Project Introduction
Heart disease is a major problem in America. According to the CDC, about 1 in 4 deaths in America are caused by heart disease. To solve this problem, researchers conduct studies to figure out ways to help prevent these deaths, either by finding ways to recover from heart disease or avoiding it altogether. Unfortunately, data collection tools like heart monitors and fitness trackers are expensive to use. This can result in these tools becoming prohibitively expensive for large studies, meaning that researchers will only be able to use them for short periods of time. Unfortunately, only using these data collection tools for a short time can result in participants giving biased or inaccurate data. This is a problem, as it means researchers will not get fully accurate information from their studies and thus will not be able to reach solid conclusions.
Our Client's Business
This is where our clients, Dr. Kyle Winfree and Dr. Eck Doerry, come in. They are working with the Wearable Informatics Lab at Northern Arizona University to fix this problem by allowing researchers to use Fitbits instead of the expensive tools that researchers normally use. This could help researchers to get more accurate data without having to worry as much about prices, resulting in better studies and potentially saving lives. This idea has a lot of promise, but their current implementation has several significant issues.
What's Broken?
Currently, our client's solution for Fitbit data gathering has a number of different issues. Listed below is our client's current workflow, which will allow us to illustrate the system's problems.
A diagram of our client's current workflow.
- Create studies
- Enroll participants
- Collect participant data
- Monitor studies
- Export data
However, this system also has a number of significant issues, including the following:
- Export times are too long, which means timeout errors can occur and prevent proper data export.
- There is no way to export data without using the website.
- Enrolling participants must be done manually, and participants cannot self-enroll.
- Participant communication is not possible with this system.
- The system does not currently collect Fitbit sleep data.
As a result of these, our client has chosen to scrap their current system and has asked us to build a new one for them that will avoid all of the previously stated issues.