Hi- Sorry, but I didn't see your reply yesterday.
I got the point
I really thank you all for the constructive conversation!!
Best
Amelia
Amelia Versace, MD
Loeffler Building
121 Meyran Avenue, room 202
Pittsburgh, PA 15213
Phone 412- 383-8202
Fax 412-383-8336
CONFIDENTIAL UPMC HEALTH SYSTEM INFORMATION Any unauthorized or improper disclosure, copying, distribution, or use of the contents of this email or attached documents is PROHIBITED. The information contained in this e-mail message is intended only for the personal and confidential use of the recipient(s) named above. If you have received this communication in error, please notify the sender IMMEDIATELY by e-mail and DELETE the original message.
-----Original Message-----
From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Steve Smith
Sent: Wednesday, March 19, 2008 2:33 AM
To: [log in to unmask]
Subject: Re: [FSL] SVC for multiple comparisons
Hi - yes, as I said yesterday, as long as you didn't decide to place
this ROI _after_ seeing the initial uncorrected results, then that's
fine - that would be true for this situation AND in FMRI.
Cheers, Steve.
On 18 Mar 2008, at 20:41, Versace, Amelia wrote:
> Not really...
> I just wanted to check that the approach I'm using to control for
> multiple tests for small regions that emerge as significant
> (p<0.001, uncorrected) in my TBSS analyses is OK.
>
> The approach I'm using for a small volume correction is essentially
> the same as the small volume correction approach employed for fMRI
> and VBM structural data I've used with SMP5. I'm using an
> anatomically-defined regional mask for the small volume correction.
> Thus, for example, for the left optic radiation cluster, I use an
> optic radiation mask for which the search volume is approx.. 1500
> voxels. I test the significance of my 17-voxel cluster against that
> mask. Is this approach OK for TBSS data? I cannot think of another
> method of employing a small volume correction approach.
>
> Thank you in advance for your advice here!
> Best Regards
> Amelia
>
>
> Amelia Versace, MD
> Loeffler Building
> 121 Meyran Avenue, room 202
> Pittsburgh, PA 15213
> Phone 412- 383-8202
> Fax 412-383-8336
>
>
> CONFIDENTIAL UPMC HEALTH SYSTEM INFORMATION Any unauthorized or
> improper disclosure, copying, distribution, or use of the contents
> of this email or attached documents is PROHIBITED. The information
> contained in this e-mail message is intended only for the personal
> and confidential use of the recipient(s) named above. If you have
> received this communication in error, please notify the sender
> IMMEDIATELY by e-mail and DELETE the original message.
>
>
> From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On
> Behalf Of Thomas Nichols
> Sent: Monday, March 17, 2008 6:07 PM
> To: [log in to unmask]
> Subject: Re: [FSL] SVC for multiple comparisons
>
> Amelia,
>
> I haven't shed any tears yet, but I'm afraid I don't quite
> understand your question either.
>
> Is it this: You are surprised that using a tightly constrained ROI
> doesn't increase your sensitivity when using FDR (relative to a more
> generous region)? This is not surprising, especially if you have a
> tiny region. The reason is that FDR is an adaptive method and works
> best with there are many 100's or 1000's of voxels, and when most of
> them are null voxels with no signal. With a tiny number of voxels
> (17?), I think your better off just averaging the data and then
> fitting a model with, e.g. SPSS.
>
> Was this the issue?
>
> -Tom
> On Sun, Mar 16, 2008 at 2:52 PM, Versace, Amelia <[log in to unmask]>
> wrote:
> I am afraid I wasn't clear enough.... Let me try to make it clear...
> I didn't try to run fdr in the same volume (number of voxels
> significantly 'uncorrected'), but I defined the anatomical WM tract
> whom my 'uncorrected' roi belong to.
> I.e.: WM-mask in the L_Cuneus (about 1500 voxels) for an roi
> (17vxls) placed in the L_optical radiation (my uncorrected roi).
> I thought in this way I can say that this region doesn't survive for
> multiple comparisons at whole brain level, but it does within the
> anatomical area it belongs to.
> I can't see why this means cheating, but ...."of course I don't want
> make Tom cry...". :-), so please, could you suggest some references
> to understand these concepts better??
>
> In the mean time, I'll try the suggested approach.
>
> Thank you very much for your attention!!!
> Amelia
>
>
>
>
> Hi - I'm afraid you're not allowed to reduce your set of considered
> voxels (in order to reduce the effects of multiple comparisons) using
> the same data before and after - that's cheating and will make Tom
> cry.
>
> You _are_ for example, allowed to find a (corrected) significant ROI
> in the FA, and then only test the MD in that ROI.
>
> But you're NOT allowed to find an (uncorrected) 'significant' ROI in
> FA, and then only test within there for multiple comparisons - this
> goes against the whole reason for needing to do multiple comparisons
> on the original full set of voxels.
>
> If you're needing to boost significance I would just recommend testing
> the -tfce option (probably using H=1 and E=1 and just 500 permutations
> to start with)
>
> Cheers.
>
>
>
> On 11 Mar 2008, at 20:59, Versace, Amelia wrote:
>
> > Dear FSL experts,
> >
> > I am trying to do small volume correction for multiple comparison in
> > DTI data, because I got significant results just in tbss_*_voxtstat*
> > image (uncorrected p value).
> >
> > I was wondering if the following steps are correct:
> >
> > 1. run tbss -i all_FA.nii.gz -o tbss_* -m mean_FA_skeletonized _mask
> > -d design.mat -t design.con -c 3 -n 10000 -v 5
> >
> > 2. define 1-voxtstat* image (fslmaths tbss_*_voxtstat* -mul -1 -add
> > 1 tbss_*_1-voxtstat*)
> >
> > 3. define a WM-mask (accordingly with mean_FA_skeleton_mask)
> > surrounding a significant roi (group of voxels with p>0.999).
> >
> > 2. run fdr -i tbss_*_1-voxtstat* -m WM_mask -q 0.05
> >
> > 4.if the output is >0, can I consider that roi as small volume
> > corrected for multiple comparison??
> >
> >
> > About point 3, is there any size limitation of WM-mask in order to
> > use FDR properly??
> >
> >
> > If this is not the proper way, can anybody suggest a better one?
> > Many thanks for your help!
> > Amelia
> >
>
>
> ---------------------------------------------------------------------------
> 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
> ---------------------------------------------------------------------------
>
>
>
> --
> ____________________________________________
> Thomas Nichols, PhD
> Director, Modelling & Genetics
> GlaxoSmithKline Clinical Imaging Centre
>
> Senior Research Fellow
> Oxford University FMRIB Centre
---------------------------------------------------------------------------
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
---------------------------------------------------------------------------
|
