Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu Sent: Wednesday, July 5, 2017 7:28 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferhttps://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.edu_mailman_listinfo_freesurfer&d=DwMC-g&c=JeTkUgVztGMmhKYjxsy2rfoWYibK1YmxXez1G3oNStg&r=bOaERSxG_JqDNQmM_8K52LCjnCr_2F55jAz7ZN-9dqg&m=DhD2aaHH4v0l556v1tTQWGefdhjxGef9QZtInzhlGAw&s=bQIJi6_pIGgShFkxwNFbAaJNwv2UyXNY15PnOqVX3Jk&e=
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Wednesday, July 5, 2017 7:28 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* 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
?When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz?
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu Sent: Thursday, July 6, 2017 12:58 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edumailto:greve@nmr.mgh.harvard.edu Sent: Wednesday, July 5, 2017 7:28 PM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferhttps://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.edu_mailman_listinfo_freesurfer&d=DwMC-g&c=JeTkUgVztGMmhKYjxsy2rfoWYibK1YmxXez1G3oNStg&r=bOaERSxG_JqDNQmM_8K52LCjnCr_2F55jAz7ZN-9dqg&m=DhD2aaHH4v0l556v1tTQWGefdhjxGef9QZtInzhlGAw&s=bQIJi6_pIGgShFkxwNFbAaJNwv2UyXNY15PnOqVX3Jk&e=
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferhttps://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.edu_mailman_listinfo_freesurfer&d=DwMD-g&c=JeTkUgVztGMmhKYjxsy2rfoWYibK1YmxXez1G3oNStg&r=bOaERSxG_JqDNQmM_8K52LCjnCr_2F55jAz7ZN-9dqg&m=TWY5bMFcl8S9uGNwfRrVXyP12FLvSoXujeIDNNeeP_s&s=LML38gfagtoUifRTCxv8DcSmq6xJlb4dNszfTDy2sls&e=
Add --no-prune
On 7/7/17 11:38 AM, Sadie Marvel wrote:
When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Thursday, July 6, 2017 12:58 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Wednesday, July 5, 2017 7:28 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* 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
So now that I am using --no-prune, I'm getting this error: group wrfx error: MRInormWeights: value less than or eq to 0. ?I found another thread in the mailing list about this error saying that there may be values in cesvar.nii that are =0, and that pruning normally masks out these values. But here, I am trying to use my own masks to do so. Is it possible now that my masks and my --y input have different sizes? If so, how can I solve this? ________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu Sent: Friday, July 7, 2017 11:57 AM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
Add --no-prune
On 7/7/17 11:38 AM, Sadie Marvel wrote:
?When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz?
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edumailto:greve@nmr.mgh.harvard.edu Sent: Thursday, July 6, 2017 12:58 PM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edumailto:freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edumailto:greve@nmr.mgh.harvard.edu Sent: Wednesday, July 5, 2017 7:28 PM To: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferhttps://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.edu_mailman_listinfo_freesurfer&d=DwMC-g&c=JeTkUgVztGMmhKYjxsy2rfoWYibK1YmxXez1G3oNStg&r=bOaERSxG_JqDNQmM_8K52LCjnCr_2F55jAz7ZN-9dqg&m=DhD2aaHH4v0l556v1tTQWGefdhjxGef9QZtInzhlGAw&s=bQIJi6_pIGgShFkxwNFbAaJNwv2UyXNY15PnOqVX3Jk&e=
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferhttps://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.edu_mailman_listinfo_freesurfer&d=DwMD-g&c=JeTkUgVztGMmhKYjxsy2rfoWYibK1YmxXez1G3oNStg&r=bOaERSxG_JqDNQmM_8K52LCjnCr_2F55jAz7ZN-9dqg&m=TWY5bMFcl8S9uGNwfRrVXyP12FLvSoXujeIDNNeeP_s&s=LML38gfagtoUifRTCxv8DcSmq6xJlb4dNszfTDy2sls&e=
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edumailto:Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurferhttps://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.edu_mailman_listinfo_freesurfer&d=DwMDaQ&c=JeTkUgVztGMmhKYjxsy2rfoWYibK1YmxXez1G3oNStg&r=bOaERSxG_JqDNQmM_8K52LCjnCr_2F55jAz7ZN-9dqg&m=3HRkrJHzDtJvcUaTv1eXqK72g83qXHE_YRbtRif2vWU&s=kJ8Lycv6GJRpQ-ROcmLLZoVEmBmFLirfHaae0TZaLYY&e=
it might not be possible to use the frame mask with wls. Does it work without it?
