yashshende / pneumonia_detection

A Flask Pneumonia Detection web app from chest X-Ray Images using CNN

Flask Web-App Pneumonia Detection from Chest X-Ray Images CNN

A Flask pneumonia detection web application

Setting up the Web-App Locally

1. Clone The Repository or Download Zip from https://github.com/YashShende/Pneumonia_Detection/archive/master.zip & Extract it.

2. Install Requirements pip install -r requirements.txt

3. To run the app on localhost run python app.py

4. app is running at http://127.0.0.1:5002

Dataset

Dataset Name : Chest X-Ray Images (Pneumonia)

Dataset Link : https://www.kaggle.com/paultimothymooney/chest-xray-pneumonia
           

Model

In this Project i Trained my model on Google Colab and saved model in Models Folder.
You can put your trained model inside the Models folder to get better results just 
keep in mind the dimenssions of image feed to trained model if you want to take a 
look how i trained my model check this link 

Link : https://colab.research.google.com/drive/1RrHsRmCa-V9EtZQRI0HnnVoaN9j8A5qk

WebApp

Home Page

Prediction

pneumonia_detection's Issues

What kind of data preparation models were used before traing models ?

Hello Yash, How are you. Are you good ? I have just created this issue, cause I need to know which data preparation technique was used for the model in this project. As you know before training models, data should be processed. I mean, its pixels, sizes, colors should be changed to improve result accuracy.
For instance : There are some techniques : SuperVision (AlexNet) Data Preparation,Train-Time Augmentation,GoogLeNet (Inception) Data Preparation

How can I train model or already trained ?

Hello, your work is good, but documentation is not enough, can you tell us where images should be uploaded ?