HeartDiseasePredictor

Inspiration

We wanted to find a way to identify if a particular individual could have heart disease, and to do so we wanted to use a set of data and create a predictive algorithm to determine.

What it does

By using a set of data of people having/not having a heart disease with various categories pertaining to the particular individual (age, sex, bmi, race, physical health, etc.), we can apply it to a prediction algorithm constructed from the data set to return a response to the user stating whether or not they could have a heart disease with 91% accuracy.

How we built it

By using pandas, a python module that allows for data analysis and prediction models, we assigned values to each category of interest, ran it through the prediction algorithms from the data set, and then returned a string saying "yes" or "no" to indicate whether or not the user has a heart disease. Then we built a website using django- implementing html, javascript, and css- to build the UI, input methods, and importing the prediction algorithm. Finally, a button at the bottom of the screen takes in the entered values and redicrects the user to a new page where the result is stated based off of the prediction algorithm.

Challenges we ran into

Having more familiarity with mobile app development, we thought of implementing python into java based android studios so that the UI and algorithms would be much simpler to access. However, we were stuck for hours being unable to find a python interpreter for android studios which would have allowed us to import the pandas module in order for our algorithm to work. We then tried to work around this setback by trying kivy and flutter, but instead we decided to go with a basic website design that could easily access the algorithm and work as necessary.

Accomplishments that we're proud of

We are proud of developing our own predictive algorithm and being able to implement the algorithm into a website that takes in values from the user and successfully output a response.