Projects

IA et données de santé (DRIAS)

CNRS (mission MITI), 2020

Artificial intelligence technologies are spreading in the fields of health, and it raises new questions about the legal responsibilities of public persons. The objective of this transdisciplinary project is to investigate the interactions between artificial intelligence (automatic and supported decisions, machine learning and big data analytics), health and administrative law.

Research Topics

• Artificial intelligence
• Machine learning
• Administrative responsabilities, administrative law
• Health data

Project members

• IRISA/Inria (Teams LACODAM and DRUID)
• IDPSP
• Centre Borely
• CERSA
• CHU Rennes & Brest
• EHESP

contact
webpage

Pharmaco-Epidemiology of Health Products (PEPS)

Funded by ANSM, 2015-2018

The objective of this project is the development of a platform for digital pharmacoepidemiology. Pharmacoepidemiology is the study of the uses and effect of drugs in real conditions. The PEPS project aims at developping tools to analyse the SNIIRAM database to conduct pharmacoepidemiological studies.

Research Topics

• Query-based interactive sequential pattern mining toolbox
• Domain-specific query language
• New sequential pattern mining operators (discriminant chronicles)
• Application of data mining tools to pharmacoepidemiology

Project members

• CHU Rennes & Brest
• EHESP
• LTSI
• IRT B<>Com
• IRISA/Inria (Teams LACODAM, DRUID and DYLISS)

add

Publications

• Erwan Drezen, Thomas Guyet, André Happe. From medico-administrative databases analysis to care trajectories analytics: an example with the French SNDS. Fundamental and Clinical Pharmacology, Wiley, 2017, hal-01631802
• Yann Dauxais, Thomas Guyet, David Gross-Amblard, André Happe. Discriminant chronicles mining: Application to care pathways analytics. Artificial Intelligence in Medicine, Jun 2017, Vienna, Austria. 2017, 16th Conference on Artificial Intelligence in Medicine. hal-01568929
• Thomas Guyet, André Happe, Yann Dauxais. Declarative Sequential Pattern Mining of Care Pathways. Conference on Artificial Intelligence in Medicine in Europe, Jun 2017, Vienna, Austria. 24, pp.1161 – 266, 2017, 16th Conference on Artificial Intelligence in Medicine., hal-01569023

HYPTSER: Hybrid-models for Time-Series Prediction

Funded by PGMO, 2017-2018

The objective of this project is to develop hybrid machine-learning models for time series prediction. It is applied to the prediction of KPI of cloud infrastructures.

Research Topics

• Time-series prediction
• Machine learning

Project members

• Orange Labs, Lannion
• Inria (Teams LACODAM and LinkMedia)

add

ITRAMI

2014-2017

Research Topics

• User preferences
• New pattern mining operators : skypatterns
• Data visualization

Project members

• LIG
• ST Microelectronics

add

SePaDec: Sequential Pattern Mining with Declarative Approaches

Funded by PEPS project and Region Bretagne, 2016-2017

Research Topics

• Declarative pattern mining
• Answer Set Programming
• Pharmaco-epidemiology

Project members

• LACODAM

add

Publications

• Ahmed Samet, Thomas Guyet, Benjamin Negrevergne. Mining rare sequential patterns with ASP. ILP 2017 – 27th International Conference on Inductive Logic Programming, 2017, hal-01569582
• Ahmed Samet, Thomas Guyet, Benjamin Negrevergne, Tien-Tuan Dao, Tuan Nha Hoang, et al.. Expert Opinion Extraction from a Biomedical Database. Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty (ECSQARU), Jul 2017, Lugano, Switzerland. Springer, 31 (LNCS 10369), pp.1 – 12, 2017, Proceedings of 14th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty hal-01584984