[Freesurfer] freesurfer@nmr.mgh.harvard.edu
Dear DougI'm looking for making the interhemispheric registration procedure of FS, I was just gertting familiar with it by reading the:http://surfer.nmr.mgh.harvard.edu/fswiki/XhemiI found that in the step # Correct for MC say that this need a new version of mri_glmfit-sim, is this version somehow available now? or it does refeer to the FS v5.1 version?I have two doubs about the whole procedure:which are the main differences between using the FS average and creating my own average?and second: if I decide to create my own atlas, after all the 1.3 procedure, should I then run the 1.2 steps subtituing the FSaverage_sym for mine?Many thanks in advanced,Gabriel ___ 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] Research Fellow/Postdoc: Functional MRI EEG Neuroimaging in Dementia
The Institute for Stroke and Dementia Research (ISD) is looking for a Research Fellow/Postdoc: Functional MRI EEG Neuroimaging in Dementia RESEARCH FOCUS and RESPONSIBILITIES: The research fellow/postdoc will work on multimodal imaging of human brain changes in early stages of Alzheimer's disease and cerebrovascular disease. The funded research project has a strong focus on the combined acquisition of functional MRI and high-density EEG along with other neuroimaging modalities. The establishment of the fMRI-EEG task-related paradigm and statistical analysis will be a key part of the responsibilities of research position. The fellow will work in a dynamically growing neuroimaging team and is encouraged to bring in own research ideas. REQUIREMENTS: Research experience in EEG and fMRI of brain changes is strongly encouraged. An independent work style and solid knowledge of programming and statistics is a plus. Familiarity with biological or clinical aspects of Alzheimer's disease is not a prerequisite. An enthusiastic, reliable and independent work style will provide a good fit for our team. OFFER: Our institute offers an excellent multidisciplinary environment and state-of-the-art multimodal neuroimaging equipment (3T MRI, high-density fMRI compatible EEG, molecular PET). The research fellow will have clearly defined task definitions, access to all data sets of the research project, and advanced technical support to facilitate efficient work conditions. The fellow will work in a friendly, dynamic, and highly productive team, headed by Prof. Michael Ewers, PhD. The University of Munich, Germany, and its clinics figure among Germany's premier and internationally competitive biomedical research sites funded by the government's German Universities Excellence Initiative. The Institute for Stroke and Dementia Research (ISD) is a thriving new research-dedicated institution of the University of Munich (http://www.klinikum.uni-muenchen.de/Institut-fuer-Schlaganfall-und-Demenzforschung/en/index.html). The ISD will be housed soon in a top-notch research facility that is currently being build and will eventually attract around 100 staff members, graduate students, scholars and foreign scholars. Salary is according to TV-L. The position is limited for two years, with a possible extension. Disabled persons will be preferentially considered in case of equal qualification. Presentation costs can unfortunately not be refunded. For more information, please contact Mr. Frühauf, Tel.: +49 (0)89 7095 7800 (E-Mail: i...@med.uni-muenchen.de). HOW TO APPLY: Your application - preferably in electronic form - with the usual documents indicating the earliest possible starting date should be directed to: Klinikum der Universität München, Institute for Stroke and Dementia Research Markus A. Frühauf, Managing Director ISD Heiglhofstr. 55 81377 Munich | Germany E-Mail: i...@med.uni-muenchen.de Viele Grüße Lilo Thomas Assistentin der Institutsleitung Institut für Schlaganfall- und Demenzforschung (ISD) Klinikum der Universität München Heiglhofstraße 55 D-81377 München Telefon: +49 (0)89 7095 - 7801 Fax: +49 (0)89 7095 - 8729 E-Mail: lilo.tho...@med.uni-muenchen.demailto:lilo.tho...@med.uni-muenchen.de Internet: www.isd.klinikum.uni-muenchen.dehttp://www.isd.klinikum.uni-muenchen.de/ ___ 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] Regarding cortical thickness
I am Bharti, a M.Tech./Ph.D student at School of Computer Systems Sciences, Jawaharlal Nehru University, New Delhi, India. My area of interest is pattern recognition and image processing. In one of my research work, i used freely available freesurfer software to find cortical thickness using recon-all -all command. But, while retrieving cortcal thickness maps using read_curv() function, i am getting different number of vertices for different brains. Are different number of cortical vertices possible for two different brains? I have gone through FAQ provided by Freesurfer but i did not find any related answer. Kindly clarify. Thanks in advance. -- Best Regards Bharti M.Tech./Ph.D. School of Computer Systems Sciences Jawaharlal Nehru University NEW DELHI, INDIA ___ 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.
Re: [Freesurfer] Regarding cortical thickness
yes, each brain has a different number, sufficient to get about 1mm isotropic resolution. You can use our spherical averaging to map them to a common coordinate system with the same # of vertices if you want (I think mri_surf2surf will do this) cheers Bruce On Wed, 21 Nov 2012, Bharti Rana wrote: I am Bharti, a M.Tech./Ph.D student at School of Computer Systems Sciences, Jawaharlal Nehru University, New Delhi, India. My area of interest is pattern recognition and image processing. In one of my research work, i used freely available freesurfer software to find cortical thickness using recon-all -all command. But, while retrieving cortcal thickness maps using read_curv() function, i am getting different number of vertices for different brains. Are different number of cortical vertices possible for two different brains? I have gone through FAQ provided by Freesurfer but i did not find any related answer. Kindly clarify. Thanks in advance. -- Best Regards Bharti M.Tech./Ph.D. School of Computer Systems Sciences Jawaharlal Nehru University NEW DELHI, INDIA ___ 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] selection of a restricted structure, easy question ?
