Documentation/4.1/Modules/SimpleRegionGrowingSegmentation

For the latest Slicer documentation, visit the read-the-docs.

Introduction and Acknowledgements

This work is part of the National Alliance for Medical Image Computing (NA-MIC), funded by the National Institutes of Health through the NIH Roadmap for Medical Research, Grant U54 EB005149. Information on NA-MIC can be obtained from the NA-MIC website. Author: Jim Miller, GE Contact: Jim Miller <email>millerjv@ge.com</email>

NA-MIC   GE   Itk

Module Description

A simple region growing segmentation algorithm based on intensity statistics. To create a list of fiducials (Seeds) for this algorithm, click on the tool bar icon of an arrow pointing to a starburst fiducial to enter the 'place a new object mode' and then use the fiducials module. This module uses the Slicer Command Line Interface (CLI) and the ITK filters CurvatureFlowImageFilter and ConfidenceConnectedImageFilter.

The fiducial tool bar icon (identified inside a red box) on the Slicer tool bar is shown.

Fiducial tool bar icon

Use Cases

This module can be used to obtain a volumetric segmentation of a region of interest. Most frequently used for these scenarios:

• Example volumetric segmentation of a tumor given a set of fiducial points selected on the tumor:

User input in the form of fiducials

Segmentation result: Multiplier=2.5, Neighborhood radius=1, Number of Iterations=5

This module is useful for obtaining a 2-class segmentation. The fiducials are placed on parts of the image that correspond to the target (foreground). The segmentation can be refined by modifying the number of iterations, the multiplier, and the neighborhood radius options available on the user interface panel. Additionally, the user may also place additional fiducials on the image.

Panels and their use

Parameters:

• Smoothing Parameters
  • Smoothing iterations: Number of smoothing iterations
  • Timestep: Timestep for curvature flow
• Segmentation Parameters
  • Number of iterations: Number of iterations of region growing
  • Multiplier: Number of standard deviations to include in intensity model
  • Neighborhood Radius: The radius of the neighborhood over which to calculate intensity model
  • Output Label Value: The integer value (0-255) to use for the segmentation results. This will determine the color of the segmentation that will be generated by the Region growing algorithm
  • Seeds: Seed point(s) for region growing
• IO
  • Input Volume: Input volume to be filtered
  • Output Volume: Output filtered

The user interface panel:

Similar Modules

• Robust Statistics Segmenter
• GrowCut segmentation in the editor module