Slicer3:LUTs and Ontologies
Back to Slicer 3 Developers pages
Contents
Introduction
The NAC CCA Core is developing tools to make computer-readable sources of medical knowledge available to the Slicer, NAC, and NA-MIC communities. This knowledge often takes the form of a hierarchy or ontology. Ontologies are available to represent a large range of biological knowledge, but they are often large and difficult to use by medical imaging researchers.
Lookup tables
In Slicer, the simplest and most widespread application of medicine-specific knowledge are the color lookup tables (LUT) that Slicer uses to map the integer values contained in label map images into structure names and colors. We have begun to use this wiki as both a centralized repository for lookup table information as well as a collaborative tool for creating new mappings. The following wiki pages are in a special format that allows conversion into Slicer lookup tables.
Converting wiki pages to and from LUTs
Web-based tools that convert between wiki and Slicer color files can be found at http://labelpage.halle.us/
Link to Lut pages
LUTs for Slicer4
LUTs for Slicer 3.6
- GenericAnatomyColors.txt: Default LUT for Slicer 3.6
- GenericColors.txt: Generic Colors LUT for Slicer3.6 (under construction)
- Preparation work
- 2010 Color lookup table with colors chosen based in part on traditional anatomic atlases. (Became default LUT for Slicer 3.6.)
- development samples of new colors for 2010 default lut
Custom LUTs
- 2010 Colors for a generic labelmap
- 2010 Label colors for brain segmentation
- Freesurfer labels
- Mori-Atlas labels
- Lauras labels
Brain Atlas LUTs
- 2010 Brain Atlas (Good example of the wiki-based labelpage format, including multiple tables.)
Ontologies and Slicer
Ontologies of anatomy such as RadLex and FMA information about many anatomic structures and rich relationships between them. We can use these ontologies to create or augment color lookup tables, form model hierarchies, interface to end-user tools such as QueryAtlas, and even drive segmentation algorithms. The CCA Core is creating simplified representations of these ontologies to make them more accessible to Slicer developers.
The CCA Core is also use local expertise in anatomy to define workable subsets of information for the SPL, Slicer, and NA-MIC communities. Florin Talos has drafted a simplified whole-body hierarchy that has in turn been augmented with FMA-derived details. Links to these works in progress can be found here.
