Differences between revisions 1 and 11 (spanning 10 versions)
Revision 1 as of 2010-11-17 00:23:44
Size: 449
Editor: KhoaNguyen
Comment:
Revision 11 as of 2011-05-26 17:38:43
Size: 2393
Editor: KhoaNguyen
Comment:
Deletions are marked like this. Additions are marked like this.
Line 2: Line 2:
*Note: These bugs are found while running the Buckner40 data set.<<BR>> ''see also: Buckner40NotesCombined'' Note: These bugs are found while running the Buckner40 data set.<<BR>> ''see also: Buckner40Notes''
||<tablewidth="903px" tableheight="327px"style="font-weight: bold; text-align: center;">Bug ||<style="font-weight: bold; text-align: center;">Example ||<style="font-weight: bold; text-align: center;">Fixed? ||
||There's a difference between the orig surf and white surf after wm edits<<BR>> --usually the orig surf is better || || ||
||White surf does weird indentations even when wm voxels are present (white matter damage?)<<BR>> --again orig surf is better here || || ||
||Surfaces between temporal lobe and the cerebral cortex are usually crazy (random wm voxels) || || ||
||Still some problems with calcarine area - aseg is correct yet the surfaces aren't ||073_dev || ||
||Pial surf isn't correct at midline<<BR>> (not incorrect but not pretty to look at) ||073_dev || ||
||Sometimes the dura is taken as wm and white surf is pulled outward, after deleting the dura from wm.mgz, the white surf still doesn't adjust - the orig surf, however, does. ||092 at 174 131 143, 097 at 117 147 153, 102 at 114 122 162 || ||
||When I click on the snapshot button, I can save a snapshot as either tiff<<BR>>or rgb. When I right or left click on the snapshot button I can only save rgb. If I<<BR>>try to save a .tif in the rgb window, tkmedit crashes with this error:<<BR>><<BR>>*** glibc detected *** tkmedit.bin: free(): invalid next size (normal):<<BR>>0x000000002370d610 ***<<BR>>======= Backtrace: =========<<BR>>/lib64/libc.so.6[0x354c27230f]<<BR>>/lib64/libc.so.6(cfree+0x4b)[0x354c27276b]<<BR>>/lib64/libc.so.6(fclose+0x14b)[0x354c260d5b]<<BR>>tkmedit.bin(cos+0xa8b)[0x41542b]<<BR>>tkmedit.bin(atan+0x2101)[0x4167c1]<<BR>> || || ||
||Watershed method to correct skull strip errors - there seems to be no in between for fine tuning the amount of brain I want present. There's either part of the brain missing (skull strip failure) or the entire skull is present. ||Tried with buckner 145_wstest || ||
Line 4: Line 13:
||<tablestyle="width: 903px; height: 327px;": style="font-weight: bold;">Bug||<: style="font-weight: bold;">Example||<: style="font-weight: bold;">Fixed?||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||
|| || || ||



*Examples are found here<<BR>> /autofs/cluster/freesurfer/test/subjects/x86_64/buckner_data/group_study_fs5.0.0_from_scratch

== Feature Requests ==
||Requests ||Granted? ||
||Quit prompt when pressing Ctrl-q when view window is highlighted in tkmedit. || ||
||Have a tool that can highlight and add wm voxels at the same time? || ||

Bug List from Stable 5.0.0

Note: These bugs are found while running the Buckner40 data set.
see also: Buckner40Notes

Bug

Example

Fixed?

There's a difference between the orig surf and white surf after wm edits
--usually the orig surf is better

White surf does weird indentations even when wm voxels are present (white matter damage?)
--again orig surf is better here

Surfaces between temporal lobe and the cerebral cortex are usually crazy (random wm voxels)

Still some problems with calcarine area - aseg is correct yet the surfaces aren't

073_dev

Pial surf isn't correct at midline
(not incorrect but not pretty to look at)

073_dev

Sometimes the dura is taken as wm and white surf is pulled outward, after deleting the dura from wm.mgz, the white surf still doesn't adjust - the orig surf, however, does.

092 at 174 131 143, 097 at 117 147 153, 102 at 114 122 162

When I click on the snapshot button, I can save a snapshot as either tiff
or rgb. When I right or left click on the snapshot button I can only save rgb. If I
try to save a .tif in the rgb window, tkmedit crashes with this error:

*** glibc detected *** tkmedit.bin: free(): invalid next size (normal):
0x000000002370d610 ***
======= Backtrace: =========
/lib64/libc.so.6[0x354c27230f]
/lib64/libc.so.6(cfree+0x4b)[0x354c27276b]
/lib64/libc.so.6(fclose+0x14b)[0x354c260d5b]
tkmedit.bin(cos+0xa8b)[0x41542b]
tkmedit.bin(atan+0x2101)[0x4167c1]

Watershed method to correct skull strip errors - there seems to be no in between for fine tuning the amount of brain I want present. There's either part of the brain missing (skull strip failure) or the entire skull is present.

Tried with buckner 145_wstest

*Examples are found here
/autofs/cluster/freesurfer/test/subjects/x86_64/buckner_data/group_study_fs5.0.0_from_scratch

Feature Requests

Requests

Granted?

Quit prompt when pressing Ctrl-q when view window is highlighted in tkmedit.

Have a tool that can highlight and add wm voxels at the same time?

BugsInProgress (last edited 2011-05-26 17:41:09 by KhoaNguyen)