[Freesurfer] Best way to convert Philips par/rec to mgz
Hello, I have T1-weighted data from Philips 3 T scanner obtained at a resolution of 0.898 X 0.898 X 0.9 mm in both dicom (8 bit) and par/rec (pixel size = int16) format. I would like to preserve the acquisition resolution of the data in par/rec format; what is the best way to convert such data to mgz format for analysis in freesurfer? A command line example would be helpful, as I am new to freesurfer. Thanks for your help, Julie Julie E. McEntee, MA, CCRP Senior Research Program Coordinator Department of Psychiatry- Neuroimaging Johns Hopkins University School of Medicine 600 N. Wolfe St./Phipps 300 Baltimore, MD 21287 Phone: 410-502-0468 Fax: 410-614-3676 ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
[Freesurfer] GM & WM bitmaps; different from FAST(FSL)?
Hi, So far, I have results from autorecon1 & -subcortseg. The aseg.stats that are obtained following these procedures contain measures of L & R-Cerebral-White-Matter and Cerebral Cortex. If I created masks (bitmaps) using these labels in FreeSurfer, would they be equivalent to whole-brain WM and GM, or do I need to use maps/surfaces created further along in the FreeSurfer processing stream (if so, which would be accurate)? In general, how would these differ from the tissue segmentations obtained using FSL-FAST? Thank-you for your help, Julie Julie E. McEntee, MA, CCRP Senior Research Program Coordinator Department of Psychiatry- Neuroimaging Johns Hopkins University School of Medicine 600 N. Wolfe St./Phipps 300 Baltimore, MD 21287 Phone: 410-502-0468 Fax: 410-614-3676 ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] aseg/ICV question
Assuming the results of -subcortseg should be corrected for whole brain volume (for group comparison or use as covariate), what number should be used: ICV, BrainSegVol, total number of voxels? Should it be obtained from an image file (e.g., aseg.mgz) or stats (e.g., aseg.stats) file? Thank-you, Julie Julie E. McEntee, MA, CCRP Senior Research Program Coordinator Department of Psychiatry- Neuroimaging Johns Hopkins University School of Medicine 600 N. Wolfe St./Phipps 300 Baltimore, MD 21287 Phone: 410-502-0468 Fax: 410-614-3676 On 12/15/09 12:49 PM, "Douglas N Greve" wrote: > aseg.stats is created by counting the number of voxels in the aseg.mgz. > Actually, it includes partial volume correction, so it won't be exactly > the same as the voxel count. The ICV is an estimate only based on the > talairach transform. It's hard to say why two subjects would have the > same ICV. Maybe they have the same ICV or it might be an artifact of the > transform. Try checking the transforms on both of them to make sure they > are accurate. > > doug > > Rebecca Dautoff wrote: >> Hi all >> I am relatively unfamiliar with the structural side of analysis and am >> curious about how aseg.stats is created during recon. I have 2 subjects >> with the same ICV but volumes for other structures are different. What are >> the chances that this is just a coincidence and not an issue with the >> data? Assuming there is a mistake, how is it possible for there to be a >> mistake with ICV but not other structures? >> >> Thanks, >> Becca >> ___ >> 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
[Freesurfer] recon-all error during SubCort Seg
Hello, The initial command to recon-all: recon-all -i $FREESURFER_HOME/subjects/PPG1156/1156MPRage.nii.gz -autorecon1 -autorecon2 -subjid MGZ exited with errors during SubCort Seg. At this point, I am attempting to obtain subcortical segmentation only (aseg.mgz, wm.mgz, aseg.stats), but would like to eventually run entire recon. I previosly tried to attach the recon-all.log, but the file is too large (>200KB). I checked the tal transform and skullstrip- both looked OK. The scan is of a subject with Huntington's Disease so the ventricles are a bit enlarged and there is significant asymmetrical atrophy such that at least one area of brainmask.mgz was labeled as 57 (Right-Lesion). Also, there seems to be intensity variations (artifact?) A-P across cortex on slices around mid-line in sagittal view. Any ideas? Please let me know if you'd like me to upload any/all files. Thank you for your help, Julie Julie E. McEntee, M.A., C.C.R.P. Senior Research Program Coordinator Department of Psychiatry- Neuroimaging Johns Hopkins University School of Medicine 600 N. Wolfe St./ Phipps 300 Baltimore, MD 21287 Phone: 410-502-0468 Fax: 410-614-3676 ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] recon-all error during SubCort Seg
