This issue should be postponed until T28523: Evaluate MITK segmentation survey (to now the new priority) is done.
- Queries
- All Stories
- Search
- Advanced Search
- Transactions
- Transaction Logs
Feed All Stories
All Stories
All Stories
Jun 4 2021
Jun 4 2021
floca added a parent task for T28524: Refactor LabelSetImage and associated classes.: T28499: Refactor/rewrite RegionGrowing3D tool.
floca added a parent task for T28523: Evaluate MITK segmentation survey: T28499: Refactor/rewrite RegionGrowing3D tool.
floca added a parent task for T27437: Migrate MITK to use ITK 5.x : T28499: Refactor/rewrite RegionGrowing3D tool.
floca triaged T28525: Design refactoring of LabelSetImage and associated classes as Unbreak Now! priority.
floca moved T28464: [Segmentation] New Add/Remove tool from Backlog to MITK Meeting on the Request for Discussion board.
floca moved T28464: [Segmentation] New Add/Remove tool from Backlog to Segmentation on the MITK (v2021.10) board.
floca raised the priority of T28142: [Segmentation] Remove plugin redundancy with MultiLabelSegmentation from Normal to Unbreak Now!.
floca updated the task description for T28479: [Segmentation] Rethink / reevaluate our existing segmentation tools / algorithms.
floca added a project to T28320: [Segmentation] Region growing does not work if seed point is already inside a segmentation: Missing Info.
floca added a comment to T28145: SetRegionTool does not work as expected (flood fill like) with concentric structures..
This issue should be postponed until T28523: Evaluate MITK segmentation survey (to now the new priority) is done.
floca added a comment to T28247: [Segmentation] Watershed tool freezes when slider and threshold are set to a low value.
This issue should be postponed until T28523: Evaluate MITK segmentation survey (to now the new priority) and T27437: Migrate MITK to use ITK 5.x (check for performance improvments due to refactored ITK) is done.
floca added a project to T28244: [Segmentation] Fast Marching Tool not really responsive for large images: Missing Info.
This issue should be postponed until T28523: Evaluate MITK segmentation survey (to now the new priority) and T27437: Migrate MITK to use ITK 5.x (check for performance improvments due to refactored ITK) is done.
floca added a comment to T28479: [Segmentation] Rethink / reevaluate our existing segmentation tools / algorithms.
Also see our internal discussion about who to address several more fundamental design issues: https://hub.dkfz.de/f/5459842
floca added a project to T28490: Segmentation plugin: "new" segmentation should not run reinit!: Cleared.
floca added a parent task for T28490: Segmentation plugin: "new" segmentation should not run reinit!: T27613: Improve reinit behavior.
floca moved T28523: Evaluate MITK segmentation survey from Backlog to Cycle on the MITK (v2021.10) board.
Jun 4 2021, 12:21 PM · Restricted Project
Jun 4 2021, 12:21 PM · Restricted Project
isensee updated the task description for T28500: Multilabel Segmentation Utilities do not work for multilabel segmentations.
isensee added a comment to T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes.
As it is an inexpensive operation my first guess would be an issue related to the amount of used memory. For example, the image above eats 3,5 GB, Activating the interpolation already triples it. Confirming adds at least another 3,5 GB. That could also explain why it is so slow (swapping). But at this point it is just a first assumption. On a 16 GB Windows machine I was able to complete the interpolation without crash but it took a while.
Jun 4 2021, 12:06 PM · Restricted Project
Without looking into it yet, I think the Nvidia solution for accelerated remote desktop is currently by far the most promising as it tackles the actual issue.
Jun 3 2021
Jun 3 2021
Running MITK within a Windows Server 2012 R2 Session (VMWare virtualized) and using the fdossena's opengl mesa libraries the current state is (no crash, though):
This seems to be still an issue. Maybe there is new information out there or it could be re-tested with newer Windows versions / drivers?
Jun 2 2021
Jun 2 2021
s434n added a comment to T28514: [Multilabel Segmentation] Changing visibility of labels on higher layers does not trigger update.
It seems to be the same bug, I just failed to find the existing task. Since I found it in the last Release version, the bugfix was just not included yet, but the problem is still solved. I guess this task can be closed then
floca moved T27684: [Segmentation] Improve Region Growing 2D algorithm and interaction from Segmentation to Cycle on the MITK (v2021.10) board.
• thomass added a comment to T28514: [Multilabel Segmentation] Changing visibility of labels on higher layers does not trigger update.
I am indeed still reading since I still have tasks to do (I will not leave beforehand just as promised :-)). I think you are referring to T28383 / D477 ... maybe it didn’t fully do the trick.
kislinsk added a comment to T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes.
