Development of a wireless health monitoring system with an efficient power management scheme based on localized time-varying data analyses

This work involves development of an integrated health monitoring system based on wireless sensor motes for continuous wireless monitoring of vital signs for large populations. The system incorporates efficient power management schemes based on localized data analysis, and will be useful in a variety of settings such as nursing homes, emergency rooms, military fields, and individual health monitoring. Remote monitoring of patients via wireless sensor systems has the potential to change the way health care is delivered. However, until now very few cost-effective wireless remote monitoring systems have been developed and put into use. One of the key challenges in making wireless monitoring ubiquitous in health care is efficiently managing the limited energy in individual sensor nodes. To this end, the current work utilizes smart sampling and sufficient processing at the sensor so that minimal data is transmitted. Our technique is based on the development of appropriate hardware and software implementations. It has the added benefit of significantly reducing the amount of data that needs to be analyzed at the central server, which further facilitates faster real-time alerts. Finally, our localized data analysis scheme also results in minimal usage of available radio transmission bandwidth, which allows for robust wireless transmission of many other vital signs' data from multiple patients in close vicinity. In this context of wireless health monitoring, this work has the following four specific aims. The first specific aim involves design and development of integrated scalable wireless motes containing an electrocardiograph, pulse oximeter sensor and other various sensors. The goal is to determine if these additional physiological parameters lead to better on-demand sampling rate strategies, better fault tolerance, and more accurate medical decision alerts. The second specific aim is to design and develop an on-demand variable sampling data transmission scheme. The sampling rate strategy will be based on localized real-time data analysis using a microprocessor attached to a mote combined with the subject's a priori medical data and the published risk factors. The third specific aim is to design wireless networking protocols for transmitting data from a large number of monitored subjects in the same physical space. The fourth specific aim involves development of new algorithms to facilitate more accurate medical decision alerts.