This project is part of Machine learning hackathon organized by sparkhub.
Every year in India around 200-300 people die with pneumonia according to previous year statistics the major age group affected with pneumonia are above 70 and below 5 years old. Early detection of pneumonia is crucial for determining the appropriate treatment of the disease and preventing it from threatening the patient’s life. So we inspired to build a web app which detects pneumonia using lung X-ray images with greater accuracy.
When the Person drops lung X-ray our deep learning model which will tell us whether the person is having pneumonia disease or not having pneumonia.
We used deep learning models VGG16 and Transfer learning to train and test the dataset each with two hidden layers. Chest X-rays dataset is taken from Kaggle which contain various x-rays images differentiated by two categories “Pneumonia” and “Normal”. Html ,CSS ,JS and python framework flask are use for frontend.
While working with this project we came across many models. Incorporating models correctly in our project is quite challenging in a short period of time. At the same time we also faced hardship while deploying using flask and gunicorn.
We're really proud after completing this project with 96% accuracy
We learnt models like VGG16 and TL which are complete new to us. Even though we are little familiar with flask deploying as web-app is challenging which we accomplished.
Further we can create a single web app for various lung diseases such that when the person drops his/her lung X-ray our web-app detects if he is having any lung disease such as Covid-19 - Asthma-Pneumonia or normal.