Dear Freesurfers, I already have some experiment in segmentation : I just used FSL to perform cerebellar segmentation. I wanted to know if it's possible to run segmentation on a restricted structure (i.e cerebellum) instead of performing segmentation on the whole brain. It ll be easier for me and eventually faster since I have a lot of MRI. I know it's possible in FSL so my question is : can I do the same in freesurfer ? Thanks a lot, Charles (Paris) ___ 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.
Re: [Freesurfer] Pixelated DTI mask from Tracula
Hi Alessia - I'm not sure what you're referring to as the dti mask. Can you give me the file name of the image that you're displaying in the screehshot? It does look like noise. Thanks, a.y On Wed, 21 Nov 2012, Alessia Sarica wrote: Dear Freesurfer Users,I'm a Ph.D. Student and I'm working on tractography using Tracula, with the aim of conducting statistical analysis on the FA values. I have a cohort of 18 Amyotrophic Lateral Sclerosis Patients, and i'm doing all the process phases, starting from recon-all, ending to trac-all path. I've attached a text file with info about DTI DICOM extracted by ImageJ. Now, my problem is about the dti mask, as you can see on the attached image, a pixelated one is obtained, even if the left and right cst tracts seem good. This problem only occurs in few patients, not in all of them, even if the script is the same for all. This is the script i'm using: #Patient_name setenv SUBJECTS_DIR $TUTORIAL_DATA/Patient_recons set dtroot = $TUTORIAL_DATA/Patient_tracula set subjlist = (Patient_name) set runlist = (1) set dcmroot = $TUTORIAL_DATA/Patient_tracula set dcmlist = (Patient_name/orig/MR00) set bvalfile = $TUTORIAL_DATA/Patient_tracula/Patient_name/bfiles/Patient_name.bval set bvecfile = $TUTORIAL_DATA/Patient_tracula/Patient_name/bfiles/Patient_name.bvec set nb0 = 1 set dob0 = 0 set doeddy = 1 set dorotbvecs = 1 set usemaskanat = 0 set thrbet = 0.3 set doregflt = 1 set doregbbr = 0 set doregmni = 1 set mnitemp = $FSLDIR/data/standard/MNI152_T1_1mm_brain.nii.gz set pathlist = (lh.cst_AS rh.cst_AS \ lh.ilf_AS rh.ilf_AS \ lh.unc_AS rh.unc_AS \ fmajor_PP fminor_PP \ lh.atr_PP rh.atr_PP \ lh.cab_PP rh.cab_PP \ lh.ccg_PP rh.ccg_PP \ lh.slfp_PP rh.slfp_PP \ lh.slft_PP rh.slft_PP) set ncpts = 5 set nburnin = 200 set usetrunc = 1 set nkeep = 5 set nsample = 5000 I want to highlight that bval and bvec files are obtained from the tool dicom2nift, because Tracula is not able to extract them from DICOM header. I will really appreciate your help. Best regards, Alessia. -- Alessia Sarica Ph.D. Student Biomedical and Informatics Engineering, Bioinformatics Laboratory, Department of Surgical and Medical Sciences, University Magna Græcia of Catanzaro, Viale Europa (Località Germaneto), 88100 CATANZARO, ITALY. Phone: +3909613694193 ___ 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.
Re: [Freesurfer] freesurfer@nmr.mgh.harvard.edu
On 11/21/2012 05:48 AM, Gabriel Gonzalez Escamilla wrote: Dear Doug I'm looking for making the interhemispheric registration procedure of FS, I was just gertting familiar with it by reading the: http://surfer.nmr.mgh.harvard.edu/fswiki/Xhemi I found that in the step # Correct for MC say that this need a new version of mri_glmfit-sim, is this version somehow available now? or it does refeer to the FS v5.1 version? Actually, you don't need the new version anymore. Just run it without the --subject-override fsaverage argument. I'll change it on the wiki. I have two doubs about the whole procedure: which are the main differences between using the FS average and creating my own average? Your own atlas might allow for a better fit to the data. and second: if I decide to create my own atlas, after all the 1.3 procedure, should I then run the 1.2 steps subtituing the FSaverage_sym for mine? Yes. I would try it with the default fsaverage_sym. I had to go through 35 iterations to make sure fsaverage_sym was really symmetrical. This takes a very long time, so if your results are satisfactory with fsaverage_sym, I would use that. doug Many thanks in advanced, Gabriel -- Douglas N. Greve, Ph.D. MGH-NMR Center gr...@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 Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ ___ 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.
Re: [Freesurfer] selection of a restricted structure, easy question ?
Hi Charles, sorry, it is not possible. doug On 11/21/2012 10:51 AM, charles laidi wrote: Dear Freesurfers, I already have some experiment in segmentation : I just used FSL to perform cerebellar segmentation. I wanted to know if it's possible to run segmentation on a restricted structure (i.e cerebellum) instead of performing segmentation on the whole brain. It ll be easier for me and eventually faster since I have a lot of MRI. I know it's possible in FSL so my question is : can I do the same in freesurfer ? Thanks a lot, Charles (Paris) ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer -- Douglas N. Greve, Ph.D. MGH-NMR Center gr...@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 Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ ___ 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] wm vol mismatch
Dear Freesurfer,I'm trying to convert come labels from the aseg.mgz into nifti volumesI'm using: mri_extract_label aseg.mgz # and aseglabel#.nii where the # is the label as appears on the freesurferColoLUT.txtand it seems to work perfectly, but the number of voxels does not match with the number of voxels output in the aseg.stats, is this normal?they vary from i.e. 2603 in the aseg.stats to 2466 in the volume.to verify that it was't a problem from the extraction, I have tryed mris_binarize to match the same # segID as the mri_extract_label,and also use the matlab functions to open the aseg.mgz and keep those voxels which value were close to the #.With these two new options I get exactly the same number of voxels as with mri_extrac_label. So I don't know what could be the problem. Does anyone have an idea?Best Regards,Gabriel. ___ 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.
