Hi Bruce, Thank you very much for your clarifications. I will ask ADNI for their permission and I will consult CMA for their definitions.
Thank you, Have a great day, John Sent from my iPhone 4S On 2012-07-18, at 6:42 AM, Bruce Fischl <fis...@nmr.mgh.harvard.edu> wrote: > Hi John > > we follow the conventions of the Center for Morphometric Analysis (CMA) here > at MGH. I'm not actually sure what landmarks they use to separate what they > call "inferior lateral ventricle" from the main one, but you can check with > them. > > As for ADNI, we might have that data but I'm not sure we are allowed to > distribute it. If you check with them about it and let us know we'll give you > what we have. > > cheers > Bruce > > > On Tue, 17 Jul 2012, John Drozd wrote: > >> Hi Bruce, >> >> I apologize for my detailist nature. >> I found the csv files on the ADNI web site containing the numerical values >> of volume in mm^3 numbers. Do you mean that there should also be the >> corresponding label maps on the ADNI web site, particularly of the third and >> lateral ventricles? I know when I ran recon-all it gave all the label maps >> of each parcellation of the brain and I could overlay these labels on the >> grey scale images and render 3D surface models in Slicer. If this is the >> case, could you confirm this. Then I'll email ADNI to see if they have the >> label maps corresponding to the csv files that they have in the ADNI Study >> Data -> MRI Images -> Image Analysis link. >> >> I am particularly interested in where Freesurfer separates the lateral >> ventricles from the third ventricle at the foramina, or at the connection >> between the posterior horns and the third ventricle. If you have any >> insight in this, that would be helpful. I am glad that Freesurfer includes >> the isolated horn pieces, as you explained. I particularly found that >> accurate segmentation of the horns was very important for the significance >> statistics in comparing AD, MCI and NEC groups. The next step for me will >> be sub-regional ventricle volumes, not only left and right but separate >> volumes of horns. I am curious how you defined the division of left from >> right for the ventricles? Was it special landmarks on the atlas? >> >> Thank you, >> John >> >> Sent from my iPhone 4S >> >> >> On 2012-07-17, at 2:13 PM, Bruce Fischl <fis...@nmr.mgh.harvard.edu> wrote: >> >>> Hi John >>> >>> have you checked with the ADNI folks? I think they have FreeSurfer >>> segmentation on many if not most of their subjects. We place no >>> connectivity constraints on the ventricular segmentations because they get >>> so thin that indeed sometimes 1mm data cannot resolve the connections. So >>> sometimes the temporal or occipital horn are disconnected, and sometimes >>> not. >>> >>> cheers >>> Bruce >>> >>> >>> On Tue, 17 Jul 2012, John Drozd wrote: >>> >>>> Hello, >>>> >>>> I have successfully developed an automated algorithm combining pixel >>>> intensity-based and shape based approaches to segment the lateral >>>> ventricles. I have applied my algorithm to a small data set of 1.5 T1 ADNI >>>> images consisting of 24 AD (subjects with Alzheimer Disease) 19 MCI >>>> (subjects with Mild Cognitive Impairment) and 24 NEC (elderly controls). >>>> In my cases, particularly AD cases, I noticed that the segmentation of the >>>> lateral ventricles is connected to the third ventricle with foramina >>>> (usually 1 on a side or on both sides). Also, sometimes I find a >>>> connection with the posterior horn and third ventricle. I have researched >>>> CSF flow and from the literature, and all of these are real connections. I >>>> am near completion of developing an automated cropping scheme with a >>>> colleague to crop the third ventricle at these junctions. >>>> >>>> Our implementation is a combination of C++ code that we have developed and >>>> modified ITK classes. We have been viewing our segmented label maps in >>>> Slicer3 and generated 3D renderings using Slicer3's Model Maker module >>>> which is based on Marching Cubes. >>>> >>>> I have viewed a Freesurfer segmentation in Slicer3 based on a tutorial on >>>> the Slicer web site. >>>> >>>> I presented a poster of our results at the AIC and AAIC conference which I >>>> am at now in Vancouver, and a few people have asked me to compare my >>>> volumes with other methods. They were particularly interested in comparing >>>> my volumes with Freesurfer ventricle volumes and cross-sectional and >>>> longitudinal data. I was told and found csv files on the ADNI web site of >>>> Freesurfer analysis. I plan to correlate my values with Freesurfer's >>>> values, but I would be interested if Freesurfer also finds these >>>> connections between the third ventricle and lateral ventricles and also if >>>> it includes isolated temporal horn pieces when the horns are discontinuous >>>> due to the MRI not being able to resolve a very thin portion of the >>>> temporal horn. >>>> >>>> So, would it be possible, if I gave you a list of my ADNI subjects, for >>>> you to provide me the freesurfer label maps that correspond to the volumes >>>> in the csv files that are on the ADNI web site please? I could run >>>> Freesurfer but I would prefer having the label maps that exactly >>>> correspond to the Freesurfer ventricle volumes that are on the ADNI >>>> website. These would be from the two versions of Freesurfer noted on the >>>> ADNI web site in the ADNI Study Data/MRI image analysis link. >>>> >>>> Thank you, >>>> John >>>> >>>> Sent from my iPhone 4S >>>> >>>> >>>> _______________________________________________ >>>> 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-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
