Scalable Integration of Multiple Health Sensor Data for Observing Medical Patterns

With an aging global population, Ambient Assisted Living (aal) attempts to improve life expectancy and quality of life through the remote monitoring of various health signals using personal and home-based sensors. Possible medical conditions can be early ascertained by observable patterns over the patients’ health data. However, aggregating multiple raw signals and matching against medical protocols can be computational and bandwidth intensive. Moreover, adding new protocols requires non-trivial expertise to define necessary rules. This paper describes a lightweight, scalable, and composable mechanism that captures, processes and infers possible health problems from raw data obtained from multiple sensors.