Difference between revisions of "2014 Project Week:Processing Pipelines"

From NAMIC Wiki
Jump to: navigation, search
(Created page with '__NOTOC__ <gallery> Image:PW-SLC2014.png|Projects List Image:Pipeline_visualization_sketch.png‎| Visualization of pipeline from code </ga…')
 
 
Line 17: Line 17:
 
We have developed simple, very lightweight software for building processing pipelines: <br />
 
We have developed simple, very lightweight software for building processing pipelines: <br />
 
https://www.github.com/rameshvs/medical-imaging-pipelines <br />
 
https://www.github.com/rameshvs/medical-imaging-pipelines <br />
At the project week, we plan to finish building basic visualization of pipelines and a framework for making pipeline output easily visualizable in tools like Slicer.
+
At the project week, we plan to finish building basic visualization of pipelines and a framework for making pipeline output easily visualizable in tools like Slicer. <br />
 +
This system is meant to be a more lightweight alternative to existing tools such as nipype (http://nipy.org/nipype/): nipype provides a broad range of features but requires learning the intricacies of its system. Our system is meant to provide the core subset of those features, but in turn will be much simpler to learn, use, and work with.
 
</div>
 
</div>
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<div style="width: 27%; float: left; padding-right: 3%;">
Line 27: Line 28:
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<h3>Progress</h3>
 
<h3>Progress</h3>
 +
* Built better visualization (see screenshot)
 +
* Had many productive discussions with ideas for pipeline
 +
* Started data provenance tracking system and partial script execution
 
</div>
 
</div>
 
</div>
 
</div>

Latest revision as of 16:01, 10 January 2014

Home < 2014 Project Week:Processing Pipelines


Key Investigators

  • Ramesh Sridharan, Adrian Dalca, Polina Golland, MIT

Project Description

Objective

We have developed simple, very lightweight software for building processing pipelines:
https://www.github.com/rameshvs/medical-imaging-pipelines
At the project week, we plan to finish building basic visualization of pipelines and a framework for making pipeline output easily visualizable in tools like Slicer.
This system is meant to be a more lightweight alternative to existing tools such as nipype (http://nipy.org/nipype/): nipype provides a broad range of features but requires learning the intricacies of its system. Our system is meant to provide the core subset of those features, but in turn will be much simpler to learn, use, and work with.

Approach, Plan

  • Implement visualization (see screenshot)
  • Discuss wishlist for pipeline (come see me if you have any thoughts!)
  • Discuss MRML/MRB output for visualizing results in Slicer

Progress

  • Built better visualization (see screenshot)
  • Had many productive discussions with ideas for pipeline
  • Started data provenance tracking system and partial script execution