 Abstract—Health monitoring at home could be an important element of care and support environment for older people. Diversity of diseases and different needs of users require universal design of a home platform. We present our work on a sensor-based multimodal platform that is trained to recognize the activities elderly person on their home. Two specific problems were investigated: configuration and functionality of central workstation as a module for data acquisition and analysis and second problem is devoted to user’s home environment monitoring. I. INTRODUCTION EALTH personalization and support for older and immobilized people is actually very important target of many national and international initiatives (e.g. Framework Program 7, Hong Kong “Care for the Elderly 2007 - Active Mind”). Different research areas are connected with those initiatives including “Wearable Sensors (WS)” [1], “Body Area Sensors (BAS)” [2], “Wireless Sensor Net- works (WSN)” [3] and telemedicine methods [4]. As a result of this research different sensors and integrated solutions were proposed, usually dedicated for a particular goal. De- signing a home platform for support of older or immobilized people different categories of existing and possible compo- nents should be considered from particular sensors to central computer stations. H In case of systems with integrated sensors many solutions were proposed like Crossbow IRIS [5], Sun SPOT [6], eWatch [7], Smart-Its [8] or other [9]. Many motes are cur- rently under constructions, however they are usually equipped with embedded sensors (e.g. temperature, light, and location), expending slots (e.g. sandwich model) and communication modules (Bluetooth or based on IEEE 802.15). Dedicated, medical extensions are often proposed like results of CodeBlue project [10], MobiHealth project [11] or UbiMon (Ubiquitous Monitoring Environment for Wearable and Implantable Sensors) project [12]. Typical so-  This work was partly supported by European Regional Development Fund concerning the project: UDA-POIG.01.03.01-22-139/09-00 -“Home assistance for elders and disabled – DOMESTIC”, Innovative Economy 2007-2013, National Cohesion Strategy. lutions used for such extensions (or standalone systems) are universal medical diagnostic devices, including ECG, pul- soxymeter, blood pressure/pulse monitors, etc. Communication interfaces allow data acquisition (especially at home) from motes to a one central station (or a middle- ware). The central station may be used to process data to as- sess user state based on many parameters and inform a user relatives or healthcare professional (a nurse, general practi- tioner) about the patient condition. The central station is of- ten required to limit data processing at sensor node and to build an integrated view on the patient (including ontology based context models [13]). Recent advances in sensor technology, cellular networks and information technology allow to improve the well-being of the elderly by assisting them in their everyday activities, monitoring their health status and environment conditions. There is more and more projects concerning the “smart” or “intelligent” homes with prototyping information-sensor systems recognizing habitant activities and abnormalities of them [14][15][16][17]. It is possible to develop systems that recognize an individual’s everyday activities and automati- cally detect changes in the behavioral patterns of people at home that indicate declining health [18]. Human-computer interface (HCI) for older citizens or immo- bilized people/patients is also very important aspect of the home-based system. Specially designed user-interfaces and interaction devices are often required. The main goal of this paper is to present a design of the mul- timodal, integrated platform for communication, training and health and environment monitoring at home. Communication includes technical and functional methods of a user commu- nication with his/her environment as well as processing of alerts from a home/user sensor network. The training is mainly related to promote mental activity of a group of pa- tients in danger (e.g. patients with dementia). Home monitor- ing is devoted to collect patient-related data and home envi- ronment data (e.g. fire detection). The very important aspect of the presented platform, is designing of a central computer station to collect data, process events/alerts, supply a proper human-computer interface, etc. In section II Multimodal platform for continuous monitoring of elderly and disabled Mariusz Kaczmarek Gdansk University of Technology Narutowicza 11/12, 80-233 Gdansk, Poland email: mariusz @biomed.eti.pg.gda.pl Jacek Ruminski Gdansk University of Technology Narutowicza 11/12, 80-233 Gdansk, Poland email: jwr@biomed.eti.pg.gda.pl Adam Bujnowski Gdansk University of Technology Narutowicza 11/12, 80-233 Gdansk, Poland email: bujnows@biomed.eti.pg.gda.pl. Proceedings of the Federated Conference on Computer Science and Information Systems pp. 393–400 ISBN 978-83-60810-22-4 978-83-60810-22-4/$25.00 c  2011 IEEE 393