Nicolas Singer

An adaptive system for home monitoring using a multiagent classification of patterns

This research takes place in the S(MA)2D project which proposes software architecture to monitor elderly people in their own homes. We want to build patterns dynamically from data about activity, movements, and physiological information of the monitored people. To achieve that, we propose a multiagent method of classification: every agent has a simple know-how of classification. Data generated at this local level are communicated and adjusted between agents to obtain a set of patterns. The patterns are used at a personal level, for example to raise an alert, but also to evaluate global risks (epidemic, heat wave). These data are dynamic; the system has to maintain the built patterns and has to create new patterns. So, the system is adaptive and can be spread on a large scale.

A sustainable software architecture for home care monitoring applications

The Ageing of population is a major concern for Western societies and leads to the development of new solutions to improve home care for elders, in order to delay their admission in specialized institutions (retirement house, healthcare facility and so on). These new solutions can be hardware or software based, and most often rely on home automation (e.g. motion sensors, temperature, light...). These sensors are used to monitor elderly or disabled people in order to detect their activities and the potential accidents that may occur. In this paper, we present a software architecture based on interoperable components for home care solutions. This architecture considers two kinds of components: the data providers, for instance a motion sensor, and the data consumers that process sensor data in order to infer higher level information such as a fall detector. The overall architecture is loosely coupled by design, in order to ease the addition of new sensors and of new functionalities.

Design of a serious game in training non-clinical skills for professionals in health care area

Most educational games and training applications for health care professionals have been developed as simulation tools dedicated to the teaching of medical knowledge in a particular area. Non-clinical skills such as communication skills or knowledge about e-Health are insufficiently focused by such tools. A serious game with consistent educational objectives offers to the learner many possibilities to acquire multiple competences in a fun and engaging learning process. This paper presents a serious game composed with extensible educational modules that concentrate on providing high-quality health care knowledge. It is designed to respect the balance between serious and fun in both educational and game elements. The proposed architecture allows the learning objective to be clearly defined and facilitate the collaborations of actors involved in the development. A prototype of the communication skills module is presented as an example of a module design.

Meta-Monitoring Using an Adaptive Agent-Based System to Support Dependent People in Place

In this paper, the authors propose software architecture to monitor elderly or dependent people in their own house. Many studies have been done on hardware aspects resulting in operational products, but there is a lack of adaptive algorithms to handle all the data generated by these products due to data being distributed and heterogeneous in a large scale environment. The authors propose a multi-agent classification method to collect and to aggregate data about activity, movements, and physiological information of the monitored people. Data generated at this local level are communicated and adjusted between agents to obtain a set of patterns. This data is dynamic; the system has to store the built patterns and has to create new patterns when new data is available. Therefore, the system is adaptive and can be spread on a large scale. Generated data is used at a local level, for example to raise an alert, but also to evaluate global risks. This paper presents specification choices and the massively multi-agent architecture that was developed; an example with a sample of ten dependant people gives an illustration.

Distributed classification: architecture and cooperation protocol in a multi-agent system for e-health

We propose an agent-based architecture to build classifications from evolutionary and distributed data. In the process, each agent builds a partial classification based on its data and a complete classification is then constructed from the partial ones by comparing them. Agents can collaborate by employing a restricted cooperation protocol. We present an application in the e-health domain, where global behavior patterns are built from activity or physiological data related to elderly people monitored at home by a variety of sensors. A step-by-step example illustrates our proposition.

A Serious Game Engine for Interview Simulation: Application to the Development of Doctor-Patient Communication Skills

In this paper we present the architecture of a conversation engine aimed to simulate an interview process between a human and a computer player. This component is a central element of many serious games where educational goal is to develop player communication skills. We demonstrate the use of our engine in AgileDoctor, a serious game project for training medical students and general practitioners to communicate with their patients, so as to improve their long-term relationship and provide a higher quality health care. Our proposed conversation engine uses a generic method to combine the game scenario and the educational objectives. The game scenario is described by an instance of a model that formalizes the general doctor-patient interview process and the skills to develop. The conversation engine is able to use this model to engage a challenging dialogue with a human player where missing skills are focused. The proposed design methodology is not bound to the health domain and is transferable to a large range of educational usages.

Prise en charge des personnes vieillissantes à leur domicile. Monitoring et classification

Le maintien a domicile des personnes âgees est devenu un enjeu important de sante publique. Nos travaux sont destines a traiter a grande echelle l’information recueillie au domicile des personnes pour en faire une information consolidee decrivant des comportements globaux. Nous proposons pour ce faire des algorithmes de classification pour identifier des profils collectifs et y rattacher les personnes suivies. Ces algorithmes tirent profit des technologies multi-agents afin de gerer l’heterogeneite des equipements et des services produisant l’information ainsi que leur totale distribution. Ils sont resistants aux deficits d’information pouvant provenir de pannes ou d’interruption de la tele-vigilance, et sont suffisamment anonymes pour assurer les besoins de confidentialite des informations echangees. Les profils obtenus sont utilises pour estimer l’etat des personnes âgees et les efforts a deployer pour qu’elles puissent continuer a vivre a leur domicile, et a un niveau plus global pour evaluer des tendances sanitaires generales (epidemie, fortes chaleurs, etc.).