medical-image-segmentation

A mini project of group 13 studying Intro to Deep Learning course at Hanoi University of Science & Technology

Instructions to run below

Features

The project will allow you to classify lung radiography images into 3 types

• Normal (Lung is healthy)
• COVID (The person has covid)
• Non-COVID (The person doesn't have COVID but suffers different lung-related issues)

If a patient is diagnosed with COVID, the pipeline will segment the COVID-diagnosed radiography image further for better prescriptions and treatments

Instructions to use

• Firstly, clone the repository's main branch into your desired directory using your git command prompt.

git clone -b main https://github.com/king17pvp/medical-image-segmentation.git

• Secondly, you can access the directory by this command.

cd medical-image-segmentation

• Thirdly, install required libraries via requirement.txt pip install -r requirements.txt

• Next, go to https://drive.google.com/drive/folders/1hcFRbwaR2sFW6akC5lkkSu3H3rJpvvcj?usp=sharing and download ALL the models and put them in the following directories, otherwise, the inference program will not work. The place where models are located should follow the directory tree below:

medical-image-segmentation/weights
├── segmentation_models
│   ├── AttUNet.pt
│   ├── R2UNet.pt
│   ├── R2AttUNet.pt
│   ├── full_lungs_resnet.pt
│   ├── ResNetUNet.pt
│   └── .gitignore
├── classification_models
│   ├── VGG16.pth
│   ├── VGG19.pth
│   ├── ResNet18.pt
│   ├── ResNet50.pt
│   └── .gitignore
└── .gitignore

• Finally, run the project by

python app.py