External Email - Use Caution
Hi. I have a question about cluster size determination within Freesurfer.
We are running linear mixed model on our Freesurfer generated volume data over the left and right cortex parcellations, and would like to look at cluster size at the p < 0.01 cluster-forming threshold level.
1) After running the models, we have created a .mgh image of the F-stats results, separately for the negative and positive results, saving them with fs_write_Y.
2) We then run these data through mri_surfcluster, which reports the maximum cluster size of 216 voxels. This was run with the command:
a. mri_surfcluster --hemi lh --in F_lhstats.mgh --o F_lhstats _negCluster.mgh --subject fsaverage --thmin 6.8 --sign pos
b. The "sign" is set to positive because this is running on F-numbers.
3) Next, we run mri_segstats on the output of the last command to get the location and sizes of the individual clusters, as follows:
a. mri_segstats --seg F_lhstats _negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
The problem is that this is the output of that file (F_lhstats_negCluster.txt):
# Title Segmentation Statistics # # generating_program mri_segstats # cvs_version $Id: mri_segstats.c,v 1.121 2016/05/31 17:27:11 greve Exp $ # cmdline mri_segstats --seg F_lhstats_negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt # sysname Linux # hostname loginnode2 # machine x86_64 # user mwagshul # anatomy_type volume # # SUBJECTS_DIR /home/mrrc-all/MRE_LAB/Holtzer/QdecAnalysis # subjectname fsaverage # SegVolFile F_lhstats_negCluster.mgh # SegVolFileTimeStamp 2019/07/18 11:21:52 # Annot fsaverage lh parc # Label fsaverage lh --seg # ExcludeSegId # Only reporting non-empty segmentations # VoxelVolume_mm3 1 # TableCol 1 ColHeader Index # TableCol 1 FieldName Index # TableCol 1 Units NA # TableCol 2 ColHeader SegId # TableCol 2 FieldName Segmentation Id # TableCol 2 Units NA # TableCol 3 ColHeader NVoxels # TableCol 3 FieldName Number of Voxels # TableCol 3 Units unitless # TableCol 4 ColHeader Volume_mm3 # TableCol 4 FieldName Volume # TableCol 4 Units mm^3 # TableCol 5 ColHeader StructName # TableCol 5 FieldName Structure Name # TableCol 5 Units NA # NRows 8 # NTableCols 5 # ColHeaders Index SegId NVoxels Volume_mm3 StructName 1 0 161454 161454.0 Seg0000 2 6 524 524.0 Seg0006 3 7 897 897.0 Seg0007 4 8 413 413.0 Seg0008 5 9 311 311.0 Seg0009 6 10 165 165.0 Seg0010 7 11 61 61.0 Seg0011 8 12 17 17.0 Seg0012
This clearly is reporting clusters larger than the 216 voxels reported by mri_surfcluster. Hope someone can help us figure out the discrepancy here, thanks!
(As an aside, what are we doing wrong with the mri_segstats command which is not printing out the actual anatomical locations for each cluster?).
Mark
____________________ Mark Wagshul, PhD Associate Professor Gruss Magnetic Resonance Research Center Albert Einstein College of Medicine Bronx, NY 10461
Ph: 718-430-4011 FAX: 718-430-3399 Email: mark.wagshul@einstein.yu.edumailto:mark.wagshul@einstein.yu.edu
[einstein-logo-rgb] This email message and any accompanying attachments may contain privileged information intended only for the named recipient(s). If you are not the intended recipient(s), you are hereby notified that the dissemination, distribution, and or copying of this message is strictly prohibited. If you receive this message in error, or are not the named recipient(s), please notify the sender at the email address above, delete this email from your computer, and destroy any copies in any form immediately.
When you run mri_surfcluster, save an output cluster number (--ocn) output, then feed that into mri_segstats.
On 7/18/19 11:44 AM, Mark Wagshul wrote:
External Email - Use Caution
Hi. I have a question about cluster size determination within Freesurfer.
We are running linear mixed model on our Freesurfer generated volume data over the left and right cortex parcellations, and would like to look at cluster size at the p < 0.01 cluster-forming threshold level.
1)After running the models, we have created a .mgh image of the F-stats results, separately for the negative and positive results, saving them with fs_write_Y.
2)We then run these data through mri_surfcluster, which reports the maximum cluster size of 216 voxels. This was run with the command:
a.mri_surfcluster --hemi lh --in F_lhstats.mgh --o F_lhstats _negCluster.mgh --subject fsaverage --thmin 6.8 --sign pos
b.The “sign” is set to positive because this is running on F-numbers.
