joel colloc
Joel Colloc at his medical studies, earned a M.S. of Information Technologies (IT) at the Business School of Lyon (IAE of Lyon). He completed his M.D. at the medical faculty of Lyon and a specialty degree of forensic medicine with a degree of clinical toxicology. He received another M.S. in computer sciences at the engineering school National Institute of Applied Sciences (INSA) of Lyon. He served as forensic physician at the Edouard Herriot Hospital in the neurological emergency department to cure drug addicted people, medical ethics, toxicology, drugs, addiction database. He went on to earn his Ph.D. in computer sciences by the INSA of Lyon. He became assistant professor at the laboratory of medical computer science and taught IT at the medical faculty. He was elected as associate professor at IAE of Lyon and he earned his accreditation to supervise researches in sciences at the Lyon 1 and he is professor in computer sciences with tenure at Le Havre Normandy university since 2003. His main research topics concern e-health and particularly: artificial intelligence, fuzzy logic, multi-agent clinical decision support systems (MADSS) and knowledge bases, Case Base Reasoning, ontologies, nervous system modeling and cognitive sciences and their applications. His human sciences researches try to conciliate the ethics of using Big Data in epidemiological studies, autonomous systems and robots and keeping ethics use of AI in order to improve clinical decision in medicine while preserving the patient-caregiver relationship, the privacy and the freewill choice of the patients.
less
Related Authors
InterestsView All (12)
Uploads
Papers by joel colloc
ABSTRACT. Decision Support Systems combine different kinds of knowledge and data acquired from different sources in a supervised or an unsupervised way and represented with qualitative based on the expert knowledge and the deductive reasoning or quantitative models relying on stochastic methods. The case based reasoning memorizes and restores the experience of problem solving. The implementation needs languages and various concepts (algorithms, distances, evaluation functions, logic rules, databases…). In this context a multi-agent system (MAS) offers good integration capabilities. We propose a cognitive agent type which is the system keystone and includes the necessary modules to provide agents with autonomy, reasoning, communication and interaction features. It is then specialised to adapt it to the variety of the knowledge models. The agent architecture, its functionalities and the development of the modules are then described. The approach is usable to build decision support MAS in many domains and is illustrated through a medical example.
Affirmation paradoxale au moment où la médecine est considérée comme un domaine type de l'application des règles alors qu'elles paraissent mal adaptées aux cas suivants : Les problèmes liés à la représentation de signes cliniques utilisant des valeurs continues, l'aspect dynamique des naladies (évolution et disparition des signes), les boucles de raisonnement générées parfois lors du déclenchement des règles.
Actuellement s'affrontent deux types de procédés : le premier utilisant les règles de production, basé sur les systèmes formels, le second représentant la connaissance sous la forme d'objets complexes hiérarchisés.
One première partie définit les contraintes liées à la nature du raisonnement médical, la deuxième partie recherche un aode de représentation des connaissances mieux adapté.Nous montrerons que modèles sémantiques de données ou langages typés apportent des éléments de solution. La troisième partie présente le système expert orienté objet que nous proposons et la dernière en décrit l'architecture.