On 07/09/2017 04:46 PM, Sadie Marvel wrote:
So now that I am using --no-prune, I'm getting this error: group wrfx error: MRInormWeights: value less than or eq to 0. I found another thread in the mailing list about this error saying that there may be values in cesvar.nii that are =0, and that pruning normally masks out these values. But here, I am trying to use my own masks to do so. Is it possible now that my masks and my --y input have different sizes? If so, how can I solve this?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Friday, July 7, 2017 11:57 AM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
Add --no-prune
On 7/7/17 11:38 AM, Sadie Marvel wrote:
When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Thursday, July 6, 2017 12:58 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Wednesday, July 5, 2017 7:28 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* 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 command will run now without --wls, but I'm still not sure the output is correct. Here is my entire output: cwd /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh cmdline mri_glmfit.bin --y c1_2/ces.nii --osgm --no-prune --frame-mask masks.nii --surface fsaverage lh --glmdir c1_2/my-glm2.wls --nii.gz sysname Linux hostname turing.rockefeller.edu machine x86_64 user smarvel FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 1 y /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh/c1_2/ces.nii logyflag 0 usedti 0 labelmask /Freiwald/lab_files/opt/freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0/subjects/fsaverage/label/lh.cortex.label maskinv 0 glmdir c1_2/my-glm2.wls IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory c1_2/my-glm2.wls Loading y from /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh/c1_2/ces.nii ... done reading. Saving design matrix to c1_2/my-glm2.wls/Xg.dat Computing normalized matrix Normalized matrix condition is 1 Matrix condition is 1 Found 149955 points in label. Found 149955 voxels in mask Saving mask to c1_2/my-glm2.wls/mask.nii.gz Reshaping mriglm->mask... search space = 74612.583051 DOF = 25 Starting fit and test Fit completed in 0.0106667 minutes Computing spatial AR1 on surface Residual: ar1mn=0.987809, ar1std=0.026903, gstd=3.653971, fwhm=8.604444 Writing results osgm maxvox sig=-6.33498 F=47.9826 at index 135220 0 0 seed=1499971484 mri_glmfit done
My question now is the part "Saving mask to c1_2/my-glm2.wls/mask.nii.gz Reshaping mriglm->mask..." Is it taking the mask I input and saving it to the new location? If so, the mask.nii.gz doesn't look like the mask I specified in the command, and it doesn't even have 26 subjects concatenated together. So instead of a 163842x1x1x26 mask it is a 163842x1x1 mask. When I look at my outputs, it seems as though it still isn't using the masks I specified. ________________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas N Greve greve@nmr.mgh.harvard.edu Sent: Wednesday, July 12, 2017 12:19 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
it might not be possible to use the frame mask with wls. Does it work without it?
On 07/09/2017 04:46 PM, Sadie Marvel wrote:
So now that I am using --no-prune, I'm getting this error: group wrfx error: MRInormWeights: value less than or eq to 0. I found another thread in the mailing list about this error saying that there may be values in cesvar.nii that are =0, and that pruning normally masks out these values. But here, I am trying to use my own masks to do so. Is it possible now that my masks and my --y input have different sizes? If so, how can I solve this?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Friday, July 7, 2017 11:57 AM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
Add --no-prune
On 7/7/17 11:38 AM, Sadie Marvel wrote:
When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Thursday, July 6, 2017 12:58 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Wednesday, July 5, 2017 7:28 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://urldefense.proofpoint.com/v2/url?u=https-3A__gate.nmr.mgh.harvard.ed... www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: https://urldefense.proofpoint.com/v2/url?u=ftp-3A__surfer.nmr.mgh.harvard.ed...
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
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 https://urldefense.proofpoint.com/v2/url?u=http-3A__www.partners.org_complia... . 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.
the command and terminal output look ok. You should compare it to the results without the frame mask (just use the regular mask). They should be the same in areas where the frame mask and regular mask completely agree.