Bruce Fischl" wrote: > Hi Julie, > > can you send us the last couple of hundred or so lines of the > recon-all.log? You can generate this with tail -n 200 recon-all.log > > cheers, > Bruce > > > On Mon, 15 > Feb 2010, Julie McEntee wrote: > >> Hello, >> >> The initial command to recon-all: >> >> recon-all -i $FREESURFER_HOME/subjects/PPG1156/1156MPRage.nii.gz -autorecon1 >> -autorecon2 -subjid MGZ >> >> exited with errors during SubCort Seg. At this point, I am attempting to >> obtain subcortical segmentation only (aseg.mgz, wm.mgz, aseg.stats), but >> would like to eventually run entire recon. I previosly tried to attach the >> recon-all.log, but the file is too large (>200KB). >> >> I checked the tal transform and skullstrip- both looked OK. The scan is of a >> subject with Huntington's Disease so the ventricles are a bit enlarged and >> there is significant asymmetrical atrophy such that at least one area of >> brainmask.mgz was labeled as 57 (Right-Lesion). Also, there seems to be >> intensity variations (artifact?) A-P across cortex on slices around mid-line >> in sagittal view. >> >> Any ideas? Please let me know if you'd like me to upload any/all files. >> >> Thank you for your help, >> >> Julie >> >> Julie E. McEntee, M.A., C.C.R.P. >> Senior Research Program Coordinator >> Department of Psychiatry- Neuroimaging >> Johns Hopkins University School of Medicine >> 600 N. Wolfe St./ Phipps 300 >> Baltimore, MD 21287 >> Phone: 410-502-0468 >> Fax: 410-614-3676 >> ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] recon-all error during SubCort Seg
Yes & Yes- over 100 GB On 2/15/10 11:25 AM, "Bruce Fischl" wrote: > Hi Julie, > > do you have write access to the file > > aseg.auto_noCCseg.label_intensities.txt > > in the subject's mri dir? Is there enough disk space free? > > cheers > Bruce > > On Mon, 15 Feb 2010, Julie McEntee > wrote: > >> Hi Bruce, >> >> Here is the last part of the recon-all.log: >> >> #...@# SubCort Seg Mon Feb 15 01:13:58 EST 2010 >> \n mri_ca_label -align -nobigventricles norm.mgz transforms/talairach.m3z >> /Applications/freesurfer/average/RB_all_2008-03-26.gca aseg.auto_noCCseg.mgz >> \n >> renormalizing sequences with structure alignment, equivalent to: >>-renormalize >>-renormalize_mean 0.500 >>-regularize 0.500 >> not handling expanded ventricles... >> reading 1 input volumes... >> reading classifier array from >> /Applications/freesurfer/average/RB_all_2008-03-26.gca... >> reading input volume from norm.mgz... >> average std[0] = 6.9 >> reading transform from transforms/talairach.m3z... >> gunzip -c transforms/talairach.m3z >> Atlas used for the 3D morph was >> /Applications/freesurfer/average/RB_all_2008-03-26.gca >> average std = 6.9 using min determinant for regularization = 4.7 >> 0 singular and 0 ill-conditioned covariance matrices regularized >> labeling volume... >> renormalizing by structure alignment >> renormalizing input #0 >> gca peak = 0.15151 (27) >> mri peak = 0.10426 ( 7) >> Left_Lateral_Ventricle (4): linear fit = 0.38 x + 0.0 (7234 voxels, >> overlap=0.000) >> Left_Lateral_Ventricle (4): linear fit = 0.40 x + 0.0 (7234 voxels, peak = >> 10), gca=10.8 >> gca peak = 0.14982 (20) >> mri peak = 0.11303 (12) >> Right_Lateral_Ventricle (43): linear fit = 0.40 x + 0.0 (6302 voxels, >> overlap=0.000) >> Right_Lateral_Ventricle (43): linear fit = 0.40 x + 0.0 (6302 voxels, peak = >> 8), gca=8.0 >> gca peak = 0.28003 (97) >> mri peak = 0.06830 (115) >> Right_Pallidum (52): linear fit = 1.15 x + 0.0 (217 voxels, overlap=0.000) >> Right_Pallidum (52): linear fit = 1.15 x + 0.0 (217 voxels, peak = 112), >> gca=111.5 >> gca peak = 0.18160 (96) >> mri peak = 0.09282 (110) >> Left_Pallidum (13): linear fit = 1.15 x + 0.0 (198 voxels, overlap=0.115) >> Left_Pallidum (13): linear fit = 1.15 x + 0.0 (198 voxels, peak = 110), >> gca=110.4 >> gca peak = 0.27536 (62) >> mri peak = 0.06623 (61) >> Right_Hippocampus (53): linear fit = 0.93 x + 0.0 (1190 voxels, >> overlap=0.999) >> Right_Hippocampus (53): linear fit = 0.93 x + 0.0 (1190 voxels, peak = 57), >> gca=57.4 >> gca peak = 0.32745 (63) >> mri peak = 0.05769 (59) >> Left_Hippocampus (17): linear fit = 0.88 x + 0.0 (1114 voxels, >> overlap=0.982) >> Left_Hippocampus (17): linear fit = 0.88 x + 0.0 (1114 voxels, peak = 55), >> gca=55.1 >> gca peak = 0.08597 (105) >> mri peak = 0.06013 (110) >> Right_Cerebral_White_Matter (41): linear fit = 1.05 x + 0.0 (21713 voxels, >> overlap=0.653) >> Right_Cerebral_White_Matter (41): linear fit = 1.05 x + 0.0 (21713 voxels, >> peak = 110), gca=110.2 >> gca peak = 0.09209 (106) >> mri peak = 0.05762 (109) >> Left_Cerebral_White_Matter (2): linear fit = 1.05 x + 0.0 (21146 voxels, >> overlap=0.661) >> Left_Cerebral_White_Matter (2): linear fit = 1.05 x + 0.0 (21146 voxels, >> peak = 111), gca=111.3 >> gca peak = 0.07826 (63) >> mri peak = 0.02176 (57) >> Left_Cerebral_Cortex (3): linear fit = 0.90 x + 0.0 (44024 voxels, >> overlap=0.000) >> Left_Cerebral_Cortex (3): linear fit = 0.90 x + 0.0 (44024 voxels, peak = >> 57), gca=56.7 >> gca peak = 0.08598 (64) >> mri peak = 0.02190 (53) >> Right_Cerebral_Cortex (42): linear fit = 0.82 x + 0.0 (48705 voxels, >> overlap=0.000) >> Right_Cerebral_Cortex (42): linear fit = 0.82 x + 0.0 (48705 voxels, peak = >> 53), gca=52.8 >> gca peak = 0.24164 (71) >> mri peak = 0.05342 (82) >> Right_Caudate (50): linear fit = 1.10 x + 0.0 (234 voxels, overlap=0.226) >> Right_Caudate (50): linear fit = 1.10 x + 0.0 (234 voxels, peak = 78), >> gca=78.1 >> gca peak = 0.18227 (75) >> mri peak = 0.06054 (80) >> Left_Caudate (11): linear fit = 0.98 x + 0.0 (309 voxels, overlap=0.746) >> Left_Caudate (11): linear fit = 0.98 x + 0.0 (309 voxels, peak = 73), >> gca=73.1 >> gca peak = 0.10629 (62) >> mri peak = 0.03469 (55) >> Left_Cerebellum_Cortex (8): linear fit = 0.95 x + 0.0 (21006 voxels, >> overlap=0.584) >> Left_Cerebellum_
Re: [Freesurfer] Caudate volume error in Huntington´ s disease
I also use FS to identify subcortical atrophy with affected HD patients; have you checked the aseg, as well as white & pial surfaces for accuracy? I have found in some of my older data at lower resolution (1.5 T) that the intensities do not differ much in subcortical regions and FS has problems differentiating structures. If this is the issue, you can give FS intensity priors to go on ... Have to ask list re: exact procedure/command lines for this, as I have not experimented with this in my own data set as of yet. Hop this helps, Julie Julie E. McEntee, MA, CCRP Senior Research Program Coordinator Department of Psychiatry- Neuroimaging Johns Hopkins University School of Medicine 600 N. Wolfe St./Phipps 300 Baltimore, MD 21287 Phone: 410-502-0468 Fax: 410-614-3676 On 3/30/10 8:32 AM, "Diego Herrera" wrote: > Hello > I tried recon-all in a patient with Huntington´s disease with severe atrophy > of caudate nucleus (visually). It seems that freesurfer subcortical > segmentation failed because the volume reported is within normal limits. How > can I correct this? > > Thanks ___ 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] subcortical segmentation
I have used: recon-all -s -subcortseg which runs through step 11 (-calabel), with the understanding re: accuracy that Bruce pointed out. -Julie On 7/7/10 8:07 AM, "Bruce Fischl" wrote: > Hi Salem, > > yes, you can run any individual options in recon-all. See the -help output > and run it through step 11 if you want. Note that we do use the surfaces to > improve the aseg accuracy at the very end, so you'll be giving that up. > > cheers > Bruce > On > Wed, 7 Jul 2010, Hannoun wrote: > >> >> >> Hi, >> >> I am a new user of freesurfer. Is it possible to do just the volumetric >> processing stages without the surface processing stages? >> >> In fact what I want is to segment the subcortical structures. >> >> Is there a command line for this purpose only or should we run all the >> recon-all steps? >> >> Thank you for your help >> >> Salem >> >> >> >> >> >> > ___ > 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
