Melanoma is the deadliest and most common type of cancer in the US. Many times, the telling signs go unnoticed for years on end due to the nature of it's symptoms; a small blemish here, a spot there. These seemingly benign signs send the affected's life down a spiral of expensive and grueling treatment. The problem with the current UV Index is that it is a blanket measure that doesn't take into account the individual. So one may be at a higher risk of getting sunburned than the UV Index leads them to believe, possibly increasing their risk of Melanoma three times over. This is where our app comes in.

What it does

Burnt is a UV index scaling Android application that takes in a location and a picture and renders a tailored UV index for that individual based on their skin tone and the current high UV Index in that location for that day.

How we built it

We did all of our work in Android Studio with the help of a few APIs and other software. This included Esri's Android SDK, Google's Cloud Vision facial recognition API, and Dark Sky's weather API. Using these, we created our own algorithms to tailor the UV index value based on the skin tone of the user.

Challenges we ran into

We first used Android Studio's built in facial recognition, but the results were fast but unreliable. We then attempted to use Microsoft's Face API for the recognition, but the current version we had of gradle in Android Studio was not compatible with that API. Finally, we settled on Google Cloud Vision facial recognition and were able to produce fairly consistent results, the only problem being that we had to sacrifice speed for reliability. In addition, while testing weather APIs, we went through a few before deciding that Dark Sky provided the most relevant information to our project.

Accomplishments that we're proud of

Overall, the event went very well for our team. We are happy to have made our proof of concept app and are proud of the amount of work we put in. Throughout the process, we got practice with Android application development, API usage, and working as a development team. In addition, our ability to adapt to unforeseen challenges without giving in showed resolve.

What we learned

We learned a great deal about the implementation of APIs and using that data in a meaningful way. In addition we got a LOT of work with Android Studio and the process of bringing a mobile application to fruition.

What's next for Burnt

The next steps for our company are to expand our research and pinpoint an accurate measure for our predictive algorithm. At the moment, our research was very limited and the studies out there for this specific problem are sparse. We would like to run studies to find reliable data that could allow us to implement a machine learning implementation of this project.

Built With

Share this project: