Inspiration
Doing research, I learned that approximately 500 million people worldwide are infected with Malaria each year. I wanted a quicker and easier way to figure out if a given person had malaria.
What it does
An app built to help fight one of the world's most terrible diseases: malaria, which every year infects over 200 million people and kills more than 200,000 children. My goal is to help healthcare workers in rural or developing areas diagnose malaria without expensive lab equipment. Our app lets a healthcare worker upload an image of blood cells under a microscope, and then uses computer vision to detect the malaria parasite.
How we built it
This web application was built using various python libraries in a google collab notebook; it was then hosted onto a website for others to access.
Challenges we ran into
I currently only have a free account for the website hosting so it is not always up and running; I need to figure out a way to have the website up at all times. For this reason, I have created a demo in case you all are unable to access the website. Also, the efficiency of the program is another challenge; approaching an efficiency of 100% is very tough.
Accomplishments that we're proud of
This web app can detect malaria at above 70% accuracy.
What we learned
I learned a lot first of all about the Malaria infection and how it is diagnosed/treated along my journey. I also learned how to use different python models that help use computer vision and learn patterns as they get fed more and more data.
What's next for Project MalariaScan
We hope to work towards a 99% accuracy and partner with some organizations to implement this to a larger audience.
Log in or sign up for Devpost to join the conversation.