The objective of this project is to significantly reduce the computational time of “real time” numerical simulations like those used for per-operative assistance / guidance in image guided telesurgery.
These simulations often rely on simplifying assumtions and approximations to optimize computational costs. These assumptions, usually acceptable in generic simulations become inacceptable when trying to specifically reproduce the behavior of a patient’s organ in the context of per-operative guidance for instance.
The applicative context will be provided by the MIMESIS team, specialized in medical numerical simulations.
To be able to accelerate computing times while minimizing code rewriting, we suggest to use the techniques and tools resulting from CAMUS‘ team research work. In particular, we will use the results from automated and semi-automatited parallelization programs and distributed asynchronous software development.
In collaboration with researchers and engineers from the CAMUS and MIMESIS teams, the engineer will take charge of the tools developed by CAMUS for software parallelization and optimization in one hand, and the SOFA Framework co-developed by MIMESIS on the second hand.
Furthermore, he / she will look into solutions to parallelize SOFA’s simulation loop, and optimize a determined subset of simulations developed by researchers.
These results could lead to scientific publications to which the engineer will be associated.
Holding a Master’s degree in computer science, or being a confirmed engineer with 2 to 7 years of engineering experience, we expecte the candidate to:
Mastering multi-core and CPU computing, and profiling techniques would be a plus.
contract duration : 24 months.
Gross Salary : 2600 à 3200 € depending on experience
Projected starting date : from the 10th of January 2016.
Offer is valid until the 3rd of December 2016.