribution of each subject to response profiles in voxels surviving a classical F-statistic criterion. The output identifies subjects who drive activation in specific cortical regions (local effects) or in voxels distributed across neural systems (global effects). The sensitivity of the approach was assessed in 38 normal subjects performing an overt naming task. FCP revealed that several subjects had either abnormally high or abnormally low responses. FCP may be particularly useful for characterizing outlier responses in rare patients or heterogeneous populations. In these cases, atypical activations may not be detected by standard tests, under parametric assumptions. The advantage of using FCP is that it searches all voxels systematically and can identify atypical activation patterns in a quantitative and unsupervised manner.

Show MeSH
Related in: MedlinePlus