Inspiration
The severe healthcare disparities that poor kids must deal with—many of them without access to basic health screenings and education—were the driving force behind the creation of HealthHomie. Our goal was to provide these kids with an easily navigable resource for managing the risks associated with diabetes and heart disease.
What it does
You enter the parameters it gives (blood pressure, glucose, etc,) and the model will determine whether or not you have diabetes or heart disease.
How we built it
Using pre-existing data sets on diabetes and heart disease, we train the model using AI algorithms to increase its accuracy and make more precise predictions. HTML, CSS, and JavaScript are used in the design of our website. We show the buttons for choosing a health concern, which takes the user to a form specific to that disease. The user enters the data they have gathered through labs or over-the-counter devices. Form data is gathered by the Flask API.
Challenges we ran into
The data values were not accurate and their many missing values in the CSV. We had to filter out the data because many of them were bringing the accuracy of the model down.
Accomplishments that we're proud of
We made a website that can help underprivileged kids easily check to see if they have any life-threatening issues. We can help save lives with this model.
What we learned
We learned how to build a model using code and we learned about each disease and the severity of both diseases.
What's Next for Health Homie - Disease Prediction Tool
In the future, we want to add more diseases to HealthHomie's coverage and take user feedback into account to make the tool even more useful and accessible for impoverished communities.
Log in or sign up for Devpost to join the conversation.