On 07/12/2017 01:54 PM, Sadie Marvel wrote:
The command will run now without --wls, but I'm still not sure the output is correct. Here is my entire output: cwd /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh cmdline mri_glmfit.bin --y c1_2/ces.nii --osgm --no-prune --frame-mask masks.nii --surface fsaverage lh --glmdir c1_2/my-glm2.wls --nii.gz sysname Linux hostname turing.rockefeller.edu machine x86_64 user smarvel FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 1 y /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh/c1_2/ces.nii logyflag 0 usedti 0 labelmask /Freiwald/lab_files/opt/freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0/subjects/fsaverage/label/lh.cortex.label maskinv 0 glmdir c1_2/my-glm2.wls IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory c1_2/my-glm2.wls Loading y from /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh/c1_2/ces.nii ... done reading. Saving design matrix to c1_2/my-glm2.wls/Xg.dat Computing normalized matrix Normalized matrix condition is 1 Matrix condition is 1 Found 149955 points in label. Found 149955 voxels in mask Saving mask to c1_2/my-glm2.wls/mask.nii.gz Reshaping mriglm->mask... search space = 74612.583051 DOF = 25 Starting fit and test Fit completed in 0.0106667 minutes Computing spatial AR1 on surface Residual: ar1mn=0.987809, ar1std=0.026903, gstd=3.653971, fwhm=8.604444 Writing results osgm maxvox sig=-6.33498 F=47.9826 at index 135220 0 0 seed=1499971484 mri_glmfit done
My question now is the part "Saving mask to c1_2/my-glm2.wls/mask.nii.gz Reshaping mriglm->mask..." Is it taking the mask I input and saving it to the new location? If so, the mask.nii.gz doesn't look like the mask I specified in the command, and it doesn't even have 26 subjects concatenated together. So instead of a 163842x1x1x26 mask it is a 163842x1x1 mask. When I look at my outputs, it seems as though it still isn't using the masks I specified. ________________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas N Greve greve@nmr.mgh.harvard.edu Sent: Wednesday, July 12, 2017 12:19 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
it might not be possible to use the frame mask with wls. Does it work without it?
On 07/09/2017 04:46 PM, Sadie Marvel wrote:
So now that I am using --no-prune, I'm getting this error: group wrfx error: MRInormWeights: value less than or eq to 0. I found another thread in the mailing list about this error saying that there may be values in cesvar.nii that are =0, and that pruning normally masks out these values. But here, I am trying to use my own masks to do so. Is it possible now that my masks and my --y input have different sizes? If so, how can I solve this?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Friday, July 7, 2017 11:57 AM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
Add --no-prune
On 7/7/17 11:38 AM, Sadie Marvel wrote:
When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Thursday, July 6, 2017 12:58 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Wednesday, July 5, 2017 7:28 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://urldefense.proofpoint.com/v2/url?u=https-3A__gate.nmr.mgh.harvard.ed... www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: https://urldefense.proofpoint.com/v2/url?u=ftp-3A__surfer.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
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 https://urldefense.proofpoint.com/v2/url?u=http-3A__www.partners.org_complia... . 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
yes, I figured it out! Thanks for your help! ________________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas N Greve greve@nmr.mgh.harvard.edu Sent: Wednesday, July 12, 2017 6:16 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
the command and terminal output look ok. You should compare it to the results without the frame mask (just use the regular mask). They should be the same in areas where the frame mask and regular mask completely agree.
On 07/12/2017 01:54 PM, Sadie Marvel wrote:
The command will run now without --wls, but I'm still not sure the output is correct. Here is my entire output: cwd /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh cmdline mri_glmfit.bin --y c1_2/ces.nii --osgm --no-prune --frame-mask masks.nii --surface fsaverage lh --glmdir c1_2/my-glm2.wls --nii.gz sysname Linux hostname turing.rockefeller.edu machine x86_64 user smarvel FixVertexAreaFlag = 1 UseMaskWithSmoothing 1 OneSampleGroupMean 1 y /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh/c1_2/ces.nii logyflag 0 usedti 0 labelmask /Freiwald/lab_files/opt/freesurfer-Linux-centos6_x86_64-stable-pub-v6.0.0/subjects/fsaverage/label/lh.cortex.label maskinv 0 glmdir c1_2/my-glm2.wls IllCondOK 0 ReScaleX 1 DoFFx 0 Creating output directory c1_2/my-glm2.wls Loading y from /Freiwald/smarvel/cooked/mygroup4/a_FOB.sm05.lh/c1_2/ces.nii ... done reading. Saving design matrix to c1_2/my-glm2.wls/Xg.dat Computing normalized matrix Normalized matrix condition is 1 Matrix condition is 1 Found 149955 points in label. Found 149955 voxels in mask Saving mask to c1_2/my-glm2.wls/mask.nii.gz Reshaping mriglm->mask... search space = 74612.583051 DOF = 25 Starting fit and test Fit completed in 0.0106667 minutes Computing spatial AR1 on surface Residual: ar1mn=0.987809, ar1std=0.026903, gstd=3.653971, fwhm=8.604444 Writing results osgm maxvox sig=-6.33498 F=47.9826 at index 135220 0 0 seed=1499971484 mri_glmfit done
My question now is the part "Saving mask to c1_2/my-glm2.wls/mask.nii.gz Reshaping mriglm->mask..." Is it taking the mask I input and saving it to the new location? If so, the mask.nii.gz doesn't look like the mask I specified in the command, and it doesn't even have 26 subjects concatenated together. So instead of a 163842x1x1x26 mask it is a 163842x1x1 mask. When I look at my outputs, it seems as though it still isn't using the masks I specified. ________________________________________ From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas N Greve greve@nmr.mgh.harvard.edu Sent: Wednesday, July 12, 2017 12:19 PM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
it might not be possible to use the frame mask with wls. Does it work without it?
