Hi Anthony
we are trying to get that piece of it published :) Hopefully soon!
Bruce On Wed, 15 Jun 2011, Anthony Dick wrote:
Good points Bruce. I guess I am more worried about the people using it, who will not be as cautious for large datasets. Also--one thing that was a huge benefit for the cortical parcellation was a quantification of the manual vs. automated parcellation differences, but this does not appear to be published yet (unless I missed it). Any added pathways should have to meet the same standards.
Anthony
On 6/15/11 1:07 PM, Bruce Fischl wrote:
Hi Anthony
we included the tracts that we found that we could segment reliably, under the guidance of a trained neuroanatomist. Certainly we can only do as well as the information contained in the diffusion data, but we have found that we can automated what someone with neuroanatomical expertise could obtain by following the Wakana protocol. I don't think we make any claims beyond that, but in and of itself it seemed like a worthy goal, opening up the field of tractography to a bigger potential user pool. Certainly that comes with potential problems, as it does in fMRI. I don't see it being significantly different than people guessing Brodmann areal identity based on visual inspection of folding patterns and presumed homology with the macaque. It's a useful tool, but like all tools you can overintepret the results if you are not cautious.
cheers Bruce
On Wed, 15 Jun 2011, Anthony Dick wrote:
Hello Anastasia,
This just reinforces my point. Tracts #7 and #8 of that paper are the inferior longitudinal fasciculus and the inferior occipitofrontal fasciculus, respectively. But you have chosen to leave one out (incidentally, both appear in the Oishi atlas). Note that in this paper, the authors recommend manual intervention for certain tracts.
When a tool like this is developed, it gets referenced in papers which gives it a kind of legitimacy. What we want to avoid here is giving legitimacy to fiber pathways that may not actually exist, or missing ones that do because the tool left them out. So what is the rule for how pathways are added or removed from this tool? It seems that simply requesting that they be added can potentially result in such an addition, or that deciding not to include a tract for whatever reason can result in its disappearance. This gives any pathway identified by this tool a legitimacy it might not otherwise have.
I am not necessarily against the tool (I think it is better for exploratory analysis), but I am against the idea, from the Wiki, that Tracula "obviates the need for manual intervention for tract solutions and thus facilitates the application of tractography to large datasets". This seems to me to be a horrible idea. Tractography is not at this point a completely data driven procedure.
Anthony
On 6/15/11 12:44 PM, Anastasia Yendiki wrote:
Hi Anthony - The labeling was done based on Wakana et al 2007. We did not include the IFOF exactly b/c of the controversy surrounding it.
a.y
On Wed, 15 Jun 2011, Anthony Dick wrote:
Hello all,
Is there any atlas to which these tracs are referenced? For cortical anatomical boundaries there is broad consensus and a published referenced atlas, which makes Freesurfer's automatic parcellation of cortical regions an excellent tool. But there is not (as is often assumed) as broad a consensus on white matter tracts (e.g., there is no middle longitudinal fasciculus in the recent Oishi et al. white matter atlas and many would argue it doesn't exist; the existence of separate inferior longitudinal fasciculus and inferior occipitofrontal fasciculus is disputed; Catani's arcuate fasciculus is based on one paper etc.). Is it such a good idea to go automatically identifying white matter tracts without saying how they are identified?
Anthony
On 6/15/11 12:07 PM, Seán Froudist Walsh wrote: Hi Anastasia,
Thanks for getting back to me and well done on making a very nice program. The first tracts I would like to have are the optic radiations, but there are others that would be nice to include: a three part arcuate fasiculus (a la Catani et al., 2005) rather than a 2 part one, Fornix, anterior commissure. If I were to manually do the dissections for the 33 subjects, what would be the most Tracula compatible way of doing it? I normally use TrackVis but I figure making all of the file conversions might be a nightmare. Which program was used in the original virtual dissections? Any further suggestions? Perhaps some of these tracts are included but I've missed them. Sean On 15 June 2011 08:37, Anastasia Yendikiayendiki@nmr.mgh.harvard.edu wrote:
Hi Seán - Which tracts do you want to add? Our existing tracts have been labeled on 33 healthy subjects. It's certainly possible to include new tracts in the atlas but someone has to do the labeling on the same or a similar set of subjects. a.y On Wed, 15 Jun 2011, Seán Froudist Walsh wrote: Hi FreeSurfing Gurus, I would like to add some new tracts to the Tracula pipeline using my own regions of interest. What would be the best way to do this? All the best and many thanks in advance, SeánThe information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mai l contains patient information, please contact the Partners Compliance HelpLin e at http://www.partners.org/complianceline . If the e-mail was sent to you in er ror but does not contain patient information, please contact the sender and prop erly dispose of the e-mail.