[FieldTrip] Plotting cluster permutation results

Tavabi, Kambiz TAVABIK at email.chop.edu
Mon Dec 13 23:02:30 CET 2010


Based on FT tutorials and user advice I have managed to create a stat structure using ft_freqstatistics.

stat =

prob [2x158x1800 double]

posclusters: [1x43 struct]

posclusterslabelmat: [2x158x1800 double]

posdistribution: [1x500 double]

negclusters: [1x63 struct]

negclusterslabelmat: [2x158x1800 double]

negdistribution: [1x500 double]

mask: [2x158x1800 logical]

stat: [2x158x1800 double]

ref: [2x158x1800 double]

dimord: 'chan_freq_time'

label: {2x1 cell} (i.e., {['MLT'];['MRT']}

freq: [1x158 double]

time: [1x1800 double]

cfg: [1x1 struct]

Please note that the cluster permuation tests were performed on source space time-frequency representations, thus data is only available for two channels. As such the statistics was done without cfg.minnbchan and cfg.neighbours=[ ]. and clustering was done on the time-frequency bins throuought the entire epoch for wide-band data (i.e., no averaging over frequencies). The initial conditional power spectra used in the statistics were 10 x 2 x 158 x 2880 (subjects x channels x frequencies x time). I did manage to use ft_singleplotTFR(cfg, stat) to plot the results for single channel -after masking time-frequency points by clusterlabels information belonging to a specific cluster (see attached). But I cannot really make much sense out of the results. Am I doing something wrong here?

When I use

cfg = [];

cfg.alpha  = 0.025;

cfg.zparam = 'stat';

cfg.zlim   = [-4 4];

ft_clusterplot(cfg, stat);

I get the following:

??? Error using ==> ft_clusterplot at 75

stat contains multiple frequencies which is not allowed because it should be

averaged over frequencies

Technically I don't have any apriori frequency band(s) of interest, thus I used the wide-band data. How does ft_clusterplot accept frequency averaging? And is it possible to work around this criterion? Is it possible to customize cfg.layout?

Thanks in advance -Kambiz



Kambiz Tavabi PhD

Biomedical Imaging Laboratory

The Children's Hospital of Philadelphia

34th Street and Civic Center Boulevard

Philadelphia, Pa. 19104

Tel: 267.426.0302

email: tavabik at email.chop.edu<mailto:tavabik at email.chop.edu>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20101213/1c31faa1/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ft_singleplotTFR.jpg
Type: image/jpeg
Size: 12468 bytes
Desc: ft_singleplotTFR.jpg
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20101213/1c31faa1/attachment-0001.jpg>

More information about the fieldtrip mailing list