Difference between revisions of "2012 Winter Project Week:GraphbasedSeg"
From NAMIC Wiki
Ktdiedrich (talk | contribs) |
|||
| (10 intermediate revisions by 2 users not shown) | |||
| Line 1: | Line 1: | ||
| − | |||
<gallery> | <gallery> | ||
Image:PW-SLC2012.png|[[2012_Winter_Project_Week#Projects|Projects List]] | Image:PW-SLC2012.png|[[2012_Winter_Project_Week#Projects|Projects List]] | ||
| − | |||
</gallery> | </gallery> | ||
| Line 23: | Line 21: | ||
* A graph based technique is used that is capable of segmenting multiple surfaces simultaneously and in an optimal manner. | * A graph based technique is used that is capable of segmenting multiple surfaces simultaneously and in an optimal manner. | ||
| − | * Our plan for the project week is | + | * Our plan for the project week is to design a cost function that could effectively work on real data including left atrium and its surrounding wall. |
</div> | </div> | ||
| − | <div style="width: | + | <div style="width: 97%; float: left;"> |
<h3>Progress</h3> | <h3>Progress</h3> | ||
| − | * So far, the algorithm has been tested successfully on the 2D synthetic data. | + | * So far, the algorithm has been tested successfully on the 2D synthetic data. |
| + | * A model using image profile has been designed and cost functions based on that have been induced to optimally segment the left atrium and its surrounding wall. | ||
| + | <ul> | ||
| + | <li>[[File:multilineblurnoise_scimg.png]]</li> | ||
| + | <li>[[File:multilineblur_scimgop.png]]</li> | ||
| + | <li>[[File:Slice20_seg.png]]</li> | ||
| + | <li>[[File:Slice20_segzoom.png]]</li> | ||
| + | </ul> | ||
| + | |||
</div> | </div> | ||
</div> | </div> | ||
| − | |||
| − | |||
Latest revision as of 16:13, 13 January 2012
Home < 2012 Winter Project Week:GraphbasedSeg
Key Partners
- Utah: Gopal Veni, Ross Whitaker
Objective
- Optimal segmentation of the left atrium and its surrounding tissue simultaneously on 3D LGE-MRI images.
Approach, Plan
- A graph based technique is used that is capable of segmenting multiple surfaces simultaneously and in an optimal manner.
- Our plan for the project week is to design a cost function that could effectively work on real data including left atrium and its surrounding wall.
Progress
- So far, the algorithm has been tested successfully on the 2D synthetic data.
- A model using image profile has been designed and cost functions based on that have been induced to optimally segment the left atrium and its surrounding wall.
