Inspiration
Two of our group members were born with jaundice so its something we thought should be more accessible to detect.
What it does
It uses machine learning to detect jaundice in hands and eyes
How we built it
We built it using python ML training and a html css interface,
Challenges we ran into
Training the model to a degree which we were comfortable with.
Accomplishments that we're proud of
Training the model in a way we didn't give it a set of rules but rather recursively allowed it to train itself allowing it to be better through variation of skin tones and other factors.
What we learned
To think of a true problem that we've had and try to fix it is a good way to make an impact.
What's next for nail'det
Get a real dataset from a medical center and test our project with that and use to better sort out data.
Log in or sign up for Devpost to join the conversation.