In-home monitoring system for assessing gait using wall-mounted RF transceivers

? DESCRIPTION (provided by applicant): The goal of this project is to develop and evaluate a new system called GaitRF that can measure walking speed and other detailed gait parameters passively within the home. The ability to move is a critical function that underlies the quality of
life for elders as well as those suffering from neurologic or physical ailments. Changes in aspects of gait such as walking speed, stride length, stride imbalance, etc. have been shown to correlate with changes in physical and cognitive health. Gait is typically measured periodically within a clinical setting. In-home monitoring offers the advantage of continuous and ongoing measurements providing significantly more information about changes in a patient's health under real life conditions. Ongoing accurate estimation of gait within the home can help to guide caregivers about the need for transitioning older adults to higher levels of care. Accurate in-home gait monitoring also has the potential to encourage older adults to maintain independence later into their lives by enabling family members and friends to monitor health status remotely. The proposed system makes use of novel wireless technology being developed at EmbedRF. An array of tiny transceivers are positioned in either a hallway or doorframe and configured to wirelessly send radio frequency (RF) signals between each other. As a person walks past the transceivers, their body disrupts the RF signal strength allowing for the estimation of gait metric without the need for the person to wear any monitoring device. Preliminary results using only 2 or 4 transceivers have already demonstrated the ability to accurately estimate walking speed. Preliminary results also show the unique ability to discriminate between individuals for use in a multi-resident home or to exclude data when a caregiver or friend visits. Specific aims in this proposal will 1) further develop prototype systems for walking speed estimation using transceiver pairs, 2) research and develop use of an array of sensors placed at base-board height in order to extract more detailed gait metrics; a second transceiver pair will be placed at approximate head height to provide multi-person discrimination capability, and 3) perform preliminary in-home evaluations of the GaitRF system. Advanced machine learning and tracking algorithms will be utilized to process the RF signals to extract gait metrics. We will evaluate the
performance of GaitRF in both a laboratory setting and within single-resident and multi- resident homes of seniors. If successful, this new system has the advantage of being low-cost, unobtrusive, and easy to install, while still providing accurate and detailed gait metrics. The system can be used as a stand-alone gait monitor or incorporated into more general health monitoring systems. Application of the technology will support independent living and may also be used in clinical trials of drugs or procedures that affect mobility.