Difference between revisions of "2017 Winter Project Week/ProstateSectorSegmentation"

From NAMIC Wiki
Jump to: navigation, search
(Created page with "__NOTOC__ <gallery> Image:PW-Winter2017.png|link=2017_Winter_Project_Week#Projects|Projects List <!-- Use the "Upload file" link on the l...")
 
 
(7 intermediate revisions by 2 users not shown)
Line 3: Line 3:
 
Image:PW-Winter2017.png|link=2017_Winter_Project_Week#Projects|[[2017_Winter_Project_Week#Projects|Projects List]]
 
Image:PW-Winter2017.png|link=2017_Winter_Project_Week#Projects|[[2017_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" -->
 
<!-- Use the "Upload file" link on the left and then add a line to this list like "File:MyAlgorithmScreenshot.png" -->
 +
File:Prostate_MRI.PNG|Transversal slice of Prostate MRI.
 +
File:SectorMap_Prostate.PNG|Sector Map according to PI-RADS v2.
 +
File:ManualSegmentations.PNG|Manual Segmentation on transversal slices.
 
</gallery>
 
</gallery>
  
Line 9: Line 12:
 
* Anneke Meyer, University of Magdeburg (Germany)
 
* Anneke Meyer, University of Magdeburg (Germany)
 
* Andrey Fedorov, BWH
 
* Andrey Fedorov, BWH
 +
* Alireza Mehrtash, BWH
 +
* Christian Hansen, University of Magdeburg (Germany)
 +
* Teodora Szasz, University of Chicago
  
 
==Project Description==
 
==Project Description==
Line 24: Line 30:
 
* Generation/ Refinement of ground truth data
 
* Generation/ Refinement of ground truth data
 
* Creation of a 3D sector model  
 
* Creation of a 3D sector model  
−
* Initialization of segmentation with user interaction or atlas-based segmentation (in order to decrease search space)
+
* Test with ProstateX challenge dataset (where zones of findings are given) to predict the zone with Deep Learning
−
* Try (model-based) segmentation approach (costs for segmentation optimization can be derived for example from supervised classification of the gland tissue). The shape of individual sector models could be used as segmentation prior
+
* if more training data is available: deep learning for an automatic sector segmentation
−
* if more training data is available: deep learning for a better cost generation or for an automatic sector segmentation
+
* Try atlas-based segmentation approach to map zone segmentations on more datasets
 +
 
 +
 
  
 
|
 
|
 
<!-- 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) -->
−
*
+
* Creation of 3 zone segmentations of the prostate (different volumes)
 +
* Prediction of zones for specific locations in ProstateX dataset with Deep Learning is in progress
 +
* Next Steps: Create 3D model with higher resolution. Atlas based segmentation on more publicly available data.
 
|}
 
|}
  
 
==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 -->
 
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->

Latest revision as of 15:52, 13 January 2017

Home < 2017 Winter Project Week < ProstateSectorSegmentation

  • Transversal slice of Prostate MRI.

  • Sector Map according to PI-RADS v2.

  • Manual Segmentation on transversal slices.

Key Investigators

  • Anneke Meyer, University of Magdeburg (Germany)
  • Andrey Fedorov, BWH
  • Alireza Mehrtash, BWH
  • Christian Hansen, University of Magdeburg (Germany)
  • Teodora Szasz, University of Chicago

Project Description

Objective Approach and Plan Progress and Next Steps
  • Segmentation of prostate and its sectors
  • Specifically, segmentation of the following prostate sectors: peripheral zones, transition zones, central zone, anterior fibromuscular stroma and urethral sphincter
  • Generation/ Refinement of ground truth data
  • Creation of a 3D sector model
  • Test with ProstateX challenge dataset (where zones of findings are given) to predict the zone with Deep Learning
  • if more training data is available: deep learning for an automatic sector segmentation
  • Try atlas-based segmentation approach to map zone segmentations on more datasets
  • Creation of 3 zone segmentations of the prostate (different volumes)
  • Prediction of zones for specific locations in ProstateX dataset with Deep Learning is in progress
  • Next Steps: Create 3D model with higher resolution. Atlas based segmentation on more publicly available data.

Background and References

Retrieved from "https://www.na-mic.org/w/index.php?title=2017_Winter_Project_Week/ProstateSectorSegmentation&oldid=95106"