Re: [Freesurfer] Pixelated DTI mask from Tracula
Hi Alessia - So it's the FA map. Yes, this doesn't look like a correct FA map, there's no contrast between the white matter and everything else. The calculation of the FA has nothing to do with the freesurfer or tracula algorithms. It just uses your DWI data set, the gradient directions and the b-values, so these 3 things must be in disagreement if the FA doesn't come out right. I'd look into whether the subjects that have this problem were scanned with a different protocol or if the gradient directions and b-values were not extracted properly for them. Hope this helps, a.y On Wed, 21 Nov 2012, Alessia Sarica wrote: Hi Anastasia,thanks for your fast reply. I mean the file named dtifit_FA.nii I've re-run twice recon-all and tracula both, but it doesn't change the result. Best regards, Alessia. 2012/11/21 Anastasia Yendiki ayend...@nmr.mgh.harvard.edu Hi Alessia - I'm not sure what you're referring to as the dti mask. Can you give me the file name of the image that you're displaying in the screehshot? It does look like noise. Thanks, a.y On Wed, 21 Nov 2012, Alessia Sarica wrote: Dear Freesurfer Users,I'm a Ph.D. Student and I'm working on tractography using Tracula, with the aim of conducting statistical analysis on the FA values. I have a cohort of 18 Amyotrophic Lateral Sclerosis Patients, and i'm doing all the process phases, starting from recon-all, ending to trac-all path. I've attached a text file with info about DTI DICOM extracted by ImageJ. Now, my problem is about the dti mask, as you can see on the attached image, a pixelated one is obtained, even if the left and right cst tracts seem good. This problem only occurs in few patients, not in all of them, even if the script is the same for all. This is the script i'm using: #Patient_name setenv SUBJECTS_DIR $TUTORIAL_DATA/Patient_recons set dtroot = $TUTORIAL_DATA/Patient_tracula set subjlist = (Patient_name) set runlist = (1) set dcmroot = $TUTORIAL_DATA/Patient_tracula set dcmlist = (Patient_name/orig/MR00) set bvalfile = $TUTORIAL_DATA/Patient_tracula/Patient_name/bfiles/Patient_name.bval set bvecfile = $TUTORIAL_DATA/Patient_tracula/Patient_name/bfiles/Patient_name.bvec set nb0 = 1 set dob0 = 0 set doeddy = 1 set dorotbvecs = 1 set usemaskanat = 0 set thrbet = 0.3 set doregflt = 1 set doregbbr = 0 set doregmni = 1 set mnitemp = $FSLDIR/data/standard/MNI152_T1_1mm_brain.nii.gz set pathlist = (lh.cst_AS rh.cst_AS \ lh.ilf_AS rh.ilf_AS \ lh.unc_AS rh.unc_AS \ fmajor_PP fminor_PP \ lh.atr_PP rh.atr_PP \ lh.cab_PP rh.cab_PP \ lh.ccg_PP rh.ccg_PP \ lh.slfp_PP rh.slfp_PP \ lh.slft_PP rh.slft_PP) set ncpts = 5 set nburnin = 200 set usetrunc = 1 set nkeep = 5 set nsample = 5000 I want to highlight that bval and bvec files are obtained from the tool dicom2nift, because Tracula is not able to extract them from DICOM header. I will really appreciate your help. Best regards, Alessia. -- Alessia Sarica Ph.D. Student Biomedical and Informatics Engineering, Bioinformatics Laboratory, Department of Surgical and Medical Sciences, University Magna Græcia of Catanzaro, Viale Europa (Località Germaneto), 88100 CATANZARO, ITALY. Phone: +3909613694193 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. -- Alessia Sarica Ph.D. Student Biomedical and Informatics Engineering, Bioinformatics Laboratory, Department of Surgical and Medical Sciences, University Magna Græcia of Catanzaro, Viale Europa (Località Germaneto), 88100 CATANZARO, ITALY. Phone: +3909613694193 ___ Freesurfer mailing list
Re: [Freesurfer] Tool for incorporating T2-SPACE data in pial surface recons?
