Société Française de Bioinformatique

Mots-Clés

deep learning data processing psychiatry medical data

Description

MyndBlue
MyndBlue is a biomedical research company developing AI algorithms for mental health. Our goal is to achieve diagnosis and prognosis based on proprietary data collected during clinical studies. Our focus is on 2 overlapping mental health diseases: major depressive disorder and post-traumatic stress disorder with a broader intent on cycling diseases. For our research, we are collaborating with academics in France and the US in machine learning, psychiatry, medicine to insure a state of the art level with the best impact possible for the patients.

Job Overview:
- Ensure a scientific watch in signal processing, deep learning, and AI in Medicine and Psychiatry
- Propose and develop algorithmic solutions for medical data
- Work and develop deep learning algorithms for prediction and forecast, with a focus on time- series data
- Be involved in the integration of algorithmic prototypes to robust, optimized production code
- Communicate the results and methodologies to our inter-disciplinary team
- Participate to scientific communications (grant applications, reports, peer-reviewed publications) by writing, graphical content and slides

Your profile and skills:
- Ph.D. in Maths, Informatics, Physics, Bioinformatics, or a related field
- Post-doctoral experience in academic or private research
- Knowledge of Unix environment and GPU functioning
- Proficiency of algorithms of machine learning and deep learning, both supervised and unsupervised. Experience with time-series data would be a plus.
- High proficiency of programming language Python and Tensorflow, or equivalent
- Usage of git-versioning and team programming
- Good at analyzing and summarizing complex scientific pipelines and results
- Innovative and independent in the research exploration
- Capable of learning on the job (state of the art data analysis methods, but also in the fields of psychiatry, pharmacology, ...)
- Very good English level