POST-DOC/RESEARCH ENGINEER IN MEDICAL DATA PROCESSING (REIMS)
4 Octobre 2021
Catégorie : Post-doctorant
Statistic expert/Data processing
Duration: 12/24 mois
Ratio: full time
Salary: with respect to degree/experience
Person in charge: Nicolas Passat, Guillaume Dollé, Stéphanie Salmon
Email: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
University: Université de Reims Champagne-Ardenne
The recruited person will be in charge to: develop new algorithms and digital methods; participate in the optimization and development of the project’s digital tools; contribute to the porting of codes on the supercomputers made available; participate in the dissemination of knowledge through conferences, articles, training sessions around the tools developed.
• Develop data processing and analysis methods (biomarkers, EEG signals)
• Develop digital calculation and visualization tools
• Exploit the ROMEO supercomputer
• Collaborate closely with the neonatology service of the CHU de Reims
• Participate in the dissemination of knowledge
• Expertise in statistics and data analysis
• Computer skills and programming languages (Python, C/C++, R)
• Knowledge of signal processing
• Knowledge of machine learning methods (GAN, autoencoders, . . . )
• Knowledge of parallelism CPU/GPU (MPI, openmp, cuda/opencl)
• Correct level in English
• Ability to work in a team and independently
• Ability to communicate
The recruited person will be integrated within the framework of a scientific project carried out in partnership between the Reims Mathematics Laboratory (LMR) UMR CNRS 9008, the CReSTIC EA 3804, and the neonatology service at the University Hospital of Reims. This project is focused on the issue of data processing from EEG / aEEG signals, biological data and MRI images for the newborn. It is funded by the National Research Agency and the American Memorial Hospital Foundation.
In this context, the work will consist more precisely in processing and analyzing data from a ancillary study of a cohort of approximately 800 term newborns as part of the LyTONEPAL study one of the objectives of which is to study the predictive factors of unfavorable outcome (neuropathologies, disorders psychomotor) at 3 years.
The work will consist in analyzing predefined biomarkers which will be confronted with characteristics extracted from standard EEG signals using statistical and machine learning tools. In parallel, it will also be a question of providing treatment and visualization tools adapted to clinical research to determine the neuroprotection measures to put in place, in particular for the management pre-hospital anoxo-ischemic encephalopathies.