[mitk-users] Question: reinit

Discussion:

[mitk-users] Question: reinit - image flip

Block, Nikolas

2015-06-16 22:10:42 UTC

Dear MITK folks,

Iâm having a bit of trouble with the global reinit vs reinit functions. It seems when I click global reinit, the image is oriented correctly. However, when I click reinit, which I need to do to perform a segmentation, the image flips (Right becomes Left), reorganizes the axial, coronal, and sagittal views, and it rotates some of these views. Is there a way to prevent this from happening while maintaining the ability to continue segmenting?

Many thanks,
Nik
---

Nikolas R. Block
UC Berkeley - UCSF Joint Medical Program
MD 2019 | MS 2017

***@gmail.com<mailto:***@gmail.com> | ***@berkeley.edu<mailto:***@berkeley.edu> | ***@ucsf.edu<mailto:***@memory.ucsf.edu>
Linkedin<http://www.linkedin.com/in/nikblock>

Goch, Caspar Jonas

2015-06-17 14:33:11 UTC

Permalink

Hi Nik,

as far as I can tell this seems to be the same issue as described in bugs 11477 and 16772 [1,2]. I believe there is currently no fix for it (apart from changing your image matrix, which is not always an option). Someone in our group is currently starting to work on it and possible solutions will be discussed in next weekâs meeting, so those bugs should be updated soonish. Any comments are welcome.

Best,
Caspar

[1] http://bugs.mitk.org/show_bug.cgi?id=11477
[2] http://bugs.mitk.org/show_bug.cgi?id=16772

Von: Block, Nikolas [mailto:***@ucsf.edu]
Gesendet: Mittwoch, 17. Juni 2015 00:11
An: mitk-***@lists.sourceforge.net
Betreff: [mitk-users] Question: reinit - image flip

Dear MITK folks,

Iâm having a bit of trouble with the global reinit vs reinit functions. It seems when I click global reinit, the image is oriented correctly. However, when I click reinit, which I need to do to perform a segmentation, the image flips (Right becomes Left), reorganizes the axial, coronal, and sagittal views, and it rotates some of these views. Is there a way to prevent this from happening while maintaining the ability to continue segmenting?

Many thanks,
Nik
---

Nikolas R. Block
UC Berkeley - UCSF Joint Medical Program
MD 2019 | MS 2017

***@gmail.com<mailto:***@gmail.com> | ***@berkeley.edu<mailto:***@berkeley.edu> | ***@ucsf.edu<mailto:***@memory.ucsf.edu>
Linkedin<http://www.linkedin.com/in/nikblock>

Fetzer, Andreas

2015-06-18 08:22:44 UTC

Permalink

Hi Nik,

the difference between the two reinit functions is:

Global reinit:

1. The orientation of the worldgeometry, which basically defines the rendering space, is set to the standard coordinate system
2. The size of the worldgeometry is calculated, so that it includes all loaded data (depends on size and position of you loaded data)
3. The spacing is set to the smallest existing spacing regarding you loaded data

Reinit:

1. The orientation of the worldgeometry, is aligned according to the orientation of the currently selected datanode
2. The size of the worldgeometry is set to the size of the currently selected datanode
3. The spacing is set to the spacing of the currently selected datanode

The segmentation plugin requires the woldgeometry to be aligned to the selected image. In most of the cases, if you just loaded one image,
this is not necessary but e.g. if you have loaded two different images with different size and origin, the manual segmentation would not work,
since the world geometry would not be aligned with the image, that you want to segment.

Best,
Andreas

Von: <Goch>, Caspar Goch <***@dkfz-heidelberg.de<mailto:***@dkfz-heidelberg.de>>
Datum: Wednesday 17 June 2015 16:33
An: "'Block, Nikolas'" <***@ucsf.edu<mailto:***@ucsf.edu>>, "mitk-***@lists.sourceforge.net<mailto:mitk-***@lists.sourceforge.net>" <mitk-***@lists.sourceforge.net<mailto:mitk-***@lists.sourceforge.net>>
Betreff: Re: [mitk-users] Question: reinit - image flip

Hi Nik,

as far as I can tell this seems to be the same issue as described in bugs 11477 and 16772 [1,2]. I believe there is currently no fix for it (apart from changing your image matrix, which is not always an option). Someone in our group is currently starting to work on it and possible solutions will be discussed in next week’s meeting, so those bugs should be updated soonish. Any comments are welcome.

