Difference between revisions of "2016 Winter Project Week/Projects/CongenitalHeart"

From NAMIC Wiki
Jump to: navigation, search
 
(6 intermediate revisions by the same user not shown)
Line 2: Line 2:
 
<gallery>
 
<gallery>
 
Image:PW-MIT2016.png|link=2016_Winter_Project_Week#Projects|[[2016_Winter_Project_Week#Projects|Projects List]]
 
Image:PW-MIT2016.png|link=2016_Winter_Project_Week#Projects|[[2016_Winter_Project_Week#Projects|Projects List]]
<!-- Use the "Upload file" link on the left and then add a line to this list like "File:MyAlgorithmScreenshot.png" -->
+
Image:congenitalMICCAI15.png|Results after user manually segments 3, 8 or 14 slices
 
</gallery>
 
</gallery>
  
Line 11: Line 11:
  
 
==Project Description==
 
==Project Description==
 +
 +
* We have developed an interactive segmentation algorithm for cardiac MR images from patients with severe congenital heart disease, in which the user manually segments a few short-axis slices and the remainder is segmented automatically using patch-based segmentation (Pace et al, MICCAI 2015)
 +
 
{| class="wikitable"
 
{| class="wikitable"
 
! style="text-align: left; width:27%" |  Objective
 
! style="text-align: left; width:27%" |  Objective
Line 18: Line 21:
 
|
 
|
 
<!-- Objective bullet points -->
 
<!-- Objective bullet points -->
*  
+
* Continue improving the method, especially focusing on correctly segmenting thin walls of the heart when few slices are manually segmented
 
|
 
|
 
<!-- Approach and Plan bullet points -->
 
<!-- Approach and Plan bullet points -->
*  
+
* Continue investigating algorithm modifications to enhance / correctly segment the thin walls.
 
|
 
|
 
<!-- Progress and Next steps bullet points (fill out at the end of project week -->
 
<!-- Progress and Next steps bullet points (fill out at the end of project week -->
*
+
* Worked on integrating new "thin wall" features to improve thin wall detection
 
|}
 
|}
  
 
==Background and References==
 
==Background and References==
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->
+
* [http://link.springer.com/chapter/10.1007/978-3-319-24574-4_10 Pace et al,, Interactive Whole-Heart Segmentation in Congenital Heart Disease, MICCAI 2015]

Latest revision as of 15:01, 8 January 2016

Home < 2016 Winter Project Week < Projects < CongenitalHeart

Key Investigators

  • Danielle Pace (MIT)
  • Adrian Dalca (MIT)
  • Polina Golland (MIT)

Project Description

  • We have developed an interactive segmentation algorithm for cardiac MR images from patients with severe congenital heart disease, in which the user manually segments a few short-axis slices and the remainder is segmented automatically using patch-based segmentation (Pace et al, MICCAI 2015)
Objective Approach and Plan Progress and Next Steps
  • Continue improving the method, especially focusing on correctly segmenting thin walls of the heart when few slices are manually segmented
  • Continue investigating algorithm modifications to enhance / correctly segment the thin walls.
  • Worked on integrating new "thin wall" features to improve thin wall detection

Background and References