Hi Doug,
Thanks for confirming how to calculate the 95% confidence interval of the slope for all my clusters. I realized that there was a small error in my formula, in that I should use the critical values from the t-distribution, not the z-distribution (so the SE multiplier will be more like 2.0 instead of 1.96, depending on my sample N).
The problem is that mri_segstats trims the output at the 10,000ths place. Here is the cluster.summary and the segstats from the two clusters identified as significantly associated with age.
mc-z.th1.3.neg.sig.cluster.summary: # ClusterNo Max VtxMax Size(mm^2) MNIX MNIY MNIZ CWP CWPLow CWPHi NVtxs Annot 1 -5.096 73213 4275.24 50.0 33.2 -7.2 0.00350 0.00280 0.00430 8455 parstriangularis 2 -3.802 51616 20825.27 4.9 -26.2 39.2 0.00010 0.00000 0.00020 41094 posteriorcingulate
gamma.mgh: # ColHeaders Index SegId NVoxels Volume_mm3 StructName Mean StdDev Min Max Range 1 1 8455 8455.0 Seg0001 -0.0089 0.0027 -0.0147 -0.0038 0.0109 2 2 41094 41094.0 Seg0002 -0.0069 0.0017 -0.0132 -0.0023 0.0108
gammavar.mgh: # ColHeaders Index SegId NVoxels Volume_mm3 StructName Mean StdDev Min Max Range 1 1 8455 8455.0 Seg0001 0.0000 0.0000 0.0000 0.0000 0.0000 2 2 41094 41094.0 Seg0002 0.0000 0.0000 0.0000 0.0000 0.0000
So, for the 95% CI of the slope for cluster #1, I would use these formulas (N=60, so the t-distribution 95% critical value is 2.0003): -0.0089+2.0003*sqrt(0.0000) -0.0089-2.0003*sqrt(0.0000)
Is there any way to add more significant digits to the mri_segstats output?
Thanks, -Kayle
On Nov 4, 2013, at 6:31 PM, Douglas N Greve greve@nmr.mgh.harvard.edu wrote:
Yes, that should work. doug
On 10/30/2013 03:35 PM, Kayle Sawyer wrote:
Hi all,
I would like to calculate the 95% confidence interval of the slope for a thickness cluster (vs age in years) produced by mri_glmfit-sim.
I have obtained values for my clusters from gamma.mgh and gammavar.mgh using these two commands:
mri_segstats --i gamma.mgh --seg mc-z.th1.3.neg.sig.ocn.mgh --excludeid 0 --sum mc-z.th1.3.neg.sig.ocn.gamma.segstats
mri_segstats --i gammavar.mgh --seg mc-z.th1.3.neg.sig.ocn.mgh --excludeid 0 --sum mc-z.th1.3.neg.sig.ocn.gammavar.segstats
Next, (for each cluster, within each segstats summary file) I would take the “mean” from gamma, and the “mean” from gammavar, and calculate the 95% CI thus: gamma + 1.96*sqrt(gammavar) gamma - 1.96*sqrt(gammavar)
Would these two values would give me the 95% confidence interval for the effect of age on that cluster, in mm/year?
Thanks, -Kayle
Yahoo Groups Links
<*> To visit your group on the web, go to: http://groups.yahoo.com/group/martinos-tech/
<*> Your email settings: Individual Email | Traditional
<*> To change settings online go to: http://groups.yahoo.com/group/martinos-tech/join (Yahoo! ID required)
<*> To change settings via email: martinos-tech-digest@yahoogroups.com martinos-tech-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to: martinos-tech-unsubscribe@yahoogroups.com
<*> Your use of Yahoo Groups is subject to: http://info.yahoo.com/legal/us/yahoo/utos/terms/
-- 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://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
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.
You can add "--mul 1000" to multiply everything by 1000 which will effectively give you more significant digits. doug
On 12/03/2013 03:36 PM, Kayle Sawyer wrote:
Hi Doug,
Thanks for confirming how to calculate the 95% confidence interval of the slope for all my clusters. I realized that there was a small error in my formula, in that I should use the critical values from the t-distribution, not the z-distribution (so the SE multiplier will be more like 2.0 instead of 1.96, depending on my sample N).
