Inspiration
1 billion people have vision impairment due to eye diseases which can cause blindness if left untreated. Many poor people or people that live in remote areas do not have easy access to an optometrist so I wanted to create a no cost web app that would allow people to check if they have an eye disease. This led me to create OptiScan.
What it does
OptiSca is a web app that uses machine learning to detect if the user has either Bulging Eyes, Cataracts, Crossed Eyes, Glaucoma, Uveitis, or is normal based on an image.
How I built it
I used Kaggle for the dataset, Tensorflow and Keras for the machine learning model, Streamlit for the UI and hosting platform, and python as the language to integrate all of these parts together.
Challenges I ran into
The Tensorflow model was not accurate in the beginning and so I had to increase the number of epochs and the training data.
Accomplishments that I'm proud of
I was able to make a Tensorflow model to detect eye diseases at a fairly high accuracy. I was able to scrape images off of google in addition to a dataset from Kaggle.
What I learned
I learned how to make a machine learning model with TensorFlow and host it so anyone can use the model.
What's next for OptiScan
In the future, I will add more diseases to the model so that less common diseases can be detected. In addition, I can market this website in remote areas to test the accuracy of this website.
Log in or sign up for Devpost to join the conversation.