Would this new tool also be able to provide a better delineation of both pial surface and inner table of the skull (assuming T2 and not FLAIR i.e. hyperintense CSF)) and thus provide the possibility of computing extra-ventricular CSF and thus cerebral atrophy? Some of us would like that I'm sure :-) -roddy -Original Message- From: freesurfer-boun...@nmr.mgh.harvard.edu [mailto:freesurfer-boun...@nmr.mgh.harvard.edu] On Behalf Of Bruce Fischl Sent: Tuesday, November 20, 2012 2:02 PM To: Winter, Warren Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Tool for incorporating T2-SPACE data in pial surface recons? Hi Warren yes, it will be part of the upcoming 5.2 release, hopefully in Dec. It will use either a FLAIR or T2 (ideally T2-SPACE for either one) cheers Bruce On Tue, 20 Nov 2012, Winter, Warren wrote: Hi all, Back in January and October Bruce mentioned that he had under development some scripts designed to utilize T2-SPACE images for better pial surface reconstruction in the presence of dura -- I was just wondering if any of that is ready for trial? Thanks! Warren -- Warren Winter Research Coordinator Boston Children's Hospital Sheridan Laboratory of Cognitive Neuroscience Division of Developmental Medicine 1 Autumn Street, AU 650 Boston, MA 02215 857-218-5224 ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ 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. UT Southwestern Medical Center The future of medicine, today. ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] the positions of white/pial surfaces and the brain.mgz
Hi Longchuan, I'm still not sure I understand what you are trying to do ultimately. Do you just want a binary map of the surface in the DTI space so that you can use each cortex voxel as a seed? Or do you want to use one of the surface-based ROIs as the seed? This should be possible without re-running recon-all at 1mm. doug On 11/20/2012 06:01 PM, Longchuan Li wrote: Hi, Doug Thank you very much for the information. What I want to do is to use the white and pial surfaces in the FreeSurfer space as the seeds for surface-based tractography implemented in FSL. In order to do that, I need a transformation between the white matter surface and diffusion MR data. For some reason, the space of the white matter surface in my monkey data is not the same as that of the other volumetric data, such as filled.mgz, orig.mgz. This is however not the case for humans, which are processed using the standard FreeSurfer pipeline. To make things worse, I used cm option in recon-all and processed the data in their original resolution(0.5mm isotropic). When I used mris_fill without -c option, the left and right white volumes will have different dimensions, making it difficult to combine them for registering with filled.mgz. If I use -c option, I can combine them since they now have the same dimension. However, they would be conformed to 1mm isotropic (originally 0.5m isotropic), which is severely downsampled. I am afraid that this downsampling from 0.5mm to 1mm will affect the registration results. Do you have any suggestions on solving the issue? I am planning to change the resolution of the image from 0.5mm to 1mm without downsampling before I feed the image to recon-all. I noticed that there are two options in mri_convert (-vs, -cs) that are supposed to be related to this. However, I found both of them downsample the images. Is there any way that I could change the resolution of the images from 0.5mm to 1mm and in the mean time, magnifying the images so that the partial volume effect of the image will not change? Thank you ! Longchuan *From:* Douglas N Greve gr...@nmr.mgh.harvard.edu *To:* Longchuan Li leonad...@yahoo.com *Cc:* freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu; Bruce Fischl fis...@nmr.mgh.harvard.edu *Sent:* Tuesday, November 20, 2012 3:19 PM *Subject:* Re: [Freesurfer] the positions of white/pial surfaces and the brain.mgz Hi Longchuan, mris_fill create a volume with a different geometry than orig.mgz (though it is in the same space). What do you want to do exactly? Map the output of mris_fill into the DTI space? The transform from the output of mris_fill to orig.mgz is easy to obtain: tkregister2 --noedit --mov mris_fill.mgz --s subject --regheader --reg mris_fill.reg.dat this will be in register.dat format/space. doug On 11/19/2012 03:35 PM, Longchuan Li wrote: Hi, Doug I tried bbregister and checked the registration using tkregister2. It seems that this program registers the diffusion image with the anatomical *volume* images in FreeSurfer, such as brain.mgz. My problem is a little unique: I am working on monkey data and I found that my ?h.white surfaces do not have the identical sto_xyz coordinates as those of the anatomical volume images, such as brain.mgz. That is, when I use mris_fill to write the ?h.white to a volume file, I find that the center of the images in the header file is not identical to that of the anatomical volume images. In my human data processed using FreeSurfer standard recon-all pipeline, they are identical. Since I am working on monkey data, I used -notalairach and -notal-check in recon-all to have the piepline work. I suspect this may be related to these options. So my questions are: (1) what do you think are the reasons causing the inconsistent sto_xyz values between the surface files (?h.white) and voluem files in the non-human data. (2) Is there any way that I could find a transformation matrix between the ?h.white and my FreeSurfer volume images (such as brain.mgz)? thank you! Longchuan PS: I also used -cm option in recon-all as suggested by Bruce. I just got the results and the problem is still there. *From:* Douglas N Greve gr...@nmr.mgh.harvard.edu mailto:gr...@nmr.mgh.harvard.edu *To:* freesurfer@nmr.mgh.harvard.edu mailto:freesurfer@nmr.mgh.harvard.edu *Sent:* Monday, November 19, 2012 10:43 AM *Subject:* Re: [Freesurfer] the positions of white/pial surfaces and the brain.mgz use bbregister, something like bbregister --s subjectname --init-fsl --t2 --mov lowb.nii --reg register.dat where lowb is the low b (usually b=0) volume from your DTI, the one that you used as the template for motion correction. doug On 11/19/2012 10:30
