Slicer3:EM

Goals

The goal of this project is to implement a Slicer3 module for creating templates that will be used in segmentation tasks. Each template contains a hierarchy/tree of tissue types, an atlas of prior probability maps for particular tissue types (along with associated T1 and T2 images), and global and node specific algorithm parameters. The template, which is written as an XML file, can be used later to segment new images. The target audience for this module is someone familiar with brain atlases and tissue labels, not a computer scientist.

This module will be demonstrated at the NAMIC All-Hands meeting in January 2007.

Features

• Slicer3 module for building templates for EM Segmentation tasks (algorithm developed by Kilian Pohl et al.)
• tree view for building hierarchy of tissue types and specifying algorithm parameters
• wizard-like implementation to aid non-expert users
• output template as XML file
• written demo with example data

Plan

We are implementing this project in C++ as a programmatic Slicer3 module. A set of workflow classes are being developed to facilitate the wizard interface. In the interest of time we will port as much code from the Slicer2.6 module as possible.

Status

6 Nov 06

• Overall:
  • Slicer3 module scaffolding (Done, Brad)
  • Demo design (underway, Brad)
• Data:
  • porting MRML nodes from Slicer2.6 to Slicer3 (underway, Brad)
• Logic: on hold until MRML nodes are finished (Brad)
• GUI:
  • porting intensity distribution widget (underway, Yumin)
  • workflow classes (underway, Sebastien)
  • implement GUI for this module (not started)