Simple and Effective Home Health Monitoring Using Spatiotemporal Data
Mentor 1
Roger Smith
Mentor 2
Nathan Spaeth
Start Date
1-5-2020 12:00 AM
Description
As more older adults are choosing to live independently in their homes, there is a need to monitor the level and nature of their daily activities. The aim of this project is to create a simple spatiotemporal prototype system to track an individual’s activity in the home. The future prototype will use a smartphone or smartwatch as its platform, be unobtrusive, low cost, and non-invasive. Methods for this project included creating and testing a semi-structured interview. The data such as that collected from the semi-structured interviews will be converted into real-time daily schedules that will include time, location, and duration of activities. In the future, we plan to work with computer scientists to create a computer learning algorithm to identify activities using spatiotemporal variables such as sequence, frequency, duration, and location in order to predict an activity an individual is participating in. We have conducted three pilot interviews and are in the process of analyzing the data. We are anticipating that the time invested in analyzing the semi-structured interviews may not justify the quality of data gathered. We may change our direction to include adding an activity configuration chart component that will be filled out by the participant with prompting from the researcher and compare the results against the data gathered from the interviews.
Simple and Effective Home Health Monitoring Using Spatiotemporal Data
As more older adults are choosing to live independently in their homes, there is a need to monitor the level and nature of their daily activities. The aim of this project is to create a simple spatiotemporal prototype system to track an individual’s activity in the home. The future prototype will use a smartphone or smartwatch as its platform, be unobtrusive, low cost, and non-invasive. Methods for this project included creating and testing a semi-structured interview. The data such as that collected from the semi-structured interviews will be converted into real-time daily schedules that will include time, location, and duration of activities. In the future, we plan to work with computer scientists to create a computer learning algorithm to identify activities using spatiotemporal variables such as sequence, frequency, duration, and location in order to predict an activity an individual is participating in. We have conducted three pilot interviews and are in the process of analyzing the data. We are anticipating that the time invested in analyzing the semi-structured interviews may not justify the quality of data gathered. We may change our direction to include adding an activity configuration chart component that will be filled out by the participant with prompting from the researcher and compare the results against the data gathered from the interviews.
