Spaces:
Sleeping
Sleeping
Updated README to be connected to the arXiv paper
Browse files
README.md
CHANGED
@@ -1,17 +1,17 @@
|
|
1 |
# Automatic lung tumor segmentation in CT
|
2 |
-
This
|
|
|
|
|
3 |
|
4 |
![sample of masked output](https://github.com/VemundFredriksen/LungTumorMask/releases/download/0.0.1/sample_images.png "Sample output of two different tumors")
|
5 |
![sample of 3d render](https://github.com/VemundFredriksen/LungTumorMask/releases/download/0.0.1/sample_renders.png "3D render of two masked outputs")
|
6 |
|
7 |
## Dependencies
|
8 |
-
In addition to the python packages specified in requirements.txt, [PyTorch](https://pytorch.org/get-started/locally/) and [
|
9 |
|
10 |
## Installation
|
11 |
```
|
12 |
-
|
13 |
pip install git+https://github.com/VemundFredriksen/LungTumorMask
|
14 |
-
|
15 |
```
|
16 |
|
17 |
## Usage
|
@@ -22,7 +22,16 @@ lungtumormask input_file output_file
|
|
22 |
|
23 |
# Example
|
24 |
lungtumormask patient_01.nii.gz mask_01.nii.gz
|
|
|
25 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
```
|
27 |
-
## Limitations
|
28 |
-
This repository is a proof of concept. It is not intended to be used in clinical or commercial use. However, it might be interesting for research, play a role as a baseline, or even aid semi-supervised lung tumor segmentation.
|
|
|
1 |
# Automatic lung tumor segmentation in CT
|
2 |
+
This is the official repository for the paper [_"Teacher-Student Architecture for Mixed Supervised Lung Tumor Segmentation"_](https://arxiv.org/abs/2112.11541), submitted to the International Journal of Computer Assisted Radiology and Surgery ([IJCARS](https://www.springer.com/journal/11548)).
|
3 |
+
|
4 |
+
A pretrained model is made available and can be used as you please. However, the current model is not intended for clinical use. The model is a result of a proof-of-concept study, and an improved model will be made available in the near future, when more training data is made available.
|
5 |
|
6 |
![sample of masked output](https://github.com/VemundFredriksen/LungTumorMask/releases/download/0.0.1/sample_images.png "Sample output of two different tumors")
|
7 |
![sample of 3d render](https://github.com/VemundFredriksen/LungTumorMask/releases/download/0.0.1/sample_renders.png "3D render of two masked outputs")
|
8 |
|
9 |
## Dependencies
|
10 |
+
In addition to the python packages specified in requirements.txt, [PyTorch](https://pytorch.org/get-started/locally/) and [lungmask](https://github.com/JoHof/lungmask) must be installed.
|
11 |
|
12 |
## Installation
|
13 |
```
|
|
|
14 |
pip install git+https://github.com/VemundFredriksen/LungTumorMask
|
|
|
15 |
```
|
16 |
|
17 |
## Usage
|
|
|
22 |
|
23 |
# Example
|
24 |
lungtumormask patient_01.nii.gz mask_01.nii.gz
|
25 |
+
```
|
26 |
|
27 |
+
## Acknowledgements
|
28 |
+
If you found this repository useful in your study, please, cite the following paper:
|
29 |
+
```
|
30 |
+
@misc{fredriksen2021teacherstudent,
|
31 |
+
title={Teacher-Student Architecture for Mixed Supervised Lung Tumor Segmentation},
|
32 |
+
author={Vemund Fredriksen and Svein Ole M. Svele and André Pedersen and Thomas Langø and Gabriel Kiss and Frank Lindseth},
|
33 |
+
year={2021},
|
34 |
+
eprint={2112.11541},
|
35 |
+
archivePrefix={arXiv},
|
36 |
+
primaryClass={eess.IV}}
|
37 |
```
|
|
|
|