Freesurfer July 2012

freesurfer@nmr.mgh.harvard.edu

128 participants
194 discussions

questions about SurfArea and meancurv
by Yang, Daniel 24 Jul '12

24 Jul '12

Dear Freesurfer, I am new to freesurfer and have a number of beginner's questions. After recon all the subjects and running the freesurferstats2table<http://surfer.nmr.mgh.harvard.edu/fswiki/freesurferstats2table> script, I am looking at the two files: aparc_area_lh.txt and aparc_meancurv_lh.txt. Everything goes by default. My questions are, 1. In the aparc_area_lh.txt, is the surface area based on the pial surface, the white surface, or both? 2. In the aparc_meancurv_lh.txt, is it based on the Gaussian Curvature of the white surface or the pial surface? I cannot find the relevant documentation in the freesurfer wiki. Any comments are highly appreciated! Best Regards, Daniel -- Daniel Yang, PhD Postdoctoral Researcher Child Study Center Yale University 203-737-5454

2 4

Hippocampal subfields and aseg
by Murray, Michael 24 Jul '12

24 Jul '12

Hello, In several of our subjects, we've noticed on the aseg that the hippocampus extends into the fusiform gyrus. I've attached an image showing this in the left hippocampus. Looking at the wm.mgz volume, there seem to be wm voxels in that region that shouldn't be there (image attached). Could these be what are causing the incorrect segmentation (in which case the voxels can just be removed from wm.mgz)? If not, do you have any other ideas as to how to fix this problem? Otherwise, it seems as though the aseg would have to be edited, which would be a large and tedious edit to make in our very large set of subjects. This is particularly a problem because we are interested in the subfields, and the issue carries over when that processing is performed. In the same subject as discussed above, the CA1 subfield of the left hippocampus also overextends into the fusiform gyrus (image attached). Since we are primarily interested in using the total subfield volume (rather than the aseg volume), it would also work if we could just specifically edit whatever the input to the subfield stream is. Would this just be aseg, or are there other volumes that would need to be edited before running the subfields flag? That may be a question for Koen, but any advice that anyone could offer would be very helpful. Thank you, Michael Murray

3 2

Interaction between two genes on cortical thickness
by Vincenzo Pierro 24 Jul '12

24 Jul '12

Please, could anyone let me know how to perform statistical analysis in QDEC to look at the interaction between two genes on cortical thickness? Vincenzo Pierro

1 0

Sample size estimates for well-powered cross-sectional cortical thickness studies
by Heath Pardoe 24 Jul '12

24 Jul '12

Hi all, Some work my colleagues and I did investigating how many subjects should be included in a study has been recently published in Human Brain Mapping ( http://www.ncbi.nlm.nih.gov/pubmed/22807270) The paper should be freely available on pubmed central fairly soon if you don't have access to Human Brain Mapping early view. Or I can send you a copy. I've provided some instructions on how to carry out a similar analysis using your own Freesurfer-processed control data here: http://www.brain.org.au/software/cortex/power The R package provided at the above link contains a function "load.mgh.R" which can be used to load mgh surface files into R & do whatever sophisticated statistical analysis takes your fancy. Let me know if you've got any questions. Best wishes Heath -- Heath R. Pardoe, PhD Florey Neuroscience Institutes Melbourne Brain Centre 245 Burgundy Street, Heidelberg 3084 Australia Phone: (+61 3) 903 57053 Email: h.pardoe(a)brain.org.au

