MedMe

Inspiration

We wanted to create a simple and easy to use tool for medical diagnoses for everyone.

What it does

Our product is capable of diagnosing medical conditions, based off of an image and/or description the user provides. We can determine the type of disease, and whether it is malignant or benign. The user can optionally take a picture of their condition and add a description, which is matched to hundreds of symptoms in our database.

How we built it

We built this using Google App Engine as the primary web host, along with AngularJS as the front-end framework. The back-end was created using NodeJS and a medical machine learning model with TensorFlow

Challenges we ran into

Choosing the right machine learning model was the hardest part of our project. We started out with the Google Vision API but switched to our own neural networks to get more relevant feedback.

Accomplishments that we're proud of

We're proud of our machine learning model, especially since none of us had experience with TensorFlow prior to this hackathon.

What we learned

We learned a lot about machine learning, TensorFlow, and dividing work efficiently as a team.

What's next for MedMe

We plan to extend MedMe's functionality to include more diseases and compile more data for more accurate results, and

Built With

• app-engine