-
-
Accuracy
-
Deep Learning Model
Inspiration
This project helps a patient to get checked for the presence of COVID-19 disease in very less time. Currently, it takes around 1-2 hours to get the test results and test kits are limited by number.
What it does
Automatic detection of COVID-19 from X-ray images of lungs using OpenCV, Keras and Tensorflow in Deep Learning.
Challenges we ran into
The main challenge was getting the dataset. Being a new disease, getting lung X-Ray images was very difficult, but we could find a publicly available dataset for our purpose. Also, another challenge was to use the right baseline model. We experimented with VGG16, VGG19, ResNet50, etc and found VGG16 most promising according to our needs and size of the dataset, thus getting maximum accuracy.
Accomplishments that we're proud of
It gives best in class accuracy of approximately 97%.
Log in or sign up for Devpost to join the conversation.