Offre en lien avec l’Action/le Réseau : – — –/– — –

Laboratoire/Entreprise : Connected Health Lab (Ecole d’ingénieurs ISIS) Ca
Durée : 36
Contact : francis.faux@univ-jfc.fr
Date limite de publication : 2025-02-28

Contexte :
The aim of the project is to trace the toxic products “inhaled” by firefighters during their various interventions.

Sujet :
The first stage of the thesis will be to make the real-time acquisition system operational and reliable,
and to contextualize it according to the type of fire (apartment, forest).
The second objective of the thesis will be:
– to study hybrid online and multi-source learning models for modeling the toxicity of different types of fire, in order to infer the duration
of exposure to different toxic products (taking into account the randomness of the context)
– to develop a medical decision support tool under uncertainty to identify at-risk firefighter profiles.
Given a firefighter’s history, it will be possible to deduce the arguments that point to a certain level
of risk associated with the development of different diseases. To this end, work on Bipolar Layered argumentative
Frameworks could be adapted to temporal data and, if necessary, enriched.

Profil du candidat :
We are looking for a candidate with a strong AI background, particularly in machine learning. Knowledge of uncertainty modeling in AI will be highly appreciated.

Formation et compétences requises :
Master2

Adresse d’emploi :
Ecole d’ingénieurs ISIS, rue Firmin Oulès, 81100 Castres (france)

Document attaché : 202501061422_Thesis-AI_hybrid.pdf