1 0

Re: [Freesurfer] stats
by Douglas Greve 23 Jul '12

23 Jul '12

Neither of these tables is the one you want. Those are tables of standard freesurfer ROIs and are unrelated to your map-based analysis. You will need to do the cluster-wise correction for multiple comparisons using the voxel-wise threshold that FDR found. Set the cwp (cluster-wise p-value) threshold to 1 so that you get all of the clusters. When you do the clusterwise correction, it will produce a file with the values you want. doug On 7/23/12 2:44 AM, Oertel, Viola wrote: > Hi Doug, > I did the correction for multiple comparisons (FDR 0,05). Then, I have seen significant clusters in the map. After that, I run the generate tables stat (or so on) button and got two tables with thickness values. But I don't know which one of the tables is the right one because this step produced two files: > > > lh.aparc.a2005s.thickness.stats and > lh.aparc.thickness.stats > > May you tell me which one is the right, and are these results only the significant areas, or all computed once? > > > Thanks, > Viola > > ________________________________________ > Von: Douglas N Greve [greve(a)nmr.mgh.harvard.edu] > Gesendet: Freitag, 20. Juli 2012 21:26 > Bis: Oertel, Viola; Freesurfer(a)nmr.mgh.harvard.edu > Betreff: Re: AW: [Freesurfer] stats > > Hi Viola, those stats tables are unrelated to your map-based exploratory > analysis. If you have clusters in your map, you should run the > correction for multiple comparisons. This will give you significant > clusters. It will also produce a table of mean thicknesses in each > cluster for each subject which you can then import to SPSS. > doug > > On 07/19/2012 04:35 PM, Oertel, Viola wrote: >> Yes. I did the contrast pat. vs. cont and see some significant regions on the display. After that, I did this "generate" stats table. My idea was to put this scores into SPSS and run the inferential statistics there. But I am not sure which table I have to use. What I want to use are the thickness values of each participant of all significant regions which I can see on the template. >> Currently, I am not sure what kind of regions the tables show? Are that all regions, only the significant ones of the specific contrast I compute? I have seen that both tables show different regions. But I currently don't how the selection works. >> >> May you help with that? >> >> Thanks and best regards, >> >> Viola >> >> ________________________________________ >> Von: freesurfer-bounces(a)nmr.mgh.harvard.edu [freesurfer-bounces(a)nmr.mgh.harvard.edu]" im Auftrag von"Douglas N Greve [greve(a)nmr.mgh.harvard.edu] >> Gesendet: Donnerstag, 19. Juli 2012 17:30 >> Bis: freesurfer(a)nmr.mgh.harvard.edu >> Betreff: Re: [Freesurfer] stats >> >> Those stats tables just gives statistics (means, standard deviations) >> for thickness in different regions. They do not have inferential >> statistics. Have you run an analysis on patients vs controls? >> doug >> >> On 07/19/2012 09:27 AM, Oertel, Viola wrote: >>> Dear Doug, >>> >>> I have a question: when I create the stats tables, there are two tables: >>> >>> lh.aparc.a2005s.thickness.stats and >>> >>> lh.aparc.thickness.stats >>> >>> which one is the right one if I want to get the thickness values of >>> the significant clusters from the main effect (pat. / controls)? >>> >>> Thanks and best regards, >>> >>> Viola >>> >>> >>> >>> _______________________________________________ >>> Freesurfer mailing list >>> Freesurfer(a)nmr.mgh.harvard.edu >>> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer >> -- >> Douglas N. Greve, Ph.D. >> MGH-NMR Center >> greve(a)nmr.mgh.harvard.edu >> Phone Number: 617-724-2358 >> Fax: 617-726-7422 >> >> Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting >> FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html >> >> _______________________________________________ >> Freesurfer mailing list >> Freesurfer(a)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. >> >> > -- > Douglas N. Greve, Ph.D. > MGH-NMR Center > greve(a)nmr.mgh.harvard.edu > Phone Number: 617-724-2358 > Fax: 617-726-7422 > > Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting > FileDrop: www.nmr.mgh.harvard.edu/facility/filedrop/index.html > >

1 0

Directed search for group analysis
by Jeni Chen 23 Jul '12

23 Jul '12

Hello Freesurfer experts, I've been trying to figure out a way to run a directed search in a group analysis. I read the tutorial on ROI analysis and some of the posts here, but we can only do it with individual subject. Simply pout, I want to restrict my search areas to 4 ROIs (determined from the fsaverage or common template) when I run the group analysis (thus lowering the corrected stats threshold based on search criteria). What's the best way of doing it? Also, some posts suggests using the --mask option when running the glmfit command, but it seems some people also used the --label option. What is the difference? Thank you for your help. Jeni

2 1

MNI to RAS
by Jared Deck 23 Jul '12

23 Jul '12

Hi Doug, I am trying to convert MNI to RAS. I know the formula to convert from RAS to MNI is: MNI305RAS = TalXFM*Norig*inv(Torig)*[tkrR tkrA tkrS 1]' In theory, I should just be able to just invert this, but I am having difficulties because TalXFM is a non-square matrix and therefore not invertible. In addition, I tried using the moore-penrose pseudoinverse of the matrix but the values obtained do not match those in tkmedit. Do you have a formula for going from MNI to RAS or any advice about how would go about doing this? Thank you, Jared

