Inspiration

In 2018, nearly 8,000 people died from melanoma in the state of Florida[1], our home state. Melanoma is one of the most costly forms of cancer; however, with early detection, there is approximately a 99 percent survival rate[2]. With these in mind, we decided to come up with a near-free way to determine if a system user should consult a dermatologist regarding a formal melanoma diagnosis.

What it does

Our system is able to predict, with good accuracy, whether or not the end user potentially has melanomia through feeding an image of their lesion through a machine learning model.

How we built it

Machine learning, hardware, and plenty of web development!

Challenges we ran into

On the machine learning side, curating the melanoma dataset was quite a challenge. On the kiosk side, interfacing the hardware with the frontend dragonboard was an initial challenge.

Accomplishments that we're proud of

The ability to predict melanoma is definitely something we're proud of!

What we learned

We learned a lot about both user experience and cancer detection. Different areas, yet they intersected in our project design.

What's next for Melanoma Kiosk and More!

Mass production of kiosks and finding volunteer professionals.

Sources

  1. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-and-statistics/annual-cancer-facts-and-figures/2018/cancer-facts-and-figures-2018.pdf

  2. https://www.skincancer.org/skin-cancer-information/skin-cancer-facts

Share this project:
×

Updates