Difference between revisions of "2013 Summer Project Week:Deformable registration validation toolkit"
From NAMIC Wiki
| (2 intermediate revisions by one other user not shown) | |||
| Line 2: | Line 2: | ||
<gallery> | <gallery> | ||
Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]] | Image:PW-MIT2013.png|[[2013_Summer_Project_Week#Projects|Projects List]] | ||
| + | Image:2013_Summer_Project_Week_Registration_Validation_1.jpg|SIFT features | ||
| + | Image:2013_Summer_Project_Week_Registration_Validation-VisAIRe.gif|VisAIRe registration validation extension in Slicer | ||
</gallery> | </gallery> | ||
| Line 25: | Line 27: | ||
</div> | </div> | ||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
| − | |||
<div style="width: 40%; float: left;"> | <div style="width: 40%; float: left;"> | ||
<h3>Progress</h3> | <h3>Progress</h3> | ||
| + | * Qualitative review method is very nice, but doesn't offer quantitative method. This method should be very useful for QA of many scans | ||
| + | * Quantitative method is required for algorithm tuning | ||
| + | * Paganelli's method only returned a small-ish number of features. Work is continuing to improve this. | ||
| + | * Toews' method returns many features, but in voxel coordinates, so cannot yet be reviewed. | ||
</div> | </div> | ||
Latest revision as of 03:14, 21 June 2013
Home < 2013 Summer Project Week:Deformable registration validation toolkit
SIFT features
VisAIRe registration validation extension in Slicer
Key Investigators
- MGH: Greg Sharp
- PMH, Toronto: Kevin Wang
- BWH: Andrey Fedorov
Objective
We are developing tools for helping users assess the accuracy of deformable image registration.
Approach, Plan
- Review Andrey Fedorov's qualitative registration review
- Review Chiara Paganelli's feature matching method
- Review Matthew Toews' feature matching method
Progress
- Qualitative review method is very nice, but doesn't offer quantitative method. This method should be very useful for QA of many scans
- Quantitative method is required for algorithm tuning
- Paganelli's method only returned a small-ish number of features. Work is continuing to improve this.
- Toews' method returns many features, but in voxel coordinates, so cannot yet be reviewed.
Delivery Mechanism
This work will be delivered to the NA-MIC Kit as:
- Extension -- python scripted
