• Embedded data management
• Spatio-temporal databases
• Mobile sensor data management

BDSTIC / MSIS – Moving objects and Sensor Information System (in collaboration with INRETS-LIVIC)

With the continuously advances in mobile devices, positioning technologies, geo-location sensors or general data sensors, moving objects databases have been an important research subject in the last years. Existing works focus on managing object motion. However, a moving object can produce additional data flows that describe the object state and that are of major interest for some applications like analyzing driving behavior or reconstructing the circumstances of an accident. This project aims at providing a framework for handling moving objects, temporal sensor data, and the underlying network infrastructure as well as their interaction.

RHEA / CIREA: Mining International Codes of Disease From Hospital Reports

In France and many other countries, it is a legal obligation to supply ICD (International Classification of Diseases) codes whenever a patient treatment reaches its end. Some hospitals employ whole-time members of the medical profession to fulfill this task. The purpose is triple: (1) basis for appropriation assignation, considering these codes represent the unit activity, (2) authorizations for nationwide and international epidemiological studies, thanks to a systematic and standardized coding of pathologies, and (3) factor for calculating indicators of the quality of treatments, such as the number of nosocomial infections, which justifies that the quality of coding itself has been proposed as a quality indicator. In the last years, machine learning approaches have demonstrated some success for the construction of automatic document categorizers. The supervised learning represents a way to help to codify hospital records. This is precisely the goal of an essential part of the project RNTS Rhea.