PyDerm

Inspiration

I was inspired to do this project through research of medical applications of machine learning.

What it does

The AI takes an image of a skin lesion as an input and makes a prediction as to what skin condition applies to the image and whether it is cancerous or not.

How I built it

I used Keras and TensorFlow, which are python libraries for machine learning. I used a dataset called HAM 10000 which contained images of skin lesions used to train the machine learning algorithm.

Challenges I ran into

One of the hardest challenges I ran into was formatting the dataset to be able to be properly formatted as an input for the AI.

Accomplishments that I'm proud of

One of my proudest accomplishments was being able to build a machine learning project.

What I learned

From this project I learned a lot about machine learning, its applications, and how the technology can better our quality of life.

What's next for PyDerm

For this project, I want to implement my own machine learning algorithm from scratch. A long-term goal of this project is to be able to detect a wider range of diseases such as lung cancer.

Built With

• keras
• python
• tensorflow