3)Next, we run mri_segstats on the output of the last command to get the location and sizes of the individual clusters, as follows:
a.mri_segstats --seg F_lhstats _negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
The problem is that this is the output of that file (F_lhstats_negCluster.txt):
# Title Segmentation Statistics
#
# generating_program mri_segstats
# cvs_version $Id: mri_segstats.c,v 1.121 2016/05/31 17:27:11 greve Exp $
# cmdline mri_segstats --seg F_lhstats_negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
# sysname Linux
# hostname loginnode2
# machine x86_64
# user mwagshul
# anatomy_type volume
#
# SUBJECTS_DIR /home/mrrc-all/MRE_LAB/Holtzer/QdecAnalysis
# subjectname fsaverage
# SegVolFile F_lhstats_negCluster.mgh
# SegVolFileTimeStamp 2019/07/18 11:21:52
# Annot fsaverage lh parc
# Label fsaverage lh --seg
# ExcludeSegId
# Only reporting non-empty segmentations
# VoxelVolume_mm3 1
# TableCol 1 ColHeader Index
# TableCol 1 FieldName Index
# TableCol 1 Units NA
# TableCol 2 ColHeader SegId
# TableCol 2 FieldName Segmentation Id
# TableCol 2 Units NA
# TableCol 3 ColHeader NVoxels
# TableCol 3 FieldName Number of Voxels
# TableCol 3 Units unitless
# TableCol 4 ColHeader Volume_mm3
# TableCol 4 FieldName Volume
# TableCol 4 Units mm^3
# TableCol 5 ColHeader StructName
# TableCol 5 FieldName Structure Name
# TableCol 5 Units NA
# NRows 8
# NTableCols 5
# ColHeaders Index SegId NVoxels Volume_mm3 StructName
1 0 161454 161454.0 Seg0000
*2 6 524 524.0 Seg0006*
* 3 7 897 897.0 Seg0007*
* 4 8 413 413.0 Seg0008*
* 5 9 311 311.0 Seg0009*
* 6 10 165 165.0 Seg0010*
* 7 11 61 61.0 Seg0011*
* 8 12 17 17.0 Seg0012*
This clearly is reporting clusters larger than the 216 voxels reported by mri_surfcluster. Hope someone can help us figure out the discrepancy here, thanks!
(As an aside, what are we doing wrong with the mri_segstats command which is not printing out the actual anatomical locations for each cluster?).
Mark
Mark Wagshul, PhD Associate Professor Gruss Magnetic Resonance Research Center Albert Einstein College of Medicine Bronx, NY 10461
Ph: 718-430-4011
FAX: 718-430-3399
Email: mark.wagshul@einstein.yu.edu mailto:mark.wagshul@einstein.yu.edu
einstein-logo-rgb
This email message and any accompanying attachments may contain privileged information intended only for the named recipient(s). If you are not the intended recipient(s), you are hereby notified that the dissemination, distribution, and or copying of this message is strictly prohibited. If you receive this message in error, or are not the named recipient(s), please notify the sender at the email address above, delete this email from your computer, and destroy any copies in any form immediately.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
External Email - Use Caution
Thanks!
Mark
____________________ Mark Wagshul, PhD Associate Professor Gruss Magnetic Resonance Research Center Albert Einstein College of Medicine Bronx, NY 10461
Ph: 718-430-4011 FAX: 718-430-3399 Email: mark.wagshul@einstein.yu.edu
This email message and any accompanying attachments may contain privileged information intended only for the named recipient(s). If you are not the intended recipient(s), you are hereby notified that the dissemination, distribution, and or copying of this message is strictly prohibited. If you receive this message in error, or are not the named recipient(s), please notify the sender at the email address above, delete this email from your computer, and destroy any copies in any form immediately.
-----Original Message----- From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu On Behalf Of Greve, Douglas N.,Ph.D. Sent: Thursday, July 18, 2019 11:49 AM To: freesurfer@nmr.mgh.harvard.edu Subject: Re: [Freesurfer] Cluster sizes discrepancy for mri_surfcluster vs mri_segstats
When you run mri_surfcluster, save an output cluster number (--ocn) output, then feed that into mri_segstats.
On 7/18/19 11:44 AM, Mark Wagshul wrote:
External Email - Use Caution
Hi. I have a question about cluster size determination within Freesurfer.
We are running linear mixed model on our Freesurfer generated volume data over the left and right cortex parcellations, and would like to look at cluster size at the p < 0.01 cluster-forming threshold level.
