COVID-19 Lung Infection Segmentation from CT Images

Medical image segmentation pipeline to segment background and three lesions:

• Ground-glass opacity
• Consilidation
• Pleural Effusion

[]

Reproducibility

• Ubuntu > 18.04
• Python > 3.6
• NVIDIA card with CUDA support (RTX 1070Ti or better)

Step-by-Step workflow:

Download the code repository via git clone to your disk. Afterwards, install all required dependencies, download the dataset and setup the file structure.

git clone https://github.com/tyommik/covid19_ct_lesion_seg
cd covid19_ct_lesion_seg

conda env create --file environment.yml
jupyter lab

Run train.ipynb using your own data

Dataset

Using private dataset. There is no option to provide it :(

Results

<demo> : http://covid19.shibaev.info