Re: [Freesurfer] contrast matrix
Hi Bo, I don't understand the contrast you are trying to make. Are you really trying to compute the interaction between four variables (factor 1, factor 2, gender, and age)? doug On 11/21/2012 10:49 AM, xiangbo_2010 wrote: Dear Donald McLaren Thank you for your reply! I make the contrast according to your method is following,but I want to make interaction between factor 1 (A,B)and factor 2(C,D,E), gender (M,F) and one continuous variable (age) as covariates, the following contrast: 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 is correct? Thanks! Bo Xiang At 2012-11-21 04:07:37,MCLAREN, Donaldmclaren.don...@gmail.com mailto:mclaren.don...@gmail.com wrote: On Tue, Nov 20, 2012 at 12:56 PM, Douglas N Greve gr...@nmr.mgh.harvard.edu mailto:gr...@nmr.mgh.harvard.edu wrote: Thanks Donald. Is this the standard way to do this? I had used 8 rows instead of 4 with the difference being that 8 rows gives you an opportunity to look for an effect in males OR females. Yes. Having 8 rows would tell you if you have an interaction between factor 1 and 2 in either males or females. My 4 rows only tell you if the interaction exists. Technically speaking, one would run the three-way interaction first. If nothing existed then you do the two-way interaction as I suggested. If there is a three-way interaction, then you would use Doug's approach of the interaction in either males or females. If there is an effect in both males and females but the effects go in opposite directions, then the 4 row implementation will resolve to 0 (no effect). Or am I misunderstanding something (again:)? Nope. You are right. If the male and female effects are different, then they could cancel each other out. If you suspect this to be the case, then you should be able to demonstrate a three-way interaction. thanks! doug On 11/20/2012 01:50 PM, MCLAREN, Donald wrote: Bo, Doug asked me to chime in on your issue. Here are some points that you (and others) will hopefully find useful. (1) Inferences are two-step process. First, you create and estimate the design matrix. Every column in the design matrix accounts can account for some of the variance in the data. Second, you have contrasts that allow you to infer specific effects. Because the model contains your covariates, you are always controlling for the covariates and by extension any factor/covariate not in the contrast. (2) Forming contrasts is often the most difficult thing to do. I assume that your three factors (1, 2, and gender) are all between-subject factors. If one of them is a within-subject factor please let me know and disregard the rest of the email. The final F-contrast will have 4 rows (factor 1 levels-1)*(factor 2 levels -1)=(3-1)*(3-1)=2*2=4 The following is an outline for creating contrasts: (a) Start simple - difference between levels of 1 factor (b) Define your null hypothesis: AO=AP=AQ (c) Make it equal to 0: AO-AP=0 AND AP-AQ=0 (d) Repeat for the other levels of the factor... BO-BP=0 AND BP-BQ=0 CO-CP=0 AND CP-CQ=0 (e) Now combine them AO-AP=BO-BP=CO-CP AND AP-AQ=BP-BQ=CP-CQ (f) Make them equal to 0: AO-AP-BO+BP=0 BO-BP-CO+CP=0 AP-AQ-BP+BQ=0 BP-BQ-CP+CQ=0 (g) Expand them to include gender, for example: AO-AP-BO+BP=0 becomes FAO-FAP-FBO+FBP+MAO-MAP-MBO+MBP=0 Since the contrast now has 2 columns per level, you should divide all values by 2. This will produce the correct amplitude and statistics. If you leave the values as 1 and -1, then you will have an incorrect amplitude, but the statistics will still be correct. (h) Fill in the respective columns of your design matrix. (3) The degrees of freedom are defined based on the rows of the F-matrix and the number of rows in the design matrix. The F-test has a numerator and denominator degrees of freedom. F(n,d). Best Regards, Donald McLaren = D.G. McLaren, Ph.D. Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School Postdoctoral Research Fellow, GRECC, Bedford VA Website: http://www.martinos.org/~mclaren Office: (773) 406-2464 = This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is intended only for the use of the individual or entity named above. If the reader of the e-mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that you are in possession of confidential and privileged information. Any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited and may be
Re: [Freesurfer] wm vol mismatch
Hi Gabriel, this is normal. The volumes listed with aseg.stats have been corrected for partial volume, so it is not as simple as counting up all the voxels. doug On 11/21/2012 12:36 PM, Gabriel Gonzalez Escamilla wrote: Dear Freesurfer, I'm trying to convert come labels from the aseg.mgz into nifti volumes I'm using: mri_extract_label aseg.mgz # and aseglabel#.nii where the # is the label as appears on the freesurferColoLUT.txt and it seems to work perfectly, but the number of voxels does not match with the number of voxels output in the aseg.stats, is this normal? they vary from i.e. 2603 in the aseg.stats to 2466 in the volume. to verify that it was't a problem from the extraction, I have tryed mris_binarize to match the same # segID as the mri_extract_label, and also use the matlab functions to open the aseg.mgz and keep those voxels which value were close to the #. With these two new options I get exactly the same number of voxels as with mri_extrac_label. So I don't know what could be the problem. Does anyone have an idea? Best Regards, Gabriel. ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer -- Douglas N. Greve, Ph.D. MGH-NMR Center gr...@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 Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ ___ 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.
