Difference between revisions of "2017 Winter Project Week/Multi-ModalitySegmentationOfUSandMRImagesForGliomaSurgery"

From NAMIC Wiki
Jump to: navigation, search
 
(5 intermediate revisions by the same user not shown)
Line 4: Line 4:
 
<!-- 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:USandMRIimagesMLSegmentation.png|Several structures of the brain segmented from US- and MR images.
 
File:USandMRIimagesMLSegmentation.png|Several structures of the brain segmented from US- and MR images.
 +
File:Dnn.png
 +
File:SkullStrip.tcsproj.mp4
 
</gallery>
 
</gallery>
  
Line 9: Line 11:
 
<!-- Add a bulleted list of investigators and their institutions here -->
 
<!-- Add a bulleted list of investigators and their institutions here -->
 
* Jennifer Nitsch, University of Bremen (Germany)
 
* Jennifer Nitsch, University of Bremen (Germany)
 +
* Hans Meine, University of Bremen (Germany)
  
  
Line 19: Line 22:
 
|
 
|
 
<!-- Objective bullet points -->
 
<!-- Objective bullet points -->
* Multi-Modual Image Segmentation of preoperative MR- and intraoperative Ultrasound(US)-images
+
* Multi-Modual Image Segmentation of preoperative MR- and intraoperative Ultrasound(US)-images for multi-modual image registration.
 
* Segmentation of the following anatomical structures: Falx cerebri, tentorium cerebelli, white matter, gray matter, CSF (ventricles), blood vessels.
 
* Segmentation of the following anatomical structures: Falx cerebri, tentorium cerebelli, white matter, gray matter, CSF (ventricles), blood vessels.
 +
* Testing applicability of Deep Learing on current data
 
|
 
|
 
<!-- Approach and Plan bullet points -->
 
<!-- Approach and Plan bullet points -->
 
* Generation/ Refinement of ground truth data
 
* Generation/ Refinement of ground truth data
 
* Generation/Creation/Refinement of characteristic (anatomical) landmarks (e.g. charactersitic representation of cerebellum, insula cortex (lateral sulcus), choroid plexus.   
 
* Generation/Creation/Refinement of characteristic (anatomical) landmarks (e.g. charactersitic representation of cerebellum, insula cortex (lateral sulcus), choroid plexus.   
 +
* Prepare images for Deep Learning and seeting up Deep learning infrastructure.
 
|
 
|
 
<!-- 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) -->
*
+
*Trained a DNN for skull stripping on 18 data sets from a public database, also segmented in GM, WM, CSF
 +
*There a a lot of databases for T1-weighted brain images, but good reference data is rare...
 
|}
 
|}
  
 
==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 14:37, 16 January 2017

Home < 2017 Winter Project Week < Multi-ModalitySegmentationOfUSandMRImagesForGliomaSurgery

Key Investigators

  • Jennifer Nitsch, University of Bremen (Germany)
  • Hans Meine, University of Bremen (Germany)


Project Description

Objective Approach and Plan Progress and Next Steps
  • Multi-Modual Image Segmentation of preoperative MR- and intraoperative Ultrasound(US)-images for multi-modual image registration.
  • Segmentation of the following anatomical structures: Falx cerebri, tentorium cerebelli, white matter, gray matter, CSF (ventricles), blood vessels.
  • Testing applicability of Deep Learing on current data
  • Generation/ Refinement of ground truth data
  • Generation/Creation/Refinement of characteristic (anatomical) landmarks (e.g. charactersitic representation of cerebellum, insula cortex (lateral sulcus), choroid plexus.
  • Prepare images for Deep Learning and seeting up Deep learning infrastructure.
  • Trained a DNN for skull stripping on 18 data sets from a public database, also segmented in GM, WM, CSF
  • There a a lot of databases for T1-weighted brain images, but good reference data is rare...

Background and References