When a medical emergency strikes, time is critical, and every decision could be a matter of life or death. health.io aims to assist in this hectic time by calculating the optimum treatment location for its users.
What it does
health.io calculates a personalized treatment success rate at all hospitals near the user, and recommends the hospital which is best suited to treating the user's specific medical ailment.
How we built it
health.io is deployed on an asynchronous, RESTful web service built from scratch with Flask and Google gPRC. The AI powering health.io is a custom neural network built with Google Tensorflow. It was trained on the John Snow Labs Hospital Comparison Database in the cloud via Amazon Web Services.
Challenges we ran into
Accomplishments that we're proud of
What's next for health.io
We hope to implement a feedback loop for our users so that they can help the health.io AI learn and improve. Additionally, we are exploring ways that we can deploy health.io live after Hophacks to make it free and available to everyone.