Re: [Freesurfer] problem with c_ras in converted AFNI .BRIK
Hi Tom, it is really necessary to use BRIKs these days? AFNI has supported nifti for years. While it is possible for me to track this down and fix it, it would take me a while (time I'd rather spend doing something else:). doug ps. On the other hand, since you are probably the only one using BIRKS, I can just applied the patch you sent. I can give you a binary to test. If it tests out, then I'll just check the patch in. Let me know. On 11/19/2012 08:21 PM, Tom Holroyd (NIH/NIMH) [E] wrote: If I do 3dcopy ortho+orig ortho.nii then 3dinfo shows that the two datasets share the same coordinate system, however when I use mri_info, the results differ. For example, 3dinfo, both .BRIK and .nii: Geometry String: MATRIX(0,0,-1,127,1,0,0,-127,0,-1,0,180):256,256,256 R-to-L extent: -128.000 [R] -to- 127.000 [L] -step- 1.000 mm [256 voxels] A-to-P extent: -127.000 [A] -to- 128.000 [P] -step- 1.000 mm [256 voxels] I-to-S extent: -75.000 {I] -to- 180.000 [S] -step- 1.000 mm [256 voxels]* mri_info ortho+orig says: voxel to ras transform: 0. 0. 1. -128.5000 -1. 0. 0. 127.5000 0. -1. 0.75.5000 0. 0. 0. 1. while mri_info ortho.nii produces (this is correct): voxel to ras transform: -0. -0. 1. -127. -1. -0. -0. 127. 0. -1. 0. 180. 0. 0. 0. 1. in particular, the center of the volume appears to be translated. So if you run FreeSurfer on the .nii file, the surface is shifted relative to what you get when you run FreeSurfer on the .BRIK file, even though the original input was the same. Also mri_info seems to have shifted things to half-integer coordinates, while the .nii version doesn't. I'm not really sure about the half-integer thing, but it's a separate issue. I think the problem is a bug in utils/afni.c, where it calculates c_ras. I've attached a patch. The problem is that FreeSurfer wants ras coordinates, which are LPI coordinates in AFNI notation, and we're staring in dataset coordinates (here ASL). First the transform matrix is computed (correctly). Then c_ras is calculated using the AFNI DELTA and ORIGIN header fields. I'm really not sure what the old code was doing, the terms seem to be associated the wrong way. The new code correctly multiplies the ORIGIN times the DELTAs to get all negative numbers in dataset coordinates (that is, the origin). This is then added to the center, also in dataset coordinates, and then this is multiplied by the rest of the transform matrix, which is a vox2ras matrix, yielding c_ras that agrees with the result from the .nii file. Except for that half-integer thing, which boils down to the afni.c code doing (w-1)/2 and the .nii file code doing w/2. Perhaps the AFNI code shouldn't subtract 1 either ? {the relevant lines are also part of the patch, but I left in the (w-1)/2} Unfortunately it seems there is no easy way to compute the correct c_ras if you started with a .BRIK, unless you have the original .BRIK. In that case 3dcopy it to a .nii file and mri_info that. Err, I haven't actually tested the patch ... I'm fairly certain it's the .nii file results that are correct, and not the other way around. It would be nice to confirm that this happens with other datasets besides the one I tried. It's easy to check if you have an AFNI dataset sitting around. Thanks, Tom ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer -- Douglas N. Greve, Ph.D. MGH-NMR Center gr...@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 Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/ ___ 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.
Re: [Freesurfer] contrast matrix
No, I want to compute the interaction between factor 1 and factor 2 , gender and age as covariates. thanks Bo Xiang 在 2012-11-22 04:22:06,Douglas N Greve gr...@nmr.mgh.harvard.edu 写道: Hi Bo, I don't understand the contrast you are trying to make. Are you really trying to compute the interaction between four variables (factor 1, factor 2, gender, and age)? doug On 11/21/2012 10:49 AM, xiangbo_2010 wrote: Dear Donald McLaren Thank you for your reply! I make the contrast according to your method is following,but I want to make interaction between factor 1 (A,B)and factor 2(C,D,E), gender (M,F) and one continuous variable (age) as covariates, the following contrast: 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 is correct? Thanks! Bo Xiang At 2012-11-21 04:07:37,MCLAREN, Donaldmclaren.don...@gmail.com mailto:mclaren.don...@gmail.com wrote: On Tue, Nov 20, 2012 at 12:56 PM, Douglas N Greve gr...@nmr.mgh.harvard.edu mailto:gr...@nmr.mgh.harvard.edu wrote: Thanks Donald. Is this the standard way to do this? I had used 8 rows instead of 4 with the difference being that 8 rows gives you an opportunity to look for an effect in males OR females. Yes. Having 8 rows would tell you if you have an interaction between factor 1 and 2 in either males or females. My 4 rows only tell you if the interaction exists. Technically speaking, one would run the three-way interaction first. If nothing existed then you do the two-way interaction as I suggested. If there is a three-way interaction, then you would use Doug's approach of the interaction in either males or females. If there is an effect in both males and females but the effects go in opposite directions, then the 4 row implementation will resolve to 0 (no effect). Or am I misunderstanding something (again:)? Nope. You are right. If the male and female effects are different, then they could cancel each other out. If you suspect this to be the case, then you should be able to demonstrate a three-way interaction. thanks! doug On 11/20/2012 01:50 PM, MCLAREN, Donald wrote: Bo, Doug asked me to chime in on your issue. Here are some points that you (and others) will hopefully find useful. (1) Inferences are two-step process. First, you create and estimate the design matrix. Every column in the design matrix accounts can account for some of the variance in the data. Second, you have contrasts that allow you to infer specific effects. Because the model contains your covariates, you are always controlling for the covariates and by extension any factor/covariate not in the contrast. (2) Forming contrasts is often the most difficult thing to do. I assume that your three factors (1, 2, and gender) are all between-subject factors. If one of them is a within-subject factor please let me