Documentation/Nightly/Extensions/ShapeRegressionExtension

From Slicer Wiki
Revision as of 23:21, 30 January 2018 by Lpascal (talk | contribs)
Jump to: navigation, search
Home < Documentation < Nightly < Extensions < ShapeRegressionExtension

For the stable Slicer documentation, visit the 4.10 page.

Introduction and Acknowledgements

Extensions: ShapeRegressionExtension
Author: Laura Pascal (Kitware Inc.)
Contributors: Beatriz Paniagua (Kitware Inc), James Fishbaugh (University of North Carolina)
Contact: Laura Pascal, <email>laura.pascal@kitware.com</email>, Beatriz Paniagua, <email>beatriz.paniagua@kitware.com </email>
Acknowledgments: NIH NIBIB R01EB021391 (Shape Analysis Toolbox for Medical Image Computing Projects)
License: Apache License, Version 2.0

Extension Description

This extension is composed by two modules :

  • RegressionComputation:
    • Computation of time-regressed shapes in a collection of 3D shape inputs associated to a linear variable thanks to shape4D CLI.
  • RegressionVisualization:
    • Plotting of the time-regressed shape volume evolution according to a linear variable
    • Visualization of the sequence of the time-regressed shapes generated (using Sequences module)

Interface

Regression Computation module:

RegressionComputation-Interface.png

Regression Visualization module:

RegressionVisualization-Interface.png


Use Cases

Regression Computation Module

This module is a user-friendly interface allowing to easily use shape4D CLI in order to compute time-regressed shapes in a collection of 3D shape inputs associated to a linear variable.

Shape4D Inputs

Regression Computation Input Shape Parameters
This section allows the user to specify the input shapes for each time point as well as their associated parameters.

To do so, the user can either specify :

  • A folder containing the input shapes. The user would then have to specify the parameters associated to each input shape. A CSV file (CSVInputshapeparameters.csv) containing all these information will be automatically created in the output folder specified by the user.
  • A CSV files containing in each row the input shape file path followed by the shape parameters "Time Point" , "Sigma W" , "Tris" and "Weight" .
Example of CSV files

Shape4D Parameters

Time Parameters

Regression Computation Time Parameters
This section allows the user to specify:
  • The first and last time points of the input shapes
  • The number of time-regressed shapes that will be computed between the first and last input shapes

Output Parameters

Shape4D CLI will compute several sets of time-regressed shapes where the basename of each shape will be: <prefix>_n_time_00i with:
  • <prefix>: the prefix can be specified by the user
  • n: the iteration number of the shape regression process (i.e. represents the number of the time-regressed shape serie)
  • i: the time-regressed shape number
The last iteration will create the final time-regressed shape set where the basename of the shapes will be <prefix>_final_time_00i.
Regression Computation Output Parameters
This section allows the user to specify :
  • The output folder where the computed time-regressed shape sets will be stored
  • The prefix of the basename of each time-regressed shapes
  • The occurrence at which the time-regressed shapes set will be saved

Optimal Parameters

Regression Computation Optimal Parameters
This section allows the user to specify:
  • the optimal method used by Shape4D CLI
  • The minimum distance that will stopped the shape regression process
  • The maximum number of iteration

Regression Visualization Module

This module allows to easily visualize the results obtained by the RegressionComputation module by using severals slicer extensions (Sequences, Colors, Tables).


Inputs

This section allows the user to create a sequence containing the time-regressed shapes computed by RegressionComputation module.
Shape Regression Inputs
In order to do that, the user would have to specify:
  • The input directory which should contain the time-regressed shape sets computed by RegressionComputation module
  • The rootname of the time-regressed shape set that should be <prefix>_n_time_ with:
    • the prefix specified by the user in the RegressionComputation module
    • n the iteration number of the shape regression process that the user want to visualize (n will be final if the user wants to display the final results computed)

Visualization

Sequence Visualization

Sequence Visualization Options
In this tab, the user could:
  • customize the color maps of the shapes contained in the sequence
  • navigate through the sequence thanks to a browser
Example of a brain growth sequence


Regression Plot

This tab allows the user to plot the time-regressed shape volume evolution according to the time.
Regression Plot
To do so, the user will have to specify the path of a CSV file containing the following information used to compute the set of time-regressed shapes in RegressionComputation module:
  • the path of each input shape
  • the time-point associated
Example of Regression Plot


Additional Information

This extension is available on 3DSlicer and on SlicerSALT.

Similar Modules

N/A

Information for Developers

The source code is available on github