Where: Imagerie et Vision Artificielle (ImViA) laboratory, Dijon, France Project description Person specification Application Closing date: 31st December 2021
Duration of the internship: six months
Stipend: about 590 € per month
Supervising team: Benoit Presles, PhD; Jean-Marc Vrigneaud, PhD
For two years now, our team has been involved in the development of a Gate model representing a digital Positron Emission Tomography (PET) system. PET is a gold standard for the evaluation of tumour metabolism but a specific mode of PET acquisition, namely dynamic PET imaging, is promising for the concurrent evaluation of tumour perfusion and metabolism. However, PET dynamic images are inherently noisier than static images. Gate is a well-known open-source platform which enables the simulation of complete PET geometries and acquisitions. As a previous work confirmed the ability of Gate to produce dynamic images, we are currently developing this aspect on a larger scale. The aim of this internship is to contribute to the building of a complete database of simulated dynamic PET images in the context of breast cancer. As such, the student will have to work on the following tasks: segment clinical images, assign kinetic modelling to the segmented structures, perform dynamic simulations in Gate and reconstruct pairs of pristine (from trues sinograms) and PET-like images (from prompt sinograms) using a proprietary toolbox. The ultimate goal of this database of coupled images is to train a deep learning algorithm for denoising.
Candidates must be engineering or master students in a relevant field such as physics, biomedical engineering, computer science, or applied mathematics.
Applications (including a CV and a covering letter outlining your motivation for the position) should be sent to Benoit Presles (benoit.presles at u-bourgogne.fr) and Jean-Marc Vrigneaud (jmvrigneaud at cgfl.fr).
Where: Imagerie et Vision Artificielle (ImViA) laboratory, Dijon, France
Closing date: 31st December 2021