The problem is that mri_segstats trims the output at the 10,000ths place. Here is the cluster.summary and the segstats from the two clusters identified as significantly associated with age.
mc-z.th1.3.neg.sig.cluster.summary: # ClusterNo Max VtxMax Size(mm^2) MNIX MNIY MNIZ CWP CWPLow CWPHi NVtxs Annot 1 -5.096 73213 4275.24 50.0 33.2 -7.2 0.00350 0.00280 0.00430 8455 parstriangularis 2 -3.802 51616 20825.27 4.9 -26.2 39.2 0.00010 0.00000 0.00020 41094 posteriorcingulate
gamma.mgh: # ColHeaders Index SegId NVoxels Volume_mm3 StructName Mean StdDev Min Max Range 1 1 8455 8455.0 Seg0001 -0.0089 0.0027 -0.0147 -0.0038 0.0109 2 2 41094 41094.0 Seg0002 -0.0069 0.0017 -0.0132 -0.0023 0.0108
gammavar.mgh: # ColHeaders Index SegId NVoxels Volume_mm3 StructName Mean StdDev Min Max Range 1 1 8455 8455.0 Seg0001 0.0000 0.0000 0.0000 0.0000 0.0000 2 2 41094 41094.0 Seg0002 0.0000 0.0000 0.0000 0.0000 0.0000
So, for the 95% CI of the slope for cluster #1, I would use these formulas (N=60, so the t-distribution 95% critical value is 2.0003): -0.0089+2.0003*sqrt(0.0000) -0.0089-2.0003*sqrt(0.0000)
Is there any way to add more significant digits to the mri_segstats output?
Thanks, -Kayle
On Nov 4, 2013, at 6:31 PM, Douglas N Greve greve@nmr.mgh.harvard.edu wrote:
Yes, that should work. doug
On 10/30/2013 03:35 PM, Kayle Sawyer wrote:
Hi all,
I would like to calculate the 95% confidence interval of the slope for a thickness cluster (vs age in years) produced by mri_glmfit-sim.
I have obtained values for my clusters from gamma.mgh and gammavar.mgh using these two commands:
mri_segstats --i gamma.mgh --seg mc-z.th1.3.neg.sig.ocn.mgh --excludeid 0 --sum mc-z.th1.3.neg.sig.ocn.gamma.segstats
mri_segstats --i gammavar.mgh --seg mc-z.th1.3.neg.sig.ocn.mgh --excludeid 0 --sum mc-z.th1.3.neg.sig.ocn.gammavar.segstats
Next, (for each cluster, within each segstats summary file) I would take the “mean” from gamma, and the “mean” from gammavar, and calculate the 95% CI thus: gamma + 1.96*sqrt(gammavar) gamma - 1.96*sqrt(gammavar)
Would these two values would give me the 95% confidence interval for the effect of age on that cluster, in mm/year?
Thanks, -Kayle
Yahoo Groups Links
<*> To visit your group on the web, go to: http://groups.yahoo.com/group/martinos-tech/
<*> Your email settings: Individual Email | Traditional
<*> To change settings online go to: http://groups.yahoo.com/group/martinos-tech/join (Yahoo! ID required)
<*> To change settings via email: martinos-tech-digest@yahoogroups.com martinos-tech-fullfeatured@yahoogroups.com
<*> To unsubscribe from this group, send an email to: martinos-tech-unsubscribe@yahoogroups.com
<*> Your use of Yahoo Groups is subject to: http://info.yahoo.com/legal/us/yahoo/utos/terms/
-- 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://gate.nmr.mgh.harvard.edu/filedrop2 www.nmr.mgh.harvard.edu/facility/filedrop/index.html Outgoing: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/
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
freesurfer@nmr.mgh.harvard.edu
-
Douglas N Greve -
Kayle Sawyer