EyeCare

Our Inspiration

We were inspired by a recent visit that one of our team members had to an optometrist, where they were referred to a eyecare specialist to be tested for glaucoma. While we're lucky enough to be able to access this sort of healthcare relatively easily, this may not be the case in many communities, and we wanted to see if there might be a better way of providing any indication of the disease.

What we used to build it

Our project was built using tkinter in python, and the model was created using tensorflow and keras.

Challenges

Our laptops aren't the most powerful, and struggled at times in the dataset that we used, which is larger than anything we've worked with before. This also made training and tuning difficult at times.

What we're proud of

We were able to learn a lot about machine learning in this weekend, and our model had an accuracy rate of over 70% after a lot of tuning and training, which we're proud of for an early interaction with neural networks.

What's next

We need to look more into our results, and understand what our model does well, and perhaps more importantly, what it doesn't, especially with regards to the individual diseases. It might not be as accurate on an even larger dataset and when other potential ocular conditions are involved.

Built With

• tensorflow