know and disregard the rest of the email. The final F-contrast will have 4 rows (factor 1 levels-1)*(factor 2 levels -1)=(3-1)*(3-1)=2*2=4 The following is an outline for creating contrasts: (a) Start simple - difference between levels of 1 factor (b) Define your null hypothesis: AO=AP=AQ (c) Make it equal to 0: AO-AP=0 AND AP-AQ=0 (d) Repeat for the other levels of the factor... BO-BP=0 AND BP-BQ=0 CO-CP=0 AND CP-CQ=0 (e) Now combine them AO-AP=BO-BP=CO-CP AND AP-AQ=BP-BQ=CP-CQ (f) Make them equal to 0: AO-AP-BO+BP=0 BO-BP-CO+CP=0 AP-AQ-BP+BQ=0 BP-BQ-CP+CQ=0 (g) Expand them to include gender, for example: AO-AP-BO+BP=0 becomes FAO-FAP-FBO+FBP+MAO-MAP-MBO+MBP=0 Since the contrast now has 2 columns per level, you should divide all values by 2. This will produce the correct amplitude and statistics. If you leave the values as 1 and -1, then you will have an incorrect amplitude, but the statistics will still be correct. (h) Fill in the respective columns of your design matrix. (3) The degrees of freedom are defined based on the rows of the F-matrix and the number of rows in the design matrix. The F-test has a numerator and denominator degrees of freedom. F(n,d). Best Regards, Donald McLaren = D.G. McLaren, Ph.D. Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School Postdoctoral Research Fellow, GRECC, Bedford VA Website: http://www.martinos.org/~mclaren Office: (773) 406-2464 = This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is intended only for the use of the individual or entity named above. If the reader of the e-mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that you are in possession of confidential
Re: [Freesurfer] the positions of white/pial surfaces and the brain.mgz
Hi, Doug Thank you for following up the issue. I think that I have solved the problem. What I wanted was to derive a binary map of the surface in the DTI space so that each cortex voxel can be used as a seed. Thank you very much for the help! Longchuan From: Douglas N Greve gr...@nmr.mgh.harvard.edu To: Longchuan Li leonad...@yahoo.com Cc: freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu Sent: Wednesday, November 21, 2012 3:20 PM Subject: Re: [Freesurfer] the positions of white/pial surfaces and the brain.mgz Hi Longchuan, I'm still not sure I understand what you are trying to do ultimately. Do you just want a binary map of the surface in the DTI space so that you can use each cortex voxel as a seed? Or do you want to use one of the surface-based ROIs as the seed? This should be possible without re-running recon-all at 1mm. doug On 11/20/2012 06:01 PM, Longchuan Li wrote: Hi, Doug Thank you very much for the information. What I want to do is to use the white and pial surfaces in the FreeSurfer space as the seeds for surface-based tractography implemented in FSL. In order to do that, I need a transformation between the white matter surface and diffusion MR data. For some reason, the space of the white matter surface in my monkey data is not the same as that of the other volumetric data, such as filled.mgz, orig.mgz. This is however not the case for humans, which are processed using the standard FreeSurfer pipeline. To make things worse, I used cm option in recon-all and processed the data in their original resolution(0.5mm isotropic). When I used mris_fill without -c option, the left and right white volumes will have different dimensions, making it difficult to combine them for registering with filled.mgz. If I use -c option, I can combine them since they now have the same dimension. However, they would be conformed to 1mm isotropic (originally 0.5m isotropic), which is severely downsampled. I am afraid that this downsampling from 0.5mm to 1mm will affect the registration results. Do you have any suggestions on solving the issue? I am planning to change the resolution of the image from 0.5mm to 1mm without downsampling before I feed the image to recon-all. I noticed that there are two options in mri_convert (-vs, -cs) that are supposed to be related to this. However, I found both of them downsample the images. Is there any way that I could change the resolution of the images from 0.5mm to 1mm and in the mean time, magnifying the images so that the partial volume effect of the image will not change? Thank you ! Longchuan *From:* Douglas N Greve gr...@nmr.mgh.harvard.edu *To:* Longchuan Li leonad...@yahoo.com *Cc:* freesurfer@nmr.mgh.harvard.edu freesurfer@nmr.mgh.harvard.edu; Bruce Fischl fis...@nmr.mgh.harvard.edu *Sent:* Tuesday, November 20, 2012 3:19 PM *Subject:* Re: [Freesurfer] the positions of white/pial surfaces and the brain.mgz Hi Longchuan, mris_fill create a volume with a different geometry than orig.mgz (though it is in the same space). What do you want to do exactly? Map the output of mris_fill into the DTI space? The transform from the output of mris_fill to orig.mgz is easy to obtain: tkregister2 --noedit --mov mris_fill.mgz --s subject --regheader --reg mris_fill.reg.dat this will be in register.dat format/space. doug On 11/19/2012 03:35 PM, Longchuan Li wrote: Hi, Doug I tried bbregister and checked the registration using tkregister2. It seems that this program registers the diffusion image with the anatomical *volume* images in FreeSurfer, such as brain.mgz. My problem is a little unique: I am working on monkey data and I found that my ?h.white surfaces do not have the identical sto_xyz coordinates as those of the anatomical volume images, such as brain.mgz. That is, when I use mris_fill to write the ?h.white to a volume file, I find that the center of the images in the header file is not identical to that of the anatomical volume images. In my human data processed using FreeSurfer standard recon-all pipeline, they are identical. Since I am working on monkey data, I used -notalairach and -notal-check in recon-all to have the piepline work. I suspect this may be related to these options. So my questions are: (1) what do you think are the reasons causing the inconsistent sto_xyz values between the surface files (?h.white) and voluem files in the non-human data. (2) Is there any way that I could find a transformation matrix between the ?h.white and my FreeSurfer volume images (such as brain.mgz)? thank you! Longchuan PS: I also used -cm option in recon-all as suggested by Bruce. I just got the results and the problem is still there.
