Kard.io

Screenshot of what the app looks like at the moment
Screenshot of inputting values into SYS/DIA

Inspiration

Our inspiration came from the current global health crisis, where stress levels in people have risen due to the added weight gotten from job loss, tight money situations and such. Stress levels as we know are one of the main factors of rising blood pressure.

What it does

The vision for the app, Kard.io, was to ask the user to input diastolic, systolic and pulse rate values in order to show the user how their overall health is in terms of an increase or decrease in their blood pressure. It compares the value received to the data stored in the backend in order to provide the results to the user. If the user's bp is lower than the threshold, that is a good sign. If the user's bp is going up but below the threshold, then a warning will be sent. If the bp is high and has reached or gone above, an alert will be sent to the user.

How we built it

For the backend, multiple models were created using Python (Machine learning portion) and Firebase. The front end was done with Flutter and Dart, in conjunction with the Android Studio for the Android emulator (virtual device).

Challenges we ran into

time constraint, cohesion, our first health topic that we have to gather more knowledge on. It was also quite difficult to brainstorm in a voice call, when it is easier to brainstorm in-person

Accomplishments that we're proud of

While we delegated different portions of development, we were glad to help one another learn about how to solve posed technical challenges. We are proud that we could come together as a team to make a product with such high potential aimed towards helping everyone avoid conditions like hypertension .

What we learned

Through participating in this hackathon, we learnt that Flutter is a really good tool that allows for one codebase to facilitate both Android and iOS versions of the app. We also gained valuable experience building and configuring API's and testing multiple machine learning models.