1)After running the models, we have created a .mgh image of the F-stats results, separately for the negative and positive results, saving them with fs_write_Y.
2)We then run these data through mri_surfcluster, which reports the maximum cluster size of 216 voxels. This was run with the command:
a.mri_surfcluster --hemi lh --in F_lhstats.mgh --o F_lhstats _negCluster.mgh --subject fsaverage --thmin 6.8 --sign pos
b.The "sign" is set to positive because this is running on F-numbers.
3)Next, we run mri_segstats on the output of the last command to get the location and sizes of the individual clusters, as follows:
a.mri_segstats --seg F_lhstats _negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
The problem is that this is the output of that file (F_lhstats_negCluster.txt):
# Title Segmentation Statistics
#
# generating_program mri_segstats
# cvs_version $Id: mri_segstats.c,v 1.121 2016/05/31 17:27:11 greve Exp $
# cmdline mri_segstats --seg F_lhstats_negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
# sysname Linux
# hostname loginnode2
# machine x86_64
# user mwagshul
# anatomy_type volume
#
# SUBJECTS_DIR /home/mrrc-all/MRE_LAB/Holtzer/QdecAnalysis
# subjectname fsaverage
# SegVolFile F_lhstats_negCluster.mgh
# SegVolFileTimeStamp 2019/07/18 11:21:52
# Annot fsaverage lh parc
# Label fsaverage lh --seg
# ExcludeSegId
# Only reporting non-empty segmentations
# VoxelVolume_mm3 1
# TableCol 1 ColHeader Index
# TableCol 1 FieldName Index
# TableCol 1 Units NA
# TableCol 2 ColHeader SegId
# TableCol 2 FieldName Segmentation Id
# TableCol 2 Units NA
# TableCol 3 ColHeader NVoxels
# TableCol 3 FieldName Number of Voxels
# TableCol 3 Units unitless
# TableCol 4 ColHeader Volume_mm3
# TableCol 4 FieldName Volume
# TableCol 4 Units mm^3
# TableCol 5 ColHeader StructName
# TableCol 5 FieldName Structure Name
# TableCol 5 Units NA
# NRows 8
# NTableCols 5
# ColHeaders Index SegId NVoxels Volume_mm3 StructName
1 0 161454 161454.0 Seg0000
*2 6 524 524.0 Seg0006*
* 3 7 897 897.0 Seg0007*
* 4 8 413 413.0 Seg0008*
* 5 9 311 311.0 Seg0009*
* 6 10 165 165.0 Seg0010*
* 7 11 61 61.0 Seg0011*
* 8 12 17 17.0 Seg0012*
This clearly is reporting clusters larger than the 216 voxels reported by mri_surfcluster. Hope someone can help us figure out the discrepancy here, thanks!
(As an aside, what are we doing wrong with the mri_segstats command which is not printing out the actual anatomical locations for each cluster?).
Mark
Mark Wagshul, PhD Associate Professor Gruss Magnetic Resonance Research Center Albert Einstein College of Medicine Bronx, NY 10461
Ph: 718-430-4011
FAX: 718-430-3399
Email: mark.wagshul@einstein.yu.edu mailto:mark.wagshul@einstein.yu.edu
einstein-logo-rgb
This email message and any accompanying attachments may contain privileged information intended only for the named recipient(s). If you are not the intended recipient(s), you are hereby notified that the dissemination, distribution, and or copying of this message is strictly prohibited. If you receive this message in error, or are not the named recipient(s), please notify the sender at the email address above, delete this email from your computer, and destroy any copies in any form immediately.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail .nmr.mgh.harvard.edu%2Fmailman%2Flistinfo%2Ffreesurfer&data=02%7C0 1%7Cmark.wagshul%40einstein.yu.edu%7C43bea5c208a74700fc9c08d70b9798e6% 7C04c70eb48f2648079934e02e89266ad0%7C1%7C0%7C636990618048314705&sd ata=KpaZ4i6TVCxsqvXR8cQEQVZ7qtKJ6gUmQWCFMcs9VJ8%3D&reserved=0
_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.m...
External Email - Use Caution
Doug,
We are still seeing differences, but I notice that the results from segstats are all whole integers, while the results from surfcluster (Max cluster size) is not. I assume this is because surfcluster is using partial volume. How do we implement this in segstats?
Thanks,
Mark _______________________ Mark Wagshul, PhD Albert Einstein College of Medicine Bronx, NY
Sent from my iPhone
On Jul 18, 2019, at 11:50 AM, Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edu wrote:
When you run mri_surfcluster, save an output cluster number (--ocn) output, then feed that into mri_segstats.
