![](assets/hjelde2020lobe.png)
The implementation was made for this study by Bouget et al. (2021). The original implementation can be found here.
The source code was tested in Python 3.6 with CUDA 10.0.
git clone git+https://github.com/andreped/PLS-Net
cd "PLS-Net"
pip install -r requirements.txt
from PLS_pytorch import PLS
network = PLS()
(Alternatively) in PyTorch-Lightning:
from PLS_lightning import PLS
network = PLS()
Disclaimer: Note that the Lightning implementation contains some hardcoded setup and Dataloaders, and thus only serves as an example. However, PyTorch implementation should work out-of-the-box.
If the source code is used in any scientific publication, please, cite the following papers:
@article{bouget2021code,
author = {David Bouget and Andr{\'e} Pedersen and Sayied Abdol Mohieb Hosainey and Johanna Vanel and Ole Solheim and Ingerid Reinertsen},
title = {{Fast meningioma segmentation in T1-weighted magnetic resonance imaging volumes using a lightweight 3D deep learning architecture}},
volume = {8},
journal = {Journal of Medical Imaging},
number = {2},
publisher = {SPIE},
pages = {024002},
keywords = {three-dimensional segmentation, deep learning, meningioma, magnetic resonance imaging, clinical diagnosis, Magnetic resonance imaging, Image segmentation, Tumors, Brain, 3D image processing, 3D modeling, Image resolution, Data modeling, Neural networks, Surgery},
year = {2021},
doi = {10.1117/1.JMI.8.2.024002},
url = {https://doi.org/10.1117/1.JMI.8.2.024002}
}
@misc{lee2019plsnet,
author = {Lee, Hoileong and Matin, Tahreema and Gleeson, Fergus and Grau, Vicente},
title = {{Efficient 3D Fully Convolutional Networks for Pulmonary Lobe Segmentation in CT Images}},
publisher = {arXiv},
year = {2019},
copyright = {arXiv.org perpetual, non-exclusive license},
keywords = {Image and Video Processing (eess.IV), Computer Vision and Pattern Recognition (cs.CV), FOS: Electrical engineering, electronic engineering, information engineering, FOS: Electrical engineering, electronic engineering, information engineering, FOS: Computer and information sciences, FOS: Computer and information sciences},
doi = {10.48550/ARXIV.1909.07474},
url = {https://arxiv.org/abs/1909.07474}
}