Opened 2 years ago
Closed 17 months ago
#9526 closed enhancement (fixed)
DICOM Segmentations: include SEG models opened from DICOM in the list of editable segs
Reported by: | Elaine Meng | Owned by: | Zach Pearson |
---|---|---|---|
Priority: | high | Milestone: | 1.8 |
Component: | DICOM | Version: | |
Keywords: | Cc: | ||
Blocked By: | Blocking: | ||
Notify when closed: | Platform: | all | |
Project: | ChimeraX |
Description
was mentioned in #9397 but is really a separate thing. Would be nice if you could take segmentations from others (or ones you made and saved to file previously) and continue refining/editing them.
I was wishy-washy about it before but now I think it would be really useful. You don't want to start over if you see you made a little mistake in the one you saved as a file, even if it were possible to restore a session with these models (which isn't possible currently as I understand it).
Attachments (1)
Change History (9)
comment:1 by , 2 years ago
comment:2 by , 2 years ago
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
comment:3 by , 18 months ago
Resolution: | fixed |
---|---|
Status: | closed → reopened |
This feature has gone away. Tested in UCSF ChimeraX version: 1.9.dev202405110020 (2024-05-11)
comment:4 by , 18 months ago
Also tested in version: 1.8rc202405140121 (2024-05-14)
Open any RIDER lung set that includes the segmentations of the tumor, then start the Segmentations tool, see that the SEG models are not listed in that tool. It definitely used to work and you could edit them, which was really nice.
comment:5 by , 17 months ago
Milestone: | → 1.8 |
---|
comment:6 by , 17 months ago
Also does not list any SEG model that was opened from a dcm file previously saved with the Segmentations tool. If I open the image data first, then open the dcm file previously saved with the Segmentations tool, then open the Segmentations tool, then that SEG model does show up in the Model Panel but not in the Segmentations tool. If I then close the SEG model from the Model Panel and try reopening it with a command, it doesn't show up anywhere (neither the Model Panel nor the Segmentations tool). I will attach nerts.dcm which is just a seg in a single plane that was previously saved with the Segmentations tool for the 4-24533 image which is the dataset provided with, i.e. linked to, the DICOM-VR tutorial.
by , 17 months ago
comment:7 by , 17 months ago
nerts.dcm saved from Segmentations tool and tested/reopened as per previous comments using the current daily build UCSF ChimeraX version: 1.9.dev202405140241 (2024-05-14)
comment:8 by , 17 months ago
Resolution: | → fixed |
---|---|
Status: | reopened → closed |
Should be fixed tomorrow. DicomGrids lacked a property that segmentations (the bundle) expects segmentations (the data structure) to have, so I have added it.
I agree. A problem is that the current DICOM code reads matrices in a strange way, directly from files instead of relying on SeriesFiles we already have read in. Once that's changed and the DicomGrid data is in memory all the time this will become feasible.