Hes-so Valais


AISLab - Applied Intelligent Systems Lab

The Applied Intelligent Systems Lab (AISLab) was founded in 2009. It is part of the Institute of Information Systems, University of Applied Sciences and Arts Western Switzerland (HES-SO). Its goal is the use of intelligent techniques for solving concrete problems, mainly in eHealth. Its methodology is bottom-up, thus starting from real requirements and trying to find the best available techniques to solve them. From this experience, we are defining concepts & techniques especially in the areas of intelligent agents.

Intelligent agents for Personalized Medicine:

AISLab is applying intelligent agent technology to pervasive healthcare applications. For that, we are researching on multi-agent middleware, and are using it for the intelligentmonitoring of chronic patients as pervasive healthcare application.

We are also developping new multi-parametric data analysis techniques that we apply to diabetes type 1 and 2, and drug monitoring for Personalized Medicine.

We are enhancing software for General Practicioners with Decision Support functionalities.

Interoperability technologies in eHealth:

AISLab is working on the urbanization of information systems and interoperability techniques, especially in the area of ehealth, with a focus on interoperability of electronic health records. The ultimate goal is to permit medical actors and patients access medical data at any time and from any location (ubiquitous access to medical information).

We have developed several interoperability platforms for exchanging medical data. For this, we base on medical standards and Integrating the Healthcare Enterprise (IHE)profiles.

Middleware, Semantic Web and Simulations:

We are generally interested in research in middleware, with different focus points. Additionally to our teaching of Java Enterprise technologies, we want to investigate in the future the integration of agent technology in standard middleware application servers.

We are currently using the agent platform GOLEM, which is built with agents, avatars, objects and containers. In GOLEM, agents are active cognitive entities with a body to perceive and produce events and a declarative mind capable of symbolic reasoning and decision making. Avatars are agents where the decision making is done by a human user via an interface.

Furthermore, the combination of Web Services and the Semantic Web can lead to a new technology infrastructure called semantic Web services, where Web services are described with the semantic Web. The aim of this research is to study how web services from distinct providers can be composed into workflows in order to solve concrete user requirements. In this context, we have developed an open source distributed directory system of semantic Web services, called WSDir (work at EPFL).

We are also working on a semantic coordination infrastructure, which allows storing metadata of electronic health records in federated nodes.

In an industrial collaboration, we are currently developing traceability solutions for the production of bio-textiles. The data is described using a knowledge server using semantic Web technologies.

Finally, we have realized agent-based micro-simulations, applying them to the simulation of the Swiss highway traffic (work at EPFL), the prediction of snow height in ski resorts, and the optimization of skier flows.