A few things that impact the handling of huge images:
kislinsk closed T28144: CorrectorTool does not properly handle unlocked lables of LabelSet Images as Resolved.
The tool was already removed from the last release because people tend to use the Add/Subtract tool for the same tasks anyway and we could reduce code/complexity.
kislinsk moved T28159: Officially support MITK using system Qt from MITK Meeting to Backlog on the Request for Discussion board.
Discussed it in the MITK meeting:
• kleina moved T28455: RT struct 3D visibility is not working correctly from Backlog to MITK Meeting on the Request for Discussion board.
kislinsk moved T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes from Backlog to Cycle on the MITK (v2021.10) board.
kislinsk added a comment to T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes.
And we seem to have a huge leak here. Even after deleting all the nodes again, half of my RAM is still in use.
kislinsk added a comment to T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes.
As it is an inexpensive operation my first guess would be an issue related to the amount of used memory. For example, the image above eats 3,5 GB, Activating the interpolation already triples it. Confirming adds at least another 3,5 GB. That could also explain why it is so slow (swapping). But at this point it is just a first assumption. On a 16 GB Windows machine I was able to complete the interpolation without crash but it took a while.
isensee updated subscribers of T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes.
kislinsk updated subscribers of T28514: [Multilabel Segmentation] Changing visibility of labels on higher layers does not trigger update.
Maybe @thomass is still reading. :) I think she reported the same thing some weeks ago but I cannot find it. It's just a missing RequestRenderWindowUpdate().
Jun 1 2021
Jun 1 2021
@kislinsk sorry for the late reply. I am not using the convert to segmentation functionality because it draws contours by default and that is annoying because it obstructs too much from the image. Multilabel color map does exactly the right thing, is easy to use and does not require and additional mouse clicks. It is just more convenient :-)
s434n triaged T28515: [Statistics] Difference in behavior between the two data selection methods as Low priority.
s434n triaged T28514: [Multilabel Segmentation] Changing visibility of labels on higher layers does not trigger update as Normal priority.
closing gVisor experiments since Kata is our first choice and Kata is able to run in our bare-metal machine, hence temporarily closing gVisor testing.
Jun 1 2021, 10:35 AM · Restricted Project
s669m closed T28350: Trial Run with Isolated Sandbox Container gVisor, a subtask of T28175: TFDA MVP v1, as Resolved.
Jun 1 2021, 10:34 AM · Restricted Project
Closing the issue since gVisor has lots of limitations with microk8s and our current Kaapana architecture.
Jun 1 2021, 10:34 AM · Restricted Project
May 31 2021
May 31 2021
paul jaeger <p.jaeger@dkfz.de> committed rMDTPUBLICf678bc5e91c3: update retina unet citation (authored by paul jaeger <p.jaeger@dkfz.de>).
update retina unet citation
May 28 2021
May 28 2021
kislinsk moved T28500: Multilabel Segmentation Utilities do not work for multilabel segmentations from Backlog to Segmentation on the MITK (v2021.10) board.
kislinsk triaged T28500: Multilabel Segmentation Utilities do not work for multilabel segmentations as High priority.
May 26 2021
May 26 2021
s434n triaged T28510: [Segmentation Utilities] Switched definitions in documentation as Normal priority.
May 25 2021
May 25 2021
isensee updated the task description for T28500: Multilabel Segmentation Utilities do not work for multilabel segmentations.
May 24 2021
May 24 2021
floca added subtasks for T28499: Refactor/rewrite RegionGrowing3D tool: T28275: [Segmentation] AdaptiveRegionGrowingTool must be refactored / 4D Segmentation returns error message when point is not on the correct timestep, T27872: [Segmentation] Region Growing 3D GUI does not provide a "Overwrite" checkbox, T27684: [Segmentation] Improve Region Growing 2D algorithm and interaction.
May 20 2021
May 20 2021
Hey Marco, I had uploaded the models there back in the days:
import numpy as np import colorsys from skimage import io
May 19 2021
May 19 2021
kalali moved T28490: Segmentation plugin: "new" segmentation should not run reinit! from Backlog to Segmentation on the MITK (v2021.10) board.
kalali triaged T28490: Segmentation plugin: "new" segmentation should not run reinit! as High priority.
I agree on that. I had some problems with that with the new mxn-multiwidget. The topic was also mentioned in T27613: Improve reinit behavior.
gaoh reopened T28380: [MultLabel Segmentation, MITK FlowBench] Segmentation not recognized as segmentation in Release v2021.02 as "Open".
The same issue seems to apply to the normal workbench, so this container has also to be updated to the current master
isensee updated the task description for T28491: [Segmentation] Slice interpolation 'confirm for all slices' sometimes crahes.
isensee updated the task description for T28490: Segmentation plugin: "new" segmentation should not run reinit!.