This repository contains a Streamlit application that classifies lung images into three categories: COVID, Pneumonia, and Normal, using a pre-trained InceptionV3 model.
The application uses a Convolutional Neural Network (CNN) based on the InceptionV3 architecture to classify lung X-ray images. It allows users to upload images and receive predictions along with confidence scores.
- Upload images in JPG, JPEG, or PNG formats. You can also upload multiple image using zip
- Get the PDF Report for each image along with the classification result and confidence level.
- Displays the uploaded image along with the classification result and confidence level.
https://covid-pneumonia-detector.streamlit.app/
- Clone this repository:
git clone https://github.com/Roberttwil/covid-pneumonia-detector.git
- Install the required packages:
pip install -r requirements.txt
- Run on Streamlit:
streamlit run app.py
The model is trained using the dataset of lung X-ray images. The training code is available in model_code.ipynb. This notebook contains the following steps:
- Data preprocessing and augmentation.
- Model architecture setup.
- Training the model with early stopping and checkpointing.
- To train the model, run the cells in the model_code.ipynb notebook, and the best model will be saved as inceptionv3_best_model.keras.
https://www.kaggle.com/datasets/sachinkumar413/covid-pneumonia-normal-chest-xray-images
This project is licensed under the MIT License - see the LICENSE file for details.