Our team's mission is to build tools that facilitates everyone who needs to track its blood glucose levels.
While it's common knowledge that diabetes is a huge problem in 21st century, we sought a solution that puts the problem at the forefront - starting with the habits that almost everyone participates in, daily tracking. We designed Diabetes Tracker to make it easy to figure out how much deviation in the sugar levels, does every patient has during the day. It's difficult to make daily habits, that is why we created motivational quotes which are revealed only when the users saves at least 9 records of its sugar levels. With Diabetes Tracker, we hope to motivate the people to use the app on a daily basis by encouraging them to record their values more frequently. All of that is achieved by using a simple but effective gamification.
What it does
Diabetes Tracker is a mobile app, which motivates its users every day to track their blood glucose levels, because the daily monitoring is essential for every person with diabetes to maintain a healthy and well-balanced lifestyle.
Apart from that the app is stores the value in its core data and uses machine learning to analyses the recorded values. If the values are not in the correct range, the app sends a notification to its user or has a feature to directly contact the the private doctor assistant, who has to take a professional check and prevent the patient from future health problems.
How we built it
The app is built on SwiftUI. The data is stored in Core Data and JSON format The notifications are triggered via Firebase For prediction and analysis we used MySQL and Python. The ML models were build on jupytor notebook.
Challenges we ran into
The main challenge was to synchronise the back-end with the from-end. During the past hours we have created the app using SwiftUI as well as the ML models which analyse the stored data and classify it as good or bad. But we didn't have enough time to integrate it in the app. As a separate project we have tested the ML model on unseen data and it performed as planned. We have also created the set-up for the remote notifications, using Firebase as well as the direct communication with the private doctor, but all of that was done within individual projects.
Accomplishments that we're proud of
We are very proud of the organisation which we managed to create. For the limited amount of time, our team has managed to accomplished all individual to-do tasks, despite the fact that most of them were very complicated and challenging for some of us. We are also very proud of the idea and the simplicity of the project, because nowadays there are many app on the market but most of them have many unnecessary features which are never used, therefore some apps are very complicated for people with low education or especially in areas with elderly people who have limited experience with smartphones. Our goal was to create a simple from design and features app, which covers all the basic and most needed for the people with diabetes, and make the user experience as intuitive as possible, so that everyone can use it, regardless of age or education.
What we learned
We learned today that sometimes the seemingly simplest things (dealing with app synchronisation for hours) can be the greatest barriers to building something that could be socially impactful. We also realised the value of well-developed, well-documented APIs for programmers who want to create great products.
What's next for Diabetes Tracker
The main goal of our team now is to finish the app, by connecting the data base with the ML model, so that prediction can be executed, as well as connecting the app with the private doctors, so that the patients can have a professional support or consultation if needed. We truly believe that the app can have a great impact, therefore we will do our best to complete is and bring it to the people with diabetes as soon as possible.