On 7/18/19 11:44 AM, Mark Wagshul wrote:
External Email - Use CautionHi. I have a question about cluster size determination within Freesurfer.
We are running linear mixed model on our Freesurfer generated volume data over the left and right cortex parcellations, and would like to look at cluster size at the p < 0.01 cluster-forming threshold level.
1)After running the models, we have created a .mgh image of the F-stats results, separately for the negative and positive results, saving them with fs_write_Y.
2)We then run these data through mri_surfcluster, which reports the maximum cluster size of 216 voxels. This was run with the command:
a.mri_surfcluster --hemi lh --in F_lhstats.mgh --o F_lhstats _negCluster.mgh --subject fsaverage --thmin 6.8 --sign pos
b.The “sign” is set to positive because this is running on F-numbers.
3)Next, we run mri_segstats on the output of the last command to get the location and sizes of the individual clusters, as follows:
a.mri_segstats --seg F_lhstats _negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
The problem is that this is the output of that file (F_lhstats_negCluster.txt):
# Title Segmentation Statistics
#
# generating_program mri_segstats
# cvs_version $Id: mri_segstats.c,v 1.121 2016/05/31 17:27:11 greve Exp $
# cmdline mri_segstats --seg F_lhstats_negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
# sysname Linux
# hostname loginnode2
# machine x86_64
# user mwagshul
# anatomy_type volume
#
# SUBJECTS_DIR /home/mrrc-all/MRE_LAB/Holtzer/QdecAnalysis
# subjectname fsaverage
# SegVolFile F_lhstats_negCluster.mgh
# SegVolFileTimeStamp 2019/07/18 11:21:52
# Annot fsaverage lh parc
# Label fsaverage lh --seg
# ExcludeSegId
# Only reporting non-empty segmentations
# VoxelVolume_mm3 1
# TableCol 1 ColHeader Index
# TableCol 1 FieldName Index
# TableCol 1 Units NA
# TableCol 2 ColHeader SegId
# TableCol 2 FieldName Segmentation Id
# TableCol 2 Units NA
# TableCol 3 ColHeader NVoxels
# TableCol 3 FieldName Number of Voxels
# TableCol 3 Units unitless
# TableCol 4 ColHeader Volume_mm3
# TableCol 4 FieldName Volume
# TableCol 4 Units mm^3
# TableCol 5 ColHeader StructName
# TableCol 5 FieldName Structure Name
# TableCol 5 Units NA
# NRows 8
# NTableCols 5
# ColHeaders Index SegId NVoxels Volume_mm3 StructName
1 0 161454 161454.0 Seg0000
*2 6 524 524.0 Seg0006*
3 7 897 897.0 Seg0007*
4 8 413 413.0 Seg0008*
5 9 311 311.0 Seg0009*
6 10 165 165.0 Seg0010*
7 11 61 61.0 Seg0011*
8 12 17 17.0 Seg0012*
This clearly is reporting clusters larger than the 216 voxels reported by mri_surfcluster. Hope someone can help us figure out the discrepancy here, thanks!
(As an aside, what are we doing wrong with the mri_segstats command which is not printing out the actual anatomical locations for each cluster?).
Mark
Mark Wagshul, PhD Associate Professor Gruss Magnetic Resonance Research Center Albert Einstein College of Medicine Bronx, NY 10461
Ph: 718-430-4011
FAX: 718-430-3399
Email: mark.wagshul@einstein.yu.edu mailto:mark.wagshul@einstein.yu.edu
einstein-logo-rgb
This email message and any accompanying attachments may contain privileged information intended only for the named recipient(s). If you are not the intended recipient(s), you are hereby notified that the dissemination, distribution, and or copying of this message is strictly prohibited. If you receive this message in error, or are not the named recipient(s), please notify the sender at the email address above, delete this email from your computer, and destroy any copies in any form immediately.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.m...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.m...
what is your mri_segstats command? If you want area in mm2, you'll need to pass an area file to mri_segstats as input (--i) and use --accumulate. If this is on fsaverage, you can use $SUBJECTS_DIR/fsaverage/surf/lh.white.avg.area.mgh
On 7/24/19 7:55 AM, Mark Wagshul wrote:
External Email - Use CautionDoug,
We are still seeing differences, but I notice that the results from segstats are all whole integers, while the results from surfcluster (Max cluster size) is not. I assume this is because surfcluster is using partial volume. How do we implement this in segstats?
