SafeSkin - Early Skin Cancer Detection

Inspiration

One of our team member's aunts has skin cancer.

What it does

Decides whether an image of a lesion is benign or malevolent.

How we built it

We used Xcode with swift to host an iOS app and we used firebase and Tensorflow to port our machine learning neural network which we built using teachable machine.

Challenges we ran into

The main challenge that we encountered was with exporting our machine learning model to our app. Downloading firebase and tensorflowlite were especially difficult as there are many different versions of cocoa pods that we had to work through.

Accomplishments that we're proud of

We were able to create a machine learning model which has a 97% accuracy at low density and 99% accuracy at high density at detecting the difference between benign and malignant tumors which will be able to help many people. We are also proud of the fact that we got the app to function on our own ios phones.