Hi again,
I've finished adding the extensions I was planning. The summary of changes is:
-New tool "Threshold Components" under the 3D segmentation tab. It allows
connected threshold segmenting over multiple seed points, and assorted UI
improvements to streamline the process (compared to 3DRegionGrow).
-New tool "Pruning" under Segmentation Utilities -> Morphological Operations
tab. This incorporates the itkBinaryPruningImageFilter to help clean up
segmentations.
-New feature added to Deformable Clipping Plane View, which lets you extract a
clipped piece of a segmentation and save it as a new segmentation. (Before, you
could calculate volumes but not actually use the tool to edit the
segmentation). Now, the Clipping Plane View functions as a 3D eraser tool where
you can clip off unwanted pieces from a segmentation.
Anything else I need to do before sending a pull request? I'm 43 commits behind
master if that's an issue. My github fork is located here:
https://github.com/ac3957/MITK/tree/T22654-SegmentationModuleExtensions
Best,
Andrew
________________________________
From: Chen, Andrew
Sent: Wednesday, March 22, 2017 1:45 PM
To: Goch, Caspar Jonas; mitk-users@lists.sourceforge.net
Subject: RE: Extending the segmentation module
Hi Caspar,
The tool I am envisioning should be usable in general for any application, and
I see it integrating well with the manual editing tools already available (for
post-processing). I see this project as accomplishing two goals: 1)
streamlining the 3D segmentation of a multi-part object (which currently
requires multiple segmentations to be run and union'ed), and 2) adding optional
morphological operations to allow for the smoothing of segmentations, all while
sensitive to user adjustments. So I will go for the tool based approach.
I thought so about the patent issue; I've dumped that fuzzy segmentation tool
onto a branch that will not make it into the final product. The rest of my work
is completely contributable.
Thank you for the advice. I've created a phabricator thread (T22654) for
further updates on this project.
Best,
Andrew
________________________________
From: Goch, Caspar Jonas [c.g...@dkfz-heidelberg.de]
Sent: Wednesday, March 22, 2017 10:02 AM
To: Chen, Andrew; mitk-users@lists.sourceforge.net
Subject: AW: Extending the segmentation module
Hi Andrew,
Thanks for letting us know. We are aware that the tool creation documentation
is pretty out of date/missing, so good job creating a new one. The tool
framework is rather historical and replacing it is somewhere on our internal
wishlist (and has been for a while). However we currently do not have a
timeline for a microservice based replacement.
So, with that context out of the way:
The question of tool/dedicated view regarding the contribution is mainly
whether it is designed to solve a general problem. If your tool is (more or
less) useful no matter what structure the end user is currently looking at, go
for a tool based approach. If it is more an answer to a specific problem where
you need only that tool and no other, I would tend towards a dedicated view.
Both could be contributed to MITK proper (even if we do not use the UI of
either tool or view, we could integrate the underlying algorithm (+tests) into
the module).
As for the patent issue, contributing to MITK (according to [1]) requires the
contributor to confirm that they have the rights to contribute the code to us.
So if you are not sure whether you are allowed to contribute the code, you
probably should not. This includes whether your employer even allows the work
they paid for being shared with the world.
Best,
Caspar
[1] http://mitk.org/wiki/How_to_contribute
Von: Chen, Andrew [mailto:ac3...@cumc.columbia.edu]
Gesendet: Dienstag, 21. März 2017 17:30
An: mitk-users@lists.sourceforge.net
Betreff: [mitk-users] Extending the segmentation module
Hello,
I am new to MITK and would like to help develop for the Segmentation module. Is
that module accepting contributions for new tools, or should I write my
software as a new plugin?
The tool I am envisioning would be similar to combining the Binary Threshold
and Picking tools, but allowing for multiple selected regions, and having the
option of eliminating holes and isthmuses in the volume. (The rationale for not
editing these independently is so one can adjust the threshold after selecting
the regions and getting a preview). Please let me know if you have any
suggestions or tips regarding this.
(Also, as a test run, I created a tool that applies ITK's fuzzy connectedness
filter to segment images. It looks like there may be potential patent issues
with that particular ITK library, but inform me if this piece might be useful.)
Best,
Andrew
------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
_______________________________________________
mitk-users mailing list
mitk-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mitk-users
