Skip to content

Latest commit

 

History

History
55 lines (41 loc) · 2.87 KB

README.md

File metadata and controls

55 lines (41 loc) · 2.87 KB

GitHub Repo stars GitHub forks GitHub contributors GitHub issues GitHub closed issues GitHub pull requests GitHub closed pull requests APM

logo

NOvid-20

One solution to all Covid-19 Queries
An online doctor and a health assistant, whenever and wherever you need
A solution to foretell the likelihood of being in vicinity of COVID-19 using ML and AI


Features

  • COVID-19 Predictor - The predictor model based on machine learning approach analyzes the likelihood of the disease based on various factors.

  • COVID-19 Detector - With the help of COVID-19 Detector, you can upload your Chest X-Ray image and get a prior prediction of whether you are COVID-19 positive, negative or it's a case of Pneumonia.

  • Chatbot - An AI assistant will resolve all your queries related to the disease and provide implementable solutions to them.

  • Current situation Overview - Latest updates and overview of the on-going situation to help you keep track of it.

Demo

Tech Stack

  • For COVID-19 Predictor, Ensemble Machine Learning classifier was used to classify the symptoms and factors entered by the user. The accuracy achieved by the model is 98.2% using Random Forest.
  • For COVID-19 Detector, Convolutional Neural Networks were used to detect the COVID-19 using Chest X-Ray images with an accuracy of 95%.
  • For both deployment of Predictor and Detector, Flask was used for backend deployment on heroku.
  • Main website is a static website hosted using Github Pages with the frontend built using Bootstrap template.
  • Chatbot is powered by SnatchBot.

Contributing

NOvid-20 is fully Open-Source and open for contributions! We request you to respect our contribution guidelines as defined in our CODE OF CONDUCT and CONTRIBUTING GUIDELINES.

Contributors