Controlling data quality in a personalized health monitoring system

Rapid advances in information and sensor technology have led to the development of tools and methods for individual health monitoring. One critical issue in these systems is the reliability and quality of the gathered data. In this paper, we develop a data monitoring system to detect issues with the collected data's reliability. We consider data from an all-in-one station-based health monitoring device that collects elderly's daily vital signs in an elderly home in Hong Kong. Due to the nature of the data both the sample sizes as well as the number of measured variables varies over time. We develop a new multivariate control chart to monitor the data quality. This new method has the ability to monitor data with varying sample sizes and varying number of parameters effectively. We illustrate the new method using the data on elderly's vital signs.