Dermitri

Inspiration

Cancer is an important and devastating issue in modern society. The group responsible for Dermitri also have some personal ties to cancer in their lives.

What it does

Our app employs state-of-the-art. cutting-edge machine learning algorithms to analyze skin lesions, providing instant and relatively accurate assessments. Users can simply upload a photo, and within moments, receive an evaluation on the type of skin cancer they have and whether or not they have.

How we built it

We gathered images and data through serpapi web scraper and kaggle datasets. We then used roboflow to manage our dataset, and train the model. Finally, we made the frontend with html, css, and javascript and accessed the model using the roboflow api.

Challenges we ran into

Accuracy was low because of the similarities between the skin cancers. We also got distracted occasionally.

Accomplishments that we're proud of

We are proud of making a full webapp that is functional in the allotted time

What we learned

We learned more about web development, and machine learning

What's next for Dermitri

We plan on improving our accuracy as well as potentially expanding the scope of our identification as well as provide further treatment/evaluations of scanned images