Breathwise

Inspiration

Our inspiration was to help reduce traffic in hospitals and tackling one of the most common reasons for medical visits, respiratory illness. This way, doctors can focus more on patients who need it. It also makes healthcare more accessible by lowering the amount of money people need to spend on medical bills and allowing them to take charge of their own health.

What it does

Our product uses both machine learning trained on audios of breathing and an algorithm to determine the probability of diseases based on patient-reported symptoms. We combine the two to further boost the accuracy of our predictions.

How we built it

We used a React.js frontend with a Node.js backend, and we custom made our own machine learning model with a validation accuracy of 92%.

Challenges we ran into

We struggled a lot with getting our model to be accurate, and also including the model weights into the frontend as we had issues making sure our models were formatted correctly.

Accomplishments that we're proud of

We're proud that we were able to combine machine learning and statistics to make a very accurate product. We are also proud that we were able to persevere through all of the debugging.

What we learned

We learned a lot about the process of integrating machine learning into a frontend, and we also learned about ways to improve the accuracy of machine learning models.