Best,
Caspar

[1] http://bugs.mitk.org/show_bug.cgi?id=11477
[2] http://bugs.mitk.org/show_bug.cgi?id=16772

Von: Block, Nikolas [mailto:***@ucsf.edu]
Gesendet: Mittwoch, 17. Juni 2015 00:11
An: mitk-***@lists.sourceforge.net<mailto:mitk-***@lists.sourceforge.net>
Betreff: [mitk-users] Question: reinit - image flip

Dear MITK folks,

I’m having a bit of trouble with the global reinit vs reinit functions. It seems when I click global reinit, the image is oriented correctly. However, when I click reinit, which I need to do to perform a segmentation, the image flips (Right becomes Left), reorganizes the axial, coronal, and sagittal views, and it rotates some of these views. Is there a way to prevent this from happening while maintaining the ability to continue segmenting?

Many thanks,
Nik
---

Nikolas R. Block
UC Berkeley - UCSF Joint Medical Program
MD 2019 | MS 2017

***@gmail.com<mailto:***@gmail.com> | ***@berkeley.edu<mailto:***@berkeley.edu> | ***@ucsf.edu<mailto:***@memory.ucsf.edu>
Linkedin<http://www.linkedin.com/in/nikblock>

------------------------------------------------------------------------------

Urban Simoncic

2015-06-18 08:29:52 UTC

Permalink

Hi,

good to know that! This explanation might be copied to user's guide.
However, I was always wondering why reinit is not done automatically
when necessary. Is any good reason for that?

Best,
Urban

Post by Goch, Caspar Jonas
Hi Nik,
1. The orientation of the worldgeometry, which basically defines the rendering space, is set to the standard coordinate system
2. The size of the worldgeometry is calculated, so that it includes all loaded data (depends on size and position of you loaded data)
3. The spacing is set to the smallest existing spacing regarding you loaded data
1. The orientation of the worldgeometry, is aligned according to the orientation of the currently selected datanode
2. The size of the worldgeometry is set to the size of the currently selected datanode
3. The spacing is set to the spacing of the currently selected datanode
The segmentation plugin requires the woldgeometry to be aligned to the selected image. In most of the cases, if you just loaded one image,
this is not necessary but e.g. if you have loaded two different images with different size and origin, the manual segmentation would not work,
since the world geometry would not be aligned with the image, that you want to segment.
Best,
Andreas
Datum: Wednesday 17 June 2015 16:33
Betreff: Re: [mitk-users] Question: reinit - image flip
Hi Nik,
as far as I can tell this seems to be the same issue as described in bugs 11477 and 16772 [1,2]. I believe there is currently no fix for it (apart from changing your image matrix, which is not always an option). Someone in our group is currently starting to work on it and possible solutions will be discussed in next week’s meeting, so those bugs should be updated soonish. Any comments are welcome.
Best,
Caspar
[1] http://bugs.mitk.org/show_bug.cgi?id=11477
[2] http://bugs.mitk.org/show_bug.cgi?id=16772
Gesendet: Mittwoch, 17. Juni 2015 00:11
Betreff: [mitk-users] Question: reinit - image flip
Dear MITK folks,
I’m having a bit of trouble with the global reinit vs reinit functions. It seems when I click global reinit, the image is oriented correctly. However, when I click reinit, which I need to do to perform a segmentation, the image flips (Right becomes Left), reorganizes the axial, coronal, and sagittal views, and it rotates some of these views. Is there a way to prevent this from happening while maintaining the ability to continue segmenting?
Many thanks,
Nik
---
Nikolas R. Block
UC Berkeley - UCSF Joint Medical Program
MD 2019 | MS 2017
Linkedin<http://www.linkedin.com/in/nikblock>
------------------------------------------------------------------------------
_______________________________________________
mitk-users mailing list
https://lists.sourceforge.net/lists/listinfo/mitk-users

------------------------------------------------------------------------------

Fetzer, Andreas

2015-06-22 09:03:34 UTC

Permalink

Hi Urban,

in case of the segmentation, this is done automatically.
However, even there we have a lot of different opinions regrading whether
this should be done automatically ;)
(Imaging you have navigated the crosshair to a certain position / rotation
and then you create a new
segmentation -> everything will be reset, which is of course annoying
sometimes)
So we do not want to force an automatic reinit too often.

