Skip to content

ConvNet that predicts whether a person has pneumonia or not based on a chest X-Ray.

Notifications You must be signed in to change notification settings

namanshah01/pneumonia-classifier

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Pneumonia Classifier

A ConvNet trained on a Kaggle Dataset that takes chest X-ray as an input and predicts whether the patient has pneumonia or not. Integrated it into a web app with Flask and deployed it on Heroku.

Details

Achieved 0.83 recall, 0.91 precision and 0.87 f1-score. The model used to overfit a lot, so used data augmentation, batchnorm and dropouts to reduce variance. Used this GitHub repo as a reference and modified it to suit the model and the website.
Link to the website: https://pnclassify.herokuapp.com/

Tech Stack

  • DL Model : Python, TensorFlow, Cv2
  • Backend : Flask
  • Frontent : HTML, CSS, Bootsrap, JavaScript
  • Deployment : Heroku

About

ConvNet that predicts whether a person has pneumonia or not based on a chest X-Ray.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages