Difference between revisions of "Projects:Iowa Longitudinal PETCT Analysis using 3DSlicer"

From NAMIC Wiki
Jump to: navigation, search
 
(5 intermediate revisions by the same user not shown)
Line 8: Line 8:
  
 
==Items under development==
 
==Items under development==
 +
Issues raised by Reinhard and Marcus on Feb 19, 2013 call:
 +
* scene save/load is not working in LongPETCT (Paul is working on this)
 +
* would like to be able to segment on the whole image volume without having to specify the ROI (Paul will look into adding this as an option)
 +
* there is a crash switching between Editor and Charts layout, only on linux (Andrey will test)
 +
* need to discuss how Markus' SUV calculation extension can be integrated with Paul's module; Steve: makes sense to package that CLI together with the Paul's module (Paul started working on this)
 +
** Markus started from the code that has not been adjusted to work with ITK4, currently does not compile with ITK4, need to apply fixes in Slicer trunk
 +
Other feature requests:
 +
* use box-and-whiskers for charts (request from Ron; problem: max/mean are not included)
 +
* report ROI SUV statistics in the findings table (Paul will implement)
 +
* allow externally supplied label to be specified for the finding segmentation (under consideration)
  
 
==Slicer-related infrastructure discussion issues==
 
==Slicer-related infrastructure discussion issues==
 +
* should there be an infrastructure for extension/module-specific parameter settings, and an interface to keep them in Slicer settings file? Should these settings be accessibel from Slicer Settings dialog, or from a panel in the extension?
 +
* It is important for reproducibility purposes to be able to keep track of the parameters and initialization (e.g., fuducials) used for certain processing operation. Should we consider how to do this on the Slicer infrastructure level, or resolve on a per-case basis?
 +
* does it make sense to do SUV correction on PET import, and store SUV volume together with the original data?
  
 
=References=
 
=References=

Latest revision as of 15:01, 6 March 2013

Home < Projects:Iowa Longitudinal PETCT Analysis using 3DSlicer

Summary

Participants

Technical details

Use-case workflow

Items under development

Issues raised by Reinhard and Marcus on Feb 19, 2013 call:

  • scene save/load is not working in LongPETCT (Paul is working on this)
  • would like to be able to segment on the whole image volume without having to specify the ROI (Paul will look into adding this as an option)
  • there is a crash switching between Editor and Charts layout, only on linux (Andrey will test)
  • need to discuss how Markus' SUV calculation extension can be integrated with Paul's module; Steve: makes sense to package that CLI together with the Paul's module (Paul started working on this)
    • Markus started from the code that has not been adjusted to work with ITK4, currently does not compile with ITK4, need to apply fixes in Slicer trunk

Other feature requests:

  • use box-and-whiskers for charts (request from Ron; problem: max/mean are not included)
  • report ROI SUV statistics in the findings table (Paul will implement)
  • allow externally supplied label to be specified for the finding segmentation (under consideration)

Slicer-related infrastructure discussion issues

  • should there be an infrastructure for extension/module-specific parameter settings, and an interface to keep them in Slicer settings file? Should these settings be accessibel from Slicer Settings dialog, or from a panel in the extension?
  • It is important for reproducibility purposes to be able to keep track of the parameters and initialization (e.g., fuducials) used for certain processing operation. Should we consider how to do this on the Slicer infrastructure level, or resolve on a per-case basis?
  • does it make sense to do SUV correction on PET import, and store SUV volume together with the original data?

References