Slicer:Image Editor

Goal

As part of the slicer 3 architecture discussion we've realized that a very commonly used tool in slicer is the Editor module for defining regions of interest in volumetric medical image data sets.

NA-MIC is in a unique position of being able to dedicate significant resources toward core software infrastructure development. Several people have commented on the fact that the image editor is a critical tool, but not likely to generate "research" results so it is a difficult project to fund through traditional grant mechanisms.

Background and Motivation

Slicer's editor module is essentially unchanged since it's initial implementation by Dave Gering in the early versions of slicer. It has been used extensively by the Core 3 DBP group at Harvard/VA for morphometric studies and by other BWH-related groups. It has significant functionality, but also has some distinct limitations.

As part of the Dartmouth Dissemination Workshop, the day-to-day importance of custom tools for morphometry was a recurring comment.

As we have been discussing areas of collaboration with summarized here.

The image editor should provide a good driver for functionality in the core of a new slicer architecture because of the need for:

• highly customized UI, yet easy to use by non-programmers
• interoperable with other programs such as automatic segmentation tools
• combination of 2D and 3D interaction

Use Cases

Morphometry

Within the BWH community, the largest use has been to outline cortical and subcortical brain structures to obtain anatomically accurte ROI analyses for a number of purposes such as to measure differences in volume of key structures between subject groups (e.g. schizophrenics vs controls).

Shape Analyses

Accurate ROIs can be further analyzed by shape analysis techniques.

Secondary Measures within ROI

Quantities derrived from fMRI (e.g. activation statistics) or DTMRI (e.g. FA) are often compared within anatomical regions.

Input for Simulation

Simbios and other simulation efforts use 3D models as input to physical simulations such as neuromusculoskeletal dynamics.

Basis for Further Computation

Labeled ROIs can be used as the starting point for seeds in tractorgraphy analyses, for example.

Surgery Planning

Manual ROI drawing is used to define items such as tumor outlines that are not extractable automatically.