 |
 |
Hi, If you are viewing cluster p-values (maxc) then the range should be (for example) 0.949 - 1 If you are viewing unthresholded tstats then it's more up to you, but a reasonable start is 3-6 or if you want to see more 2-6. Cheers. On 15 May 2007, at 09:38, Ruiwang Huang wrote: > Dear Steve, > > Many thanks for your explanation. > > In reality, I would like to know that for the given tstat images or > the pvalue > images (maxc), how to choose the range of [-b] value if I use > fslview to > display p<0.05 result? (see the example in TBSS-website > (http://www.fmrib.ox.ac.uk/fsl/tbss/index.html), > > Cheers > > Rhuang > > > On Tuesday 15 May 2007 09:39, you wrote: >> Hi - the first example shows how to bring up the cluster p-values and >> threshold them; in the second example, the example fslview command is >> bringing up the raw tstat images, and the text after that example >> command line call is describing how also to load in the pvalue images >> (maxc) and threshold them. >> >> Cheers. >> >> On 14 May 2007, at 17:55, R Huang wrote: >>> Dear All, >>> >>> >>> I am puzzled by the descriptions of TBSS--fslview >>> in the practical and the TBSS-website. >>> >>> >>> (1) In the practical of 'Analysis of Diffusion-Weighted MR Data >>> (the last paragraph of >>> http://www.fmrib.ox.ac.uk/fslcourse/lectures/practicals/fdt/ >>> index.htm), >>> >>> ... Thresholding clusters at 0.949 (corresponding to thresholding >>> the >>> p-values at 0.05, because randomise outputs p-values as 1-p for >>> convenience of display - so that higher values are more >>> significant) shows >>> several areas of reduced FA in the MS patients. >>> The following shows unthresholded t-stats in red-yellow and >>> thresholded >>> clusters in blue: >>> >>> fslview MNI152 mean_FA_skeleton -l Green -b 2000,8000 tbss_tstat1 -l >>> Red-Yellow -b 1.5,3 tbss_maxc_tstat1 -l Blue-Lightblue -b 0.949,1 >>> >>> (2) But in the TBSS >>> (http://www.fmrib.ox.ac.uk/fsl/tbss/index.html), the description >>> >>> fslview MNI152 mean_FA_skeleton -l Green -b 2000,8000 tbss_tstat1 -l >>> Red-Yellow -b 3,6 tbss_tstat2 -l Blue-Lightblue -b 3,6 >>> >>> ...The cluster p-value images (fully corrected for multiple >>> comparisons >>> across space) are tbss_maxc_tstat1 and tbss_maxc_tstat2 (note, >>> these are >>> actually 1-p for convenience of display, so thresholding at .95 >>> gives >>> significant clusters). >>> >>> >>> Does any one know why 'fslview' in the two locations use different >>> '-b' >>> values? >>> >>> Many thanks in advance >>> >>> Rhuang >> >> --------------------------------------------------------------------- >> --- >> --- >> Stephen M. Smith, Professor of Biomedical Engineering >> Associate Director, Oxford University FMRIB Centre >> >> FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK >> +44 (0) 1865 222726 (fax 222717) >> [log in to unmask] http://www.fmrib.ox.ac.uk/~steve >> --------------------------------------------------------------------- >> --- >> --- ------------------------------------------------------------------------ --- Stephen M. Smith, Professor of Biomedical Engineering Associate Director, Oxford University FMRIB Centre FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK +44 (0) 1865 222726 (fax 222717) [log in to unmask] http://www.fmrib.ox.ac.uk/~steve ------------------------------------------------------------------------ ---