Hi,
We have been using your exellent software for some time now. We work on several studies and in one of them we now asess the hippocampus volume. The information is gathered from the Freesurfer aseg.stats file. The data for this group has been run in both versions 3.0 and 4.0. When comparing results from these runs we find discrepant results.
We have four questions:
1) I assume that some degree of difference in outcome between the two FS versions is to be expected. But how large a difference would you consider "normal" or acceptable? Our differences seem to be quite large. I have attached the outputs for lh. for two subjects with extensions fs3 (Freesurfer 3.0) and fs4 (Freesurfer 4.0).
2) I have inspected the subjects with the largest discrepancies and so far it seems that it is v4.0 that is producing the errors. The tiffs attached showing where I believe there are obvious errors (the fs4´s).
3) Is it possible to edit aseg results? (or is that only for the cortical reconstructions?)
4) Can you give any advise on which of the data sets we should do calculations on?
Have also attached a Excel-file with the volumes for all the subjects in the study (average and STD on the bottom...). If this large variations is to be expected, small changes in subject groups (about 3-5 per cent) could be missed.
Thank you !
Petr
.Bold { font-weight: bold; } .Title { font-weight: bold; font-size: 18px; color: #cc3300; } .Code { border: #8b4513 1px solid; padding-right: 5px; padding-left: 5px;color: #000066; font-family: 'Courier New' , Monospace;background-color: #ff9933; } My recent paper could be help for your question.Neuroradiology, 2008 Validation of hippocampal volumes measured using a manual method and two automated methods (FreeSurfer and IBASPM) in chronic major depressive disorder.http://www.ncbi.nlm.nih.gov/pubmed/18414838?ordinalpos=1&itool=EntrezSys..., Tae Ph.DChucheon, KoreaBest regards!--- Original Message ---From : p.s.bjerkan@studmed.uio.noTo : freesurfer@nmr.mgh.harvard.eduDate : 2008/05/16 �ݿ��� ���� 1:55:05Subject : [Freesurfer] Hippocampal volumesHi, We have been using your exellent software for some time now. We work on several studies and in one of them we now asess the hippocampus volume. The information is gathered from the Freesurfer aseg.stats file. The data for this group has been run in both versions 3.0 and 4.0. When comparing results from these runs we find discrepant results. We have four questions: 1) I assume that some degree of difference in outcome between the two FS versions is to be expected. But how large a difference would you consider "normal" or acceptable? Our differences seem to be quite large. I have attached the outputs for lh. for two subjects with extensions fs3 (Freesurfer 3.0) and fs4 (Freesurfer 4.0). 2) I have inspected the subjects with the largest discrepancies and so far it seems that it is v4.0 that is producing the errors. The tiffs attached showing where I believe there are obvious errors (the fs4?s). 3) Is it possible to edit aseg results? (or is that only for the cortical reconstructions?) 4) Can you give any advise on which of the data sets we should do calculations on? Have also attached a Excel-file with the volumes for all the subjects in the study (average and STD on the bottom...). If this large variations is to be expected, small changes in subject groups (about 3-5 per cent) could be missed. Thank you ! Petr_______________________________________________ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Hi Petr,
can you try the latest release? I think there was a bug in 4.0 that caused this kind of thing which hopefully is now fixed. You can certainly edit the asegs if you want (in tkmedit), but hopefully if you rerun it will resolve these issues.
cheers, Bruce On Thu, 15 May 2008 p.s.bjerkan@studmed.uio.no wrote:
Hi,
We have been using your exellent software for some time now. We work on several studies and in one of them we now asess the hippocampus volume. The information is gathered from the Freesurfer aseg.stats file. The data for this group has been run in both versions 3.0 and 4.0. When comparing results from these runs we find discrepant results.
We have four questions:
- I assume that some degree of difference in outcome between the two FS
versions is to be expected. But how large a difference would you consider "normal" or acceptable? Our differences seem to be quite large. I have attached the outputs for lh. for two subjects with extensions fs3 (Freesurfer 3.0) and fs4 (Freesurfer 4.0).
- I have inspected the subjects with the largest discrepancies and so far
it seems that it is v4.0 that is producing the errors. The tiffs attached showing where I believe there are obvious errors (the fs4´s).
- Is it possible to edit aseg results? (or is that only for the cortical
reconstructions?)
- Can you give any advise on which of the data sets we should do
calculations on?
Have also attached a Excel-file with the volumes for all the subjects in the study (average and STD on the bottom...). If this large variations is to be expected, small changes in subject groups (about 3-5 per cent) could be missed.
Thank you !
Petr
freesurfer@nmr.mgh.harvard.edu
-
??? -
Bruce Fischl -
p.s.bjerkan＠studmed.uio.no