 |
 |
Hi Gwenaelle, thank you very much for your answer. My pathological sample is formed by malnourished subjects with anorexia nervosa. The information in my previous message was partially wrong, because I have found only a trend towards a decrease in FA, a significant increase in MD (both L1 and L23) especially in thalamic radiation, fornix, superior longitudinal fasciculum (more the left). Now I have run randomise with MO using TBSS, and in the same tracts I have found a significant decrease in MO (so from linear to planar). Why did not FA show a significant decrease? is it because in regions of crossing fibers there can be a difference in degeneration across different types of fibers (as you hypothesized in your paper on AD/MCI)? how can I test this or other hypotheses? Thank you for your help there are so few examples of the use of MO in the literature! Angela > Hi Angela, > > you can use the mode of anisotropy as you would use any other DTI indices. > It proved to be more sensitive than FA in regions of crossing fibres in > MCI/AD, but it really depends on your own question/population. > > You simply need to run tbss_non_FA on the dtifit output images *_MO. If > they have not been created by dtifit, then you need to update your version > of FSL. > > Cheers, > Gwenaelle > >> De: Angela Favaro <[log in to unmask]> >> Objet: [FSL] a question about TBSS mode of anisotropy (MO) >> À: [log in to unmask] >> Date: Vendredi 15 juillet 2011, 9h23 >> Dear FSL experts, >> I am performing a study with TBSS. >> My pathological group did not show significant differences >> analysing FA >> (although there are trends between 0.06 and 0.08 TFCE >> corrected that show >> decreased FA) or MD, but I have found 4 clusters (more than >> 100 voxel) of >> increased L1. In order to better understand this result I >> have decided to >> analyse the mode of anisotropy (MO file). >> >> My questions are: >> may I use TBSS to analyse MO? >> I have read a paper of Gwenaëlle about MCI where a >> voxel-based approach >> has been used for MO files. Is this always more correct or >> it has been >> used because of the particular population? IN the case of >> voxel-based >> analysis, do I need to smooth the images and how much? do I >> need to use a >> mask excluding gray matter or do I perform a whole-brain >> analysis? >> any help or suggestion is welcome! >> >> Angela >> >> Angela Favaro, MD, PhD >> Psychiatric Clinic >> Department of Neurosciences >> via Giustiniani 3 >> 35128 Padova >> Italy > -------------------------------------------------------------------- > > Gwenaëlle Douaud, PhD > > FMRIB Centre, University of Oxford > John Radcliffe Hospital, Headington OX3 9DU Oxford UK > > Tel: +44 (0) 1865 222 523 Fax: +44 (0) 1865 222 717 > > www.fmrib.ox.ac.uk/~douaud > > -------------------------------------------------------------------- > >