Senior citizens often face many health concerns when it comes to daily life. Geriatric hospital care solves many of these issues, but most elderly patients would rather avoid such a time commitment. However, leaving elderly people, especially those with dementia, places them at risk of becoming lost, forgetting to eat, and getting dehydrated. In fact, dehydration was sole reason for 99 deaths in UK hospitals in 2015. There are many ways to track hydration, but most solutions leave gaps in service. Urine is proven to be a reliable method in detecting hydration of patients, where the concentration of dissolved particles and color can determine hydration.

What it does

The Stay Hydrated application alerts elderly users with an onscreen message when urine is detected to be too dark. The application rates various shades of urine, providing messages telling users exactly how much water should be drunk and when the hydration must occur. A community reward system has also been implemented where users gain points, incentivizing the application’s use among elderly people.

How we built it

We researched several methods in determining dehydration from non-invasive tests. Papers generally focused on blood pressure, skin elasticity, and urine. Urine proved to be the most readily accessed information from elderly people. We proposed a device that would funnel urine in a toilet to be photographed by a downward-facing camera in order to obtain the color of urine, with clear urine implying hydration and darker shades implying dehydration. The photo was to be sent through Bluetooth over to a mobile device. In Java, the image then was broken into individual pixels and their RGB values, which were averaged to determine the darkness of the urine. The darkness index was then classified into one of 6 indices, which were then displayed on the application, which was built using MIT’s App Inventor.

Challenges we ran into

We had issues connecting with the server as our personal devices were on different IP addresses. We also had problems with combining our block code in MIT App Innovator with our Java/jsp image processor. Other issues popped up here and there during development, but we were able to work through most of them.

Accomplishments that we're proud of

We are proud of implanting an image processor/color detector to determine the concentration of urine in testing for dehydration. Coming from non-technological or high school backgrounds, we were able to specialize in our skills and majors to make the most use of our team. Ultimately, we are just excited to be able to release our own program to be showcased to other brilliant innovators.

What we learned

We learned that there are any ways to improve geriatric care, but many fall short due to existing handicaps placed on elderly people. Fortunately, urine is a commonly available form of measurement that provides many important statistics about elderly health. We also learned through sufficient research that there are numerous creative methods in building a sensor, and color and its detection provides a surprisingly powerful means of measuring dehydration.

What's next for StayHydrated

We hope to continually implement new features on StayHydrated, such as functional improvements to facilitate cleaning the toilet device and an improved incentive system for continued use. Funding for a fully functional prototype of the StayHydrated toilet sensor and device would also provide invaluable data and opportunities to this project.

Share this project: