Trajectory-based Ambient Assistive Care for Elders

UN Department of Economic and Social Affairs predicts a worldwide increase in life expectancy coupled with a decline in the number of children and total fertility; thus a drastic reduction in support provided by young individuals working in home and health care elders. This calls for an automated home-based system, capable of unobtrusive monitoring of the status of elders living alone independently. The system should have some level of intelligence or decision support properties and provide sufficient analysis to enable the elder to safely live an independent life style. This paper presents a smart home sub-system for monitoring activity levels for elders. The proposed system uses specialized ceiling-mounted intelligent activity monitoring sensors and a central hub to infer the level of activity and behaviour.

The sub-system has the ability to characterise behaviour as patterns of transition between learned resting locations, to measure overall activity levels and to detect specific events such as potential falls. We build a probabilistic spatial map of resting locations using the head position of the subject, represented as a mixture of Gaussians in 2D space. The map is used in conjunction with a Hidden Markov Model (HMM) framework, to construct two models: one for normal activity and the other for arbitrary behaviour. Similarly this is used to model the level of activity and to distinguish between normal and abnormal behaviour. By definition, usual or abnormal events are rare, and so unusual behaviour is best detected by deviation from a model of normal behaviour. This basic assumption is also used to detect falls in our system.

The sensor includes a movement detector, video camera, processing unit and wireless connection. To protect privacy and minimise data transmission the sensor processes the data locally and only transmits vital positional and activity level data for further analysis at the central hub. To support ceiling mounting a wide angle “fish-eye lens” camera is used, making the sensor unobtrusive and reducing the installation cost. The sensor is energy efficient and can be battery powered, obviating any need for wiring. Video processing is only triggered when there is sufficient movement in the environment, significantly reducing power consumption. Both sensor and hub use low-cost components, making the installation of an affordable multi-sensor system in different rooms feasible. The sub-system is part of the wider development of TOTALCARE, an end-to-end assistive care solution including multiple sensors, secure central data storage analysis, and integrated support for multiple communities of users, including clinicians, carers and social networking groups.

Rate article
Family Medicine for America's Health
Add a comment