Prolipsi

Prolipsi is a medical web application designed to help users during this pandemic. The program uses a neural network to predict if the uploaded image of the user's lung is COVID positive.

Comment

Inspiration

Our group has family members and friends in the medical field. During these difficult times their lives have become very stressful due to the amount of people seeking medical attention. We came up with this idea to help with diagnosing COVID, in hopes to lessen the burden and make their lives a bit easier.

What it does

Uses a neural network from Keras and machine learning from Tensorflow. The web application is able to analyze images from a database of COVID-19 x-ray images of both normal and COVID positive lungs to predict uploaded COVID positive lungs.

How I built it

A public dataset containing x-ray images of lungs of patients with normal lungs as well as COVID positive lungs was used to train this model that would look for COVID in uploaded x-ray images. Using Tensorflow and Keras, we were able to create and train this model and determine the accuracy of this model through tests.

Challenges I ran into

Being beginners, we have minimal to no experience in this field as this is idea is something that has never crossed our minds. This is our first attempt at creating a website and our first time playing around with machine learning.

Accomplishments that I'm proud of

This entire project was a step out of our comfort zone and we all feel proud of what we have accomplished in the past three days.

What I learned

Learned how to use django, html, css, connecting the front end of a web application to the back end as well using the Tensorflow and Keras to develop the COVID-19 detector.

What's next for Prolipsi

Further improvements to the web application regarding speed which the model for the COVID detector is made as well as the overall aesthetics of the web application.

Built With

Share this project:

Updates