Do you have specific situations in mind, were you think an automatic
reinit might be useful / necessary?

Thank you, good point! We will update the documentation regrading that.

Best regards,
Andreas

Am 18/06/15 10:29 schrieb "Urban Simoncic" unter

Post by Urban Simoncic
Hi,
good to know that! This explanation might be copied to user's guide.
However, I was always wondering why reinit is not done automatically
when necessary. Is any good reason for that?
Best,
Urban

"
t>>

Betreff: Re: [mitk-users] Question: reinit - image flip
Hi Nik,
as far as I can tell this seems to be the same issue as described in
bugs 11477 and 16772 [1,2]. I believe there is currently no fix for it
(apart from changing your image matrix, which is not always an option).
Someone in our group is currently starting to work on it and possible
solutions will be discussed in next week¹s meeting, so those bugs should
be updated soonish. Any comments are welcome.
Best,
Caspar
[1] http://bugs.mitk.org/show_bug.cgi?id=11477
[2] http://bugs.mitk.org/show_bug.cgi?id=16772
Gesendet: Mittwoch, 17. Juni 2015 00:11
Betreff: [mitk-users] Question: reinit - image flip
Dear MITK folks,
I¹m having a bit of trouble with the global reinit vs reinit functions.
It seems when I click global reinit, the image is oriented correctly.
However, when I click reinit, which I need to do to perform a
segmentation, the image flips (Right becomes Left), reorganizes the
axial, coronal, and sagittal views, and it rotates some of these views.
Is there a way to prevent this from happening while maintaining the
ability to continue segmenting?
Many thanks,
Nik
---
Nikolas R. Block
UC Berkeley - UCSF Joint Medical Program
MD 2019 | MS 2017
Linkedin<http://www.linkedin.com/in/nikblock>
-------------------------------------------------------------------------
-----
_______________________________________________
mitk-users mailing list
https://lists.sourceforge.net/lists/listinfo/mitk-users

--------------------------------------------------------------------------
----
_______________________________________________
mitk-users mailing list
https://lists.sourceforge.net/lists/listinfo/mitk-users

Urban Simoncic

2015-06-22 10:12:07 UTC

Permalink

Hi Andreas,

I don't have in mind any specific situation where automatic reinit would
be necessary. I just thought that it would be always useful to have
automatic reinit (If I should do what program ask me to do, then the
program can do it itself :-)). However, I have limited experiences with
MITK and I was not aware that some setting/information/work can be lost
with reinit, and I also have not came yet to the situation when MITK
suggested reinit and I decided not to do it.

Best regards,
Urban

Post by Fetzer, Andreas
Hi Urban,
in case of the segmentation, this is done automatically.
However, even there we have a lot of different opinions regrading whether
this should be done automatically ;)
(Imaging you have navigated the crosshair to a certain position / rotation
and then you create a new
segmentation -> everything will be reset, which is of course annoying
sometimes)
So we do not want to force an automatic reinit too often.
Do you have specific situations in mind, were you think an automatic
reinit might be useful / necessary?
Thank you, good point! We will update the documentation regrading that.
Best regards,
Andreas
Am 18/06/15 10:29 schrieb "Urban Simoncic" unter

"
t>>

------------------------------------------------------------------------------
Monitor 25 network devices or servers for free with OpManager!
OpManager is web-based network management software that monitors
network devices and physical & virtual servers, alerts via email & sms
for fault. Monitor 25 devices for free with no restriction. Download now
http://ad.doubleclick.net/ddm/clk/292181274;119417398;o
_______________________________________________
mitk-users mailing list
https://lists.sourceforge.net/lists/listinfo/mitk-users