Inactivity patterns and alarm generation in senior citizens' houses

A methodology for deriving alarms about emerging health risks in senior citizens' homes is described. Instead of sensors worn on the body, a state of the art home automation installation is chosen as technical base. The incoming signals are interpreted as human activity leading to activity profiles. For a straightforward alarm generation a new type of profile based on inactivity patterns is defined. Their advantages are used to establish an alarm generation concept which allows a trade-off between the quickness of detection and the rate of false alarms. The MTFA (Mean Time between False Alarms) is introduced to adjust the trade-off according to the user's preference. The methodology is based on real data from a German Assisted Living project comprising 20 flats inhabited by seniors.