Thanks,
Mark _______________________ Mark Wagshul, PhD Albert Einstein College of Medicine Bronx, NY
Sent from my iPhone
On Jul 18, 2019, at 11:50 AM, Greve, Douglas N.,Ph.D. DGREVE@mgh.harvard.edu wrote:
When you run mri_surfcluster, save an output cluster number (--ocn) output, then feed that into mri_segstats.
On 7/18/19 11:44 AM, Mark Wagshul wrote:
External Email - Use CautionHi. I have a question about cluster size determination within Freesurfer.
We are running linear mixed model on our Freesurfer generated volume data over the left and right cortex parcellations, and would like to look at cluster size at the p < 0.01 cluster-forming threshold level.
1)After running the models, we have created a .mgh image of the F-stats results, separately for the negative and positive results, saving them with fs_write_Y.
2)We then run these data through mri_surfcluster, which reports the maximum cluster size of 216 voxels. This was run with the command:
a.mri_surfcluster --hemi lh --in F_lhstats.mgh --o F_lhstats _negCluster.mgh --subject fsaverage --thmin 6.8 --sign pos
b.The “sign” is set to positive because this is running on F-numbers.
3)Next, we run mri_segstats on the output of the last command to get the location and sizes of the individual clusters, as follows:
a.mri_segstats --seg F_lhstats _negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
The problem is that this is the output of that file (F_lhstats_negCluster.txt):
# Title Segmentation Statistics
#
# generating_program mri_segstats
# cvs_version $Id: mri_segstats.c,v 1.121 2016/05/31 17:27:11 greve Exp $
# cmdline mri_segstats --seg F_lhstats_negCluster.mgh --annot lh.aparc.annot lh parc --slabel fsaverage lh --seg --sum F_lhstats_negCluster.txt
# sysname Linux
# hostname loginnode2
# machine x86_64
# user mwagshul
# anatomy_type volume
#
# SUBJECTS_DIR /home/mrrc-all/MRE_LAB/Holtzer/QdecAnalysis
# subjectname fsaverage
# SegVolFile F_lhstats_negCluster.mgh
# SegVolFileTimeStamp 2019/07/18 11:21:52
# Annot fsaverage lh parc
# Label fsaverage lh --seg
# ExcludeSegId
# Only reporting non-empty segmentations
# VoxelVolume_mm3 1
# TableCol 1 ColHeader Index
# TableCol 1 FieldName Index
# TableCol 1 Units NA
# TableCol 2 ColHeader SegId
# TableCol 2 FieldName Segmentation Id
# TableCol 2 Units NA
# TableCol 3 ColHeader NVoxels
# TableCol 3 FieldName Number of Voxels
# TableCol 3 Units unitless
# TableCol 4 ColHeader Volume_mm3
# TableCol 4 FieldName Volume
# TableCol 4 Units mm^3
# TableCol 5 ColHeader StructName
# TableCol 5 FieldName Structure Name
# TableCol 5 Units NA
# NRows 8
# NTableCols 5
# ColHeaders Index SegId NVoxels Volume_mm3 StructName
1 0 161454 161454.0 Seg0000
*2 6 524 524.0 Seg0006*
3 7 897 897.0 Seg0007*
4 8 413 413.0 Seg0008*
5 9 311 311.0 Seg0009*
6 10 165 165.0 Seg0010*
7 11 61 61.0 Seg0011*
8 12 17 17.0 Seg0012*
This clearly is reporting clusters larger than the 216 voxels reported by mri_surfcluster. Hope someone can help us figure out the discrepancy here, thanks!
(As an aside, what are we doing wrong with the mri_segstats command which is not printing out the actual anatomical locations for each cluster?).
Mark
Mark Wagshul, PhD Associate Professor Gruss Magnetic Resonance Research Center Albert Einstein College of Medicine Bronx, NY 10461
Ph: 718-430-4011
FAX: 718-430-3399
Email: mark.wagshul@einstein.yu.edu mailto:mark.wagshul@einstein.yu.edu
einstein-logo-rgb
This email message and any accompanying attachments may contain privileged information intended only for the named recipient(s). If you are not the intended recipient(s), you are hereby notified that the dissemination, distribution, and or copying of this message is strictly prohibited. If you receive this message in error, or are not the named recipient(s), please notify the sender at the email address above, delete this email from your computer, and destroy any copies in any form immediately.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.m...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail.nmr.m...
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Greve, Douglas N.,Ph.D. -
Mark Wagshul