On 07/09/2017 04:46 PM, Sadie Marvel wrote:
So now that I am using --no-prune, I'm getting this error: group wrfx error: MRInormWeights: value less than or eq to 0. I found another thread in the mailing list about this error saying that there may be values in cesvar.nii that are =0, and that pruning normally masks out these values. But here, I am trying to use my own masks to do so. Is it possible now that my masks and my --y input have different sizes? If so, how can I solve this?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Friday, July 7, 2017 11:57 AM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
Add --no-prune
On 7/7/17 11:38 AM, Sadie Marvel wrote:
When I use a script like this,
mri_glmfit --y ces.nii --wls cesvar.nii --osgm --frame-mask masks.nii --surface fsaverage rh --glmdir my-glm2.wls --nii.gz
using the individual binary masks for all the subjects, the command seems to revert to using mask.nii, the single mask that was averaged over subjects. The output then just looks like it was using the single mask instead of being subject specific. Any idea why?
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Thursday, July 6, 2017 12:58 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data {Disarmed}
yes
On 7/6/17 12:50 PM, Sadie Marvel wrote:
From using isxconcat-sess to concatenate all of my subjects together, the command creates binary masks called masks.nii and mask.nii, masks.nii being all of the binary masks for each subject. Is this what I would use with --frame-mask?
Thanks,
Sadie
*From:* freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu on behalf of Douglas Greve greve@nmr.mgh.harvard.edu *Sent:* Wednesday, July 5, 2017 7:28 PM *To:* freesurfer@nmr.mgh.harvard.edu *Subject:* Re: [Freesurfer] Subject with missing fMRI data
We don't have any tools to do imputation. It is possible to have a subject specific mask with the --frame-mask option to mri_glmfit. The frame mask must be the same size as the input (--y) with 1s and 0s to indicate whether a given voxel for a given subject should be included in the model. I've not been happy with this method as it can create some strange results as different voxels will now have different DOFs.
On 6/29/17 8:49 AM, Sadie Marvel wrote:
Hi List,
I am currently doing a group analysis of 26 subjects, one of which has missing fMRI data where the signal cuts off through parts of the occipital and temporal lobes. I was wondering if freesurfer has any way of dealing with this, such as multiple imputation, or other methods of filling the missing data. Alternatively, is there a way to only use the portions of the brain that have data when doing mri_glmfit? Does freesurfer do this automatically?
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu *MailScanner has detected a possible fraud attempt from "urldefense.proofpoint.com" claiming to be* https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://urldefense.proofpoint.com/v2/url?u=https-3A__gate.nmr.mgh.harvard.ed... www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: https://urldefense.proofpoint.com/v2/url?u=ftp-3A__surfer.nmr.mgh.harvard.ed...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
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 https://urldefense.proofpoint.com/v2/url?u=http-3A__www.partners.org_complia... . 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://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
-- Douglas N. Greve, Ph.D. MGH-NMR Center greve@nmr.mgh.harvard.edu Phone Number: 617-724-2358 Fax: 617-726-7422
Bugs: surfer.nmr.mgh.harvard.edu/fswiki/BugReporting FileDrop: https://urldefense.proofpoint.com/v2/url?u=https-3A__gate.nmr.mgh.harvard.ed... www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: https://urldefense.proofpoint.com/v2/url?u=ftp-3A__surfer.nmr.mgh.harvard.ed...
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://urldefense.proofpoint.com/v2/url?u=https-3A__mail.nmr.mgh.harvard.ed...
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 https://urldefense.proofpoint.com/v2/url?u=http-3A__www.partners.org_complia... . 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@nmr.mgh.harvard.edu
-
Douglas Greve -
Douglas N Greve
-
Sadie Marvel