With technology and the internet of things (IoT), smart health care is no longer a dream. The devices that wouldn’t usually be generally expected to have an internet connection can communicate with the network independent of human action, and this is referred to as the internet of things. These devices, in our case, are the sensors placed in the elderly assisted living facility Tiger Place. Sensors include the ballistocardiogram (bed sensor) and motion sensors placed in the living room, bathroom, kitchen, etc. These sensors help in measuring nine major features, which defines the per-day activities of each elderly resident in the facility, like time in bed, time spent in the bathroom, stride time etc. The sensors are connected to a network where the data is collected and stored in an enterprise database, linked to each person. The data placed in the database is collected on an everyday basis and is analyzed automatically to determine if the features are in the normal range of mean and standard deviation. If anything goes above and beyond the normal range, alerts are generated and sent to the facility health care takers. The data for these alerts is available from the past two years. The alerts include date and the information of the abnormal feature(s) associated. Our current research is to discover trends in these alerts over the time for an individual patient, as well as for the overall population of elderly living in the facility. I have used data mining association rules algorithm to find associated features which occur together and repeatedly, over time in elderly. In the end, the aim is to associate these repeating trends with some chronic or acute health conditions; For example, we may discover a resident has suddenly decreased walking speed and increased bed restlessness, which may be indicative of a recent injury due to fall.
Please contact Robert Sanders (SandersRL@missouri.edu) for Zoom information.