Inspiration
COVID-19 [Corona Virus] is a severe global pandemic that has crippled world's economy and killed many people around the world.One of the primary ways to decrease the casualties is the infected person’s identification at the proper time.AI can play a significant role in these cases by monitoring and detecting infected persons in early-stage so that it saves time and lives
In this project ,we aim to propose a fully-automated method to detect/screen COVID-19 from the patient’s chest X-ray scanned images with out needing a clinical technician.
What it does
This is a demonstration of a cloud based AI Web Application which helps medical professionals or diagnostic centers to quickly screen/diagnoise the patient in just a single click.
How I built it
I've teamed up with AI enthusiasts and got inspired from the several news articles to create a end to end web application using Deep learning.We have collected datasets from Mendeley and Qatar University which were posted on kaggle website.We augumented the images and trained the model to detect the hidden patterns in the images and using "Transfer Learning" approach we have used the model to deploy it on cloud platform and hooked it via flask server.
Challenges I ran into
The main challenge was the dataset of X-ray images from medical institutions. Unitl now the Covid19 dataset is not shared to public by most of the countries which could have paved the way for more developers or scientists to find out a better way to deal with this pandemic
Accomplishments that I'm proud of
I'm proud that I've built an application for the masses leveraging AI and cost effective and is available to everyone through internet. The overall user experience and performance is exceptional with best results
What I learned
I've learned to create, design and deploy an AI application on cloud which is a end to end project
What's next for AI-Heroku-Cloud-Deployment
There is always a room for improvements and obviously the application can be improved by adding more datasets of Covid 19 and thereby the application could be medically approved by the doctors and other medical professionals.Also the application can be leveraged to predict other deadly diseases like Dengue,Malaria,Cancer..etc
Built With
- ai
- cloud
- css
- flask
- html
- javascript
- keras
- python
- salesforce
- tensorflow

Log in or sign up for Devpost to join the conversation.