X-ray is one of the most commonly used diagnostic technology as it is widely available, low cost, non-invasive, and easy to acquire. Chest radiography is the most popular and important imaging modality used to diagnose various pulmonary diseases. The automatic segmentation of the lung region from chest X-ray (CXR) can help doctors diagnose many lung diseases.
This code was developed with Python 3.10.12 on Linux Mint 21.3. Python requirements can installed by:
pip install -r requirements.txtThe Chest Xray dataset is used for this task. All images are resized to 512*512 and samples of the data are in the assets folder.
The segmentation_models_pytorch library is used to create the segmentation model which is defined in model.py file.
The train.py is used to train the model. Pretrained model run for 50 epoch is saved in assets folder. The inference.py is used to test the model on the test dataset.
