 |
 |
Thank you! I will check this out. Best, Aga On 26.01.2008 8:35 Uhr, "Steve Smith" <[log in to unmask]> wrote: > Hi, > > Ah, I think I know what's happening now. It sounds like you are > getting wrapping from very large numbers round to being negative as > the scaling being applied in the script is too large. Please see the > TBSS manual where it is described how to set the scaling factor as an > argument to tbss_1_preproc so that the preprocessed data ends up with > range mostly 0:10000. If you have done this correctly for most voxels > but still have a few high intensity outliers then you could just hard > limit those before running tbss_1_preproc with the -min option in > fslmaths. > > Cheers. > > > On 25 Jan 2008, at 18:30, Agnieszka Burzynska wrote: > >> Hi, >> Thank you! I applied one more time fslstats to my images. Indeed, the >> location of the negative voxels may not be too problematic, but the >> negative >> values are of the same range as positive (eg. -31688 and 30531, I >> did not >> divide by 3 to get it closer to 10000) and on the origdata histogram >> I see >> really nothing below zero, too. >> I just wonder if I did any basic mistake or something else could have >> happened. >> >> Best wishes, >> Aga >> >> >> On 25.01.2008 17:47 Uhr, "Steve Smith" <[log in to unmask]> wrote: >> >>> Hi, >>> >>> I don't think that tbss_1_preproc (at least the version with FSL4) >>> can >>> create new negative values, but it does multiply the original values >>> to make them a lot larger, so maybe you started with VERY slightly >>> negative values that are sufficiently small that you didn't notice >>> them, and after the rescaling, they were more obvious? It's probably >>> not a problem anyway if they are around the edge of the brain? >>> >>> Cheers, Steve. >>> >>> >>> >>> On 25 Jan 2008, at 10:36, Agnieszka Burzynska wrote: >>> >>>> Dear all, I resend this message, because I still do not know what is >>>> wrong: >>>> >>>> >>>> Hi, >>>> I applied tbss_1_preproc to my S0 images. In origdata, none of them >>>> had >>>> values below 0, however, after the scaling in the FAi folder there >>>> are 2 >>>> cases, in which I see negative values around the brainstem. >>>> >>>> How did this happen and what should I do? I thought it is just a >>>> simple >>>> mulitiplication by a given number (here by 10). >>>> >>>> Thank you and best wishes, >>>> Aga >>>> >>> >>> >>> --------------------------------------------------------------------------- >>> 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 > ---------------------------------------------------------------------------