2 1

issues with the Talairach transformation
by Mojdeh Zamyadi 23 Jul '12

23 Jul '12

Hi all, I've just started to do cortical thickness analysis on some subjects and I'm noticing some miss-registrations, more specifically the Talairach image is smaller than the target image no matter what subject I look at! I just wanted to make sure if this is something normal and to what extend it's going to effect the individual cortical thickness calculations. I guess I'm not sure how the Talairach trasnformation is used in further steps of the cortical thickness calculation. I've attached example coronal, sagittal and horizontal slices of one of my subjects. I greatly appreciate your advice. Thanks, Mojdeh Zamyadi, MSc Clinical Research Project Assistant The Hospital for Sick Children 555 University Avenue Toronto ON M5G 1X8

2 6

Questions about correction over 2 hemispheres and MCC
by Reem Jan 22 Jul '12

22 Jul '12

Hi Doug I hope it's okay that I double check I'm doing the correction over 2 hemispheres correctly and ask a question regarding correction for multiple comparisons. I will briefly describe what I have done: I have 2 groups of subjects - Drug addicts (n= 17) and controls (n=20). 1. I ran a surface thickness study: · For the 'mri_glmfit' command, I used DOSS and specified a contrast of +1 -1 0 (Controls > MA) · I ran a pre-cached simulation with the following command: mri_glmfit-sim --glmdir lh.group_age.glmdir --cache 4 pos · This simulation should only have corrected for the left hemisphere as I understand it. My cluster summary text file showed me that a cluster survived in the insula. I interpreted this as 'controls had higher grey matter thickness in this cluster located in the insula than drug addicts'. However, this was only corrected over the left hemisphere (when I ran the same simulation in the right hemisphere, nothing survived multiple comparison correction). · I now wanted to correct the lh results for 2 hemispheres. From the help menu of 'mri_glmfit-sim', I understood you can do this in 2 ways, are both of these correct and give the same result? a. mri_glmfit-sim --glmdir lh.group_age.glmdir --cache 4 pos --2 spaces --no-sim csdbase b. mri_glmfit-sim --glmdir lh.group_age.glmdir --cache 4 pos --cwpvalthresh 0.025 Ø No clusters survived the Bonferroni correction over both hemispheres. So I assume I can no longer report this result? 2. I ran a subcortical volume study: · I used SPSS to perform the statistical analysis on each of 14 subcortical structures (left and right). Is there a good way to correct for multiple comparisons, apart from Bonferroni, which would be very conservative? Many thanks for your advice. Kind regards Reem Reem Jan BPharm (Hons), RegPharmNZ PhD Student / Pharmacist School of Pharmacy, Faculty of Medical & Health Sciences, The University of Auckland, Private Bag 92019, Auckland, New Zealand. Ph: +64 9 373 7599 ext 81138. DDI: +64 9 923 1138 F: +64 9 367 7192 From: freesurfer-bounces(a)nmr.mgh.harvard.edu [mailto:freesurfer-bounces@nmr.mgh.harvard.edu] On Behalf Of Douglas Greve Sent: Friday, 20 July 2012 11:52 a.m. To: freesurfer(a)nmr.mgh.harvard.edu Subject: Re: [Freesurfer] slice-by-slice predictors Sorry, not possible. On 7/19/12 6:56 PM, Caspar M. Schwiedrzik wrote: Hi! Is it possible to have slice-by-slice predictors in Freesurfer, and if so, how? Thanks, Caspar _______________________________________________ Freesurfer mailing list Freesurfer(a)nmr.mgh.harvard.edu<mailto:Freesurfer@nmr.mgh.harvard.edu> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer

5 6

mris_refine_surfaces
by Colin Reveley 20 Jul '12

20 Jul '12

Hi - I'm using freesurfer with an ex-vivo MTR of a monkey, at 0.25mm. unfortunately the data doesn't fit into 256^3, and anyway things are much faster, and decent results are to be had by downsampling to 0.5mm. I did notice this mris_refine_surfaces. That looked great. I can fit 0.32mm into 256^3, so I did that, and called it 0.64mm, and treated it the same way I treated the 0.5mm (nu_correct etc etc) to get my "hires.mgz" for mris_refine_surfaces then I noticed mris_refine_surfaces wants ?h_hires.label as well. a) how would I make ?h_hires.label b) does my hires.mgz even have to be 256^3? If not, I could use the actual 250um data to refine the surface, and that would be nice. c) I don't suppose there would be any chance of not only using a hi res volume to refine the surfaces, but also upsampling the surfaces too? There is a reason I might want them upsampled. It would be nice for the ?h.white mesh to sort of approximately have a node for each voxel face. It's to do with more accurately mapping certain data in CARET later on. I'd settle for just how to make ?h_hires.label and use mris_refine_surfaces though :) cheers, Colin

2 3

← Newer
1
...
4
5
6
7
8
9
10
...
20
Older →

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Freesurfer July 2012