Inspiration
We have always been aware of the accessibility issue of certain technologies to all doctors across the world. Some people don't have the same access to medical care as we do just because of where they are from. Health care shouldn't be a luxury, it should be available to everyone. This is what inspired us to build an application that could help doctors and even individuals diagnose different types of eye conditions. Our app increases accessibility because it doesn't require the technology that doctors usually use to diagnose these conditions, which not everyone has access to.
What it does
Users upload an image of their eye the website. The image then passes through our prediction software which uses machine learning to develop prediction and display it on the screen. It can be used to help doctors and individuals, especially those that are from low-income countries.
How we built it
We used data from Shanggong Medical Technology Co., Ltd. to create the machine learning models behind our prediction software. This was built using Python. The web application includes file upload sections for users to upload a picture of their eye that will be passed through the ML prediction software.
Challenges we ran into
We came across a few challenges with our website. Our first challenge was simple. We could not figure out how to keep different colors for different parts of the website. Eventually we used some information we learned from others and the internet and were able to solve it. Another problem we faced was that we did not know how to make the pictures accessible from different computers. We solved that by finding some examples online. Also, training the ML models was taking a bit longer than what we expected, so we cut down the conditions that our web app could detect to three main ones.
Accomplishments that we're proud of
This was the first time building a real website for most of us, so we are proud that we were able to learn how to build it and actually make the website in the time that we had. Also, we were able to train the ML models using the data we had to make a prediction with a good accuracy. This was a great accomplishment.
What we learned
We learned more about website design like including a file upload and sizing images. We also learned how to connect the ML code to the web app so the prediction software could be run through the web application.
What's next for aEye Technologies
he first thing we have to do is improve on our current web application. We should work on making it look nicer and getting a domain for it. We could also try to increase the accuracy of of the ML software and make it as accurate as possible.
Log in or sign up for Devpost to join the conversation.