View source for Project Week 25/CNN for PseudoCT Generation from T1T2 MR

__NOTOC__

Back to [[Project_Week_25#Projects|Projects List]]

==Key Investigators==
<!-- Key Investigator bullet points -->
*[http://www.imagenglab.com/pileggi_29.html Giampaolo Pileggi] (Magna Graecia University, Italy/German Cancer Research Center (DKFZ), Germany)
*[http://www.imagenglab.com/newsite/paolo_zaffino/ Paolo Zaffino] (Magna Graecia University, Italy)
*[http://www.imagenglab.com/newsite/salvatore_scaramuzzino/ Salvatore Scaramuzzino] (Magna Graecia University/ASL Vercelli, Italy)
*[http://www.imagenglab.com/newsite/mf_spadea/ Maria Francesca Spadea] (Magna Graecia University, Italy)
*[http://isgwww.cs.uni-magdeburg.de/cas/team.php Gino Gulamhussene] (University of Magdeburg, Germany)
*[http://isgwww.cs.uni-magdeburg.de/isg/meyer.html Anneke Meyer] (University of Magdeburg, Germany)

==Project Description==
This tool allows the user to generate an HU map (Pseudo-CT) from T1/T2 input MRI. The tool is still in the early development stages with good early results in terms of Mean Absolute Error and Bias.

{| class="wikitable"
! style="text-align: left; width:27%" |   Objective
! style="text-align: left; width:27%" |   Approach and Plan
! style="text-align: left; width:27%" |   Progress and Next Steps
|- style="vertical-align:top;"
|<!-- Objective bullet points -->
The main object is to understand which topology of CNN is the most suited for the Pseudo-CT generation task.
|<!-- Approach and Plan bullet points -->
Test of different structures with training on low-res images in order to speed-up computational time.
Analysis of the different output with MAE and Bias metrics
|<!-- Progress and Next steps (fill out at the end of project week), bullet points -->
The most suited CNN to be used for this task consisted of an edited U-Net that does not make use of the final Dense Layer, useful for segmentation but not for regression purposes.
Also unpooling layer was used in the decoding part, with zero-filling instead of repetitions for the missing values.
Next planned steps will be the fine tuning of the CNN in order to lower the MAE and BIAS, the translation to TensorFlow (now the code uses Theano as a backend) and the data parallelization. 
Interactions with Annette, Gino and Gabriele were important in order to understand how to improve the CNN topology. 
|}

==Illustrations==
[[File:MRI2CT.png]]

==Background and References==
<!-- Use this space for information that may help people better understand your project, like links to papers, source code, or data -->