Difference between revisions of "2012 Winter Project Week:RTSS"
From NAMIC Wiki
| (One intermediate revision by the same user not shown) | |||
| Line 7: | Line 7: | ||
==Key Investigators== | ==Key Investigators== | ||
| − | Greg Sharp, Steve Pieper, Andras Lasso | + | Greg Sharp, Steve Pieper, Andras Lasso, Kevin Wang, Csaba Pinter |
<div style="margin: 20px;"> | <div style="margin: 20px;"> | ||
| Line 27: | Line 27: | ||
* Preliminary discussions, requirement specification, identification of design options completed. | * Preliminary discussions, requirement specification, identification of design options completed. | ||
* Mockup of UI available for feedback | * Mockup of UI available for feedback | ||
| − | |||
| − | |||
| − | |||
</div> | </div> | ||
Latest revision as of 00:53, 13 January 2012
Home < 2012 Winter Project Week:RTSS
Sketch of editor design
Key Investigators
Greg Sharp, Steve Pieper, Andras Lasso, Kevin Wang, Csaba Pinter
Objective
Find a good representation of RT structure sets in Slicer. There were some preliminary work in Slicer3 (using Plastimatch) and new efforts (Plastimatch, DCMTK/DCMRT, SparKit).
Approach, Plan
- Specify requirements for representing structure sets in Slicer
- Decide on the design
- Implement the solution in Slicer4
Progress
- Preliminary discussions, requirement specification, identification of design options completed.
- Mockup of UI available for feedback
Notes
- Requirements
- Support up to 100 structures
- Support overlapping structures
- Display a name for each structure
- Show/hide each structure
- Manual editing of each structure
- Rigid/deformable transform of the structures
- DICOM-RT export
- Design options
- Store each structure as one labelmap (lots of memory, data loss when converted from contours)
- Store multiple structure in a labelmap as bitfield (limited to 32 structures, complicated editing, data loss when converted from contours)
- Store as model or annotation contours (not the usual representation for segmented structures, models cannot be edited)
- Considerations
- Structure contours must be associated with background volume DICOM data in order to be exported correctly. This effort should be coordinated with the DICOM SR/AIM annotation effort
- How should these be represented in ITK and the SEM?
