PhD position (France): High-speed emission tomography reconstruction on GPU architecture
[b][CLOSED] PhD position: High-speed emission tomography reconstruction on GPU architecture[/b]

Emission Tomography (ET) is a type of nuclear medicine imaging that produces three-dimensional image of important body functions, such as blood flow, oxygen use, and sugar metabolism. ET is used to diagnose variety of disease like cancer, damaged heart tissue or brain disorders. Current ET algorithms do not allow fast reconstruction in 3D and certainly not if all necessary corrections are taken into account from a Monte Carlo modelling of interactions between particles and matter. A solution based on the use of computer clusters to solve the intensive computational issues is not realistic within the routine clinical environment. Graphics processing units (GPU) represent today a cheap solution to getting high computation power in medical imaging. However, few GPU adapted methods have been proposed within the ET reconstruction context. The aim of this project is precisely to propose new approaches to achieve a fully corrected ET reconstruction on GPU architecture. This thesis will be conducted in the different following sub-objectives (i) improve the ray-projector which plays a key-role in current line of response based ET reconstruction, (ii) develop new methods to achieve a regular reconstruction on a GPU architecture, (iii) incorporate corrections from a Monte Carlo simulation performed on the GPU.

This PhD position is open to the Laboratory of Medical Information Processing (LaTIM) in the French research unit U650 of INSERM (National Institute of Health & Medical Research). The LaTIM is a research team made up of the University Hospital of Brest, the Faculty of Medicine and Telecom Bretagne (graduate engineering school). Funding for this thesis is supported in part by the European project MEDIATE “patient friendly medical intervention” and in part by “Région Bretagne” (Brittany region). The position is located at Brest (France) and will start around September/October 2011 for a period of 3 years.

Applicants need to have a M.Sc. degree in the fields of physics, engineering or computer science. Good communication skills and excellent English are requested. Mastering the French language is a plus but not required. Expertise in the following field is desirable: GPU programming (CUDA), C/C++ programming, image processing or image reconstruction or medical imaging. Some others skills will be appreciated but not required: tomographic reconstruction, emission tomography, Python programming, and Monte Carlo simulation.

Please send your application (CV) to Julien Bert by email [email="julien.bert@univ-brest.fr"]here[/email]

[b]THIS POSITION IS CLOSED[/b]
[CLOSED] PhD position: High-speed emission tomography reconstruction on GPU architecture



Emission Tomography (ET) is a type of nuclear medicine imaging that produces three-dimensional image of important body functions, such as blood flow, oxygen use, and sugar metabolism. ET is used to diagnose variety of disease like cancer, damaged heart tissue or brain disorders. Current ET algorithms do not allow fast reconstruction in 3D and certainly not if all necessary corrections are taken into account from a Monte Carlo modelling of interactions between particles and matter. A solution based on the use of computer clusters to solve the intensive computational issues is not realistic within the routine clinical environment. Graphics processing units (GPU) represent today a cheap solution to getting high computation power in medical imaging. However, few GPU adapted methods have been proposed within the ET reconstruction context. The aim of this project is precisely to propose new approaches to achieve a fully corrected ET reconstruction on GPU architecture. This thesis will be conducted in the different following sub-objectives (i) improve the ray-projector which plays a key-role in current line of response based ET reconstruction, (ii) develop new methods to achieve a regular reconstruction on a GPU architecture, (iii) incorporate corrections from a Monte Carlo simulation performed on the GPU.



This PhD position is open to the Laboratory of Medical Information Processing (LaTIM) in the French research unit U650 of INSERM (National Institute of Health & Medical Research). The LaTIM is a research team made up of the University Hospital of Brest, the Faculty of Medicine and Telecom Bretagne (graduate engineering school). Funding for this thesis is supported in part by the European project MEDIATE “patient friendly medical intervention” and in part by “Région Bretagne” (Brittany region). The position is located at Brest (France) and will start around September/October 2011 for a period of 3 years.



Applicants need to have a M.Sc. degree in the fields of physics, engineering or computer science. Good communication skills and excellent English are requested. Mastering the French language is a plus but not required. Expertise in the following field is desirable: GPU programming (CUDA), C/C++ programming, image processing or image reconstruction or medical imaging. Some others skills will be appreciated but not required: tomographic reconstruction, emission tomography, Python programming, and Monte Carlo simulation.



Please send your application (CV) to Julien Bert by email [email="julien.bert@univ-brest.fr"]here[/email]



THIS POSITION IS CLOSED

#1
Posted 05/18/2011 02:37 PM   
Scroll To Top