 |
 |
Hi Gwenaelle and Stam, A belated thank you for your advice regarding this thread. So if my interpretation of your comments is correct, changes in the mode (ie an increase from negative to more positive values) might represent a shift from a "disc" to more "cigar" shaped anisotropy especially if this increase is in areas where one expects significant amounts of crossing fibres. In areas where there is less likely to be crossing fibres, a similar increase in the mode may alternatively represent a shift to a more linear anisotropy, which may for example be due to an increase in the 1st eigen value in the tensor model or a decrease in the 2nd/3rd eigenvalues. Would use of tbss_x aid interpretation further. I understand that it ensures that f1 and f2 belong to the same fibre bundles across subjects, so that it is then possible to compare measures across subjects that are related to the same fibres. If therefore an increase in mode of anistrophy/FA picked up by TBSS is due to selective degeneration of a fibre bundle, would one then expect to see decreases in F2 in that same area ? That is, would one expect to see a decreased proportion of the diffusion signal being modelled by f2 in those regions ? Conversely if one sees an increase in f1 in those regions would that again suggest that the modal changes seen represent a more linear anisotropy, which as Gwenaelle points out you would maybe see in healthy subjects with, say, motor learning for instance, just like an increase of FA would then also be seen. Thank you for your help. Mahinda On Thu, Jun 24, 2010 at 8:07 PM, Gwenaëlle DOUAUD <[log in to unmask]>wrote: > Hi Stefano, > > yes it does (labelled dti_MO.nii.gz). > > Cheers, > Gwenaelle > > --- En date de : Jeu 24.6.10, Marenco, Stefano (NIH/NIMH) [E] < > [log in to unmask]> a écrit : > > > De: Marenco, Stefano (NIH/NIMH) [E] <[log in to unmask]> > > Objet: Re: [FSL] Anisotropy mode > > À: [log in to unmask] > > Date: Jeudi 24 juin 2010, 19h58 > > Does DTIfit output mode? If not, how > > should it be calculated? Stefano Marenco > > > > -----Original Message----- > > From: Gwenaëlle DOUAUD [mailto:[log in to unmask]] > > > > Sent: Thursday, June 24, 2010 2:50 PM > > To: [log in to unmask] > > Subject: Re: [FSL] Anisotropy mode > > > > Hi Mahinda, > > > > to add a few words to Stam's answer: very generally, you > > can see an increase of mode when there is a transition from > > "disc" shape to more "cigar" shape anisotropy. > > > > This can mean a selective degeneration in two crossing > > fibres in a neurodegenerative disorder for instance (as in > > my poster that you're mentioning), this can also mean a more > > linear anisotropy, which you would maybe see in healthy > > subjects with, say, motor learning for instance, just like > > an increase of FA would then also be seen. > > > > Hope this helps, > > Gwenaelle > > > > > > --- En date de : Mer 23.6.10, Stamatios Sotiropoulos > > <[log in to unmask]> > > a écrit : > > > > > De: Stamatios Sotiropoulos <[log in to unmask]> > > > Objet: Re: [FSL] Anisotropy mode > > > À: [log in to unmask] > > > Date: Mercredi 23 juin 2010, 22h03 > > > Hi Mahinda, > > > > > > The mode characterises directly the tensor shape. A > > > negative mode indicates an oblate profile (i.e. l1 > > roughly > > > equal to l2, both much larger than l3) and a positive > > mode a > > > prolate profile (l1 much larger than both l2 and l3), > > where > > > l's are the tensor eigenvalues. Therefore, you can > > interpret > > > directly changes in the mode to changes in the > > relative > > > relationship between the tensor eigenvalues. > > > > > > It has been empirically observed that voxels with two > > > crossing fibres tend to have an oblate tensor shape > > (and > > > this approximation improves as the crossing becomes > > more > > > perpendicular, the fibres more similar and the volume > > > fractions more identical). That's why a negative mode > > value > > > may indirectly suggest a fibre crossing. > > > > > > Cheers, > > > Stam > > > > > > > > > > > > On 23 Jun 2010, at 17:33, Mahinda Y wrote: > > > > > > > Dear All, > > > > > > > > I have seen some recent and papers using the mode > > of > > > anisotropy to distinguish between where changes in FA > > are > > > due to changes in voxels containing 2 fibre bundles to > > 1 > > > fibre bundle (causing an increase in the mode to more > > > positive values) - is this interpretation only > > valid > > > in areas where one knows there are crossing fibres - > > how > > > does one interpret such changes in the mode in areas > > where > > > crossing fibres are less likely ? > > > > > > > > Thanks. > > > > > > > > Mahinda > > > > > > > > > > -------------------------------------------------------------------- > > > > 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 <http://www.fmrib.ox.ac.uk/%7Edouaud> > > > > -------------------------------------------------------------------- > > > > >