Difference between revisions of "2014 Summer Project Week:Intelligent Steering"

From NAMIC Wiki
Jump to: navigation, search
 
(4 intermediate revisions by the same user not shown)
Line 2: Line 2:
 
<gallery>
 
<gallery>
 
Image:PW-MIT2014.png|[[2014_Summer_Project_Week#Projects|Projects List]]
 
Image:PW-MIT2014.png|[[2014_Summer_Project_Week#Projects|Projects List]]
Image:steered_rotations0_h.png | 2D slice views of fixed image
+
Image:Steered_rotations0_h.png| 2D slice views of fixed image using different PCA axis combinations
Image:steered_rotations1_h.png | 2D slice views of moving image
+
Image:Steered_rotations1_h.png| 2D slice views of moving image using different PCA axis combinations
 
</gallery>
 
</gallery>
  
Line 21: Line 21:
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<div style="width: 27%; float: left; padding-right: 3%;">
 
<h3>Approach, Plan</h3>
 
<h3>Approach, Plan</h3>
* Investigate data exploration techniques for highlighting rotational errors
+
* Investigate data exploration techniques for highlighting common registration errors (local rotation, shift, and scaling)
 
* Implement efficient, responsive extensions using the Slicer 4 Python extension platform and PyOpenCL
 
* Implement efficient, responsive extensions using the Slicer 4 Python extension platform and PyOpenCL
 
</div>
 
</div>

Latest revision as of 04:18, 27 June 2014

Home < 2014 Summer Project Week:Intelligent Steering

Key Investigators

  • Marcel Prastawa, GE
  • Jim Miller, GE
  • Steve Pieper, Isomics

Project Description

Steered image registration in Slicer using intelligent interaction rather than direct interaction where user specifies deformation changes.

Objective

  • Develop interaction schemes that allows users to indicate incorrect regions
  • Provide intelligent interfaces so users provide minimal interaction without detailed control of image deformations

Approach, Plan

  • Investigate data exploration techniques for highlighting common registration errors (local rotation, shift, and scaling)
  • Implement efficient, responsive extensions using the Slicer 4 Python extension platform and PyOpenCL

Progress

  • Testing the use of robust PCA for providing hints on rotational misalignment
  • Investigating the modeling of appropriate correction steps given the possible misalignments

References

Video of Direct Steering Interaction