HealthMax

Inspiration

We wanted to create a project with a real life application in healthcare which could help doctors in their prognosis and ill individuals who are unsure about what disease they may have.

What it does

Predicts the likeliest disease in an individual with a particular combination of medical symptoms.

How we built it

We first obtained a dataset which listed the range of symptoms in individuals with different illnesses. We then built a random forest machine learning classifier using this, which was chosen due to the binary nature of the data. The data was processed and split to train the model. Then, with the model trained, we added user inputs to select the specific symptoms present for the model to predict the likeliest disease. We then added a website to allow for a convenient user interface.

Challenges we ran into

The ESP32 could not port to any of our devices and we had many difficulties in wiring hence we resulted in using an Arduino UNO. We also ran out of time towards the end when making our website, preventing us from adding more stylistic elements.

Accomplishments that we're proud of

The model had a high accuracy in spite of the size of the dataset, and we were able to add a fully functional website with a simple UI.

What we learned

1. We learnt a great deal about random forest classifiers
2. We don't have much experience in web design, so this project taught us a lot about HTML, CSS and Javascript ## What's next for HealthMax The predictive model could be expanded by also suggesting suitable medicines for the individual to take

Built With

• flask
• python