Re: [Freesurfer] contrast matrix
Please explain what are the columns represent. Best Regards, Donald McLaren = D.G. McLaren, Ph.D. Research Fellow, Department of Neurology, Massachusetts General Hospital and Harvard Medical School Postdoctoral Research Fellow, GRECC, Bedford VA Website: http://www.martinos.org/~mclaren Office: (773) 406-2464 = This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is intended only for the use of the individual or entity named above. If the reader of the e-mail is not the intended recipient or the employee or agent responsible for delivering it to the intended recipient, you are hereby notified that you are in possession of confidential and privileged information. Any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited and may be unlawful. If you have received this e-mail unintentionally, please immediately notify the sender via telephone at (773) 406-2464 or email. On Wed, Nov 21, 2012 at 2:50 PM, xiangbo_2...@126.com wrote: No, I want to compute the interaction between factor 1 and factor 2 , gender and age as covariates. thanks Bo Xiang 在 2012-11-22 04:22:06,Douglas N Greve gr...@nmr.mgh.harvard.edu 写道: Hi Bo, I don't understand the contrast you are trying to make. Are you really trying to compute the interaction between four variables (factor 1, factor 2, gender, and age)? doug On 11/21/2012 10:49 AM, xiangbo_2010 wrote: Dear Donald McLaren Thank you for your reply! I make the contrast according to your method is following,but I want to make interaction between factor 1 (A,B)and factor 2(C,D,E), gender (M,F) and one continuous variable (age) as covariates, the following contrast: 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 -2 0 -2 2 0 2 -2 0 -2 2 0 0 0 0 0 0 0 0 0 0 0 0 is correct? Thanks! Bo Xiang At 2012-11-21 04:07:37,MCLAREN, Donaldmclaren.don...@gmail.com mailto:mclaren.don...@gmail.com wrote: On Tue, Nov 20, 2012 at 12:56 PM, Douglas N Greve gr...@nmr.mgh.harvard.edu mailto:gr...@nmr.mgh.harvard.edu wrote: Thanks Donald. Is this the standard way to do this? I had used 8 rows instead of 4 with the difference being that 8 rows gives you an opportunity to look for an effect in males OR females. Yes. Having 8 rows would tell you if you have an interaction between factor 1 and 2 in either males or females. My 4 rows only tell you if the interaction exists. Technically speaking, one would run the three-way interaction first. If nothing existed then you do the two-way interaction as I suggested. If there is a three-way interaction, then you would use Doug's approach of the interaction in either males or females. If there is an effect in both males and females but the effects go in opposite directions, then the 4 row implementation will resolve to 0 (no effect). Or am I misunderstanding something (again:)? Nope. You are right. If the male and female effects are different, then they could cancel each other out. If you suspect this to be the case, then you should be able to demonstrate a three-way interaction. thanks! doug On 11/20/2012 01:50 PM, MCLAREN, Donald wrote: Bo, Doug asked me to chime in on your issue. Here are some points that you (and others) will hopefully find useful. (1) Inferences are two-step process. First, you create and estimate the design matrix. Every column in the design matrix accounts can account for some of the variance in the data. Second, you have contrasts that allow you to infer specific effects. Because the model contains your covariates, you are always controlling for the covariates and by extension any factor/covariate not in the contrast. (2) Forming contrasts is often the most difficult thing to do. I assume that your three factors (1, 2, and gender) are all between-subject factors. If one of them is a within-subject factor please let me know and disregard the rest of the email. The final F-contrast will have 4 rows (factor 1 levels-1)*(factor 2 levels -1)=(3-1)*(3-1)=2*2=4 The following is an outline for creating contrasts: (a) Start simple - difference between levels of 1 factor (b) Define your null hypothesis: AO=AP=AQ (c) Make it equal to 0: AO-AP=0 AND AP-AQ=0 (d) Repeat for the other levels of the factor... BO-BP=0 AND BP-BQ=0 CO-CP=0 AND CP-CQ=0 (e) Now combine them AO-AP=BO-BP=CO-CP AND AP-AQ=BP-BQ=CP-CQ (f) Make them equal to 0: AO-AP-BO+BP=0 BO-BP-CO+CP=0 AP-AQ-BP+BQ=0 BP-BQ-CP+CQ=0 (g) Expand them to include gender, for example: AO-AP-BO+BP=0 becomes FAO-FAP-FBO+FBP+MAO-MAP-MBO+MBP=0 Since the contrast now has 2 columns per level, you should divide all values by 2. This will produce the correct amplitude and statistics. If you leave the
Re: [Freesurfer] Tool for incorporating T2-SPACE data in pial surface recons?
Yes, although we've only done the pial part so far Bruce On Nov 21, 2012, at 3:06 PM, Roderick McColl roderick.mcc...@utsouthwestern.edu wrote: Would this new tool also be able to provide a better delineation of both pial surface and inner table of the skull (assuming T2 and not FLAIR i.e. hyperintense CSF)) and thus provide the possibility of computing extra-ventricular CSF and thus cerebral atrophy? Some of us would like that I'm sure :-) -roddy -Original Message- From: freesurfer-boun...@nmr.mgh.harvard.edu [mailto:freesurfer-boun...@nmr.mgh.harvard.edu] On Behalf Of Bruce Fischl Sent: Tuesday, November 20, 2012 2:02 PM To: Winter, Warren Cc: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Tool for incorporating T2-SPACE data in pial surface recons? Hi Warren yes, it will be part of the upcoming 5.2 release, hopefully in Dec. It will use either a FLAIR or T2 (ideally T2-SPACE for either one) cheers Bruce On Tue, 20 Nov 2012, Winter, Warren wrote: Hi all, Back in January and October Bruce mentioned that he had under development some scripts designed to utilize T2-SPACE images for better pial surface reconstruction in the presence of dura -- I was just wondering if any of that is ready for trial? Thanks! Warren -- Warren Winter Research Coordinator Boston Children's Hospital Sheridan Laboratory of Cognitive Neuroscience Division of Developmental Medicine 1 Autumn Street, AU 650 Boston, MA 02215 857-218-5224 ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ 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. UT Southwestern Medical Center The future of medicine, today. ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer