https://github.com/zabir-nabil/fibro-cosanet
Idiopathic pulmonary fibrosis (IPF) is a restrictive interstitial lung disease that causes lung function decline by lung tissue scarring. Although lung function decline is assessed by the forced vital capacity (FVC), determining the accurate progression of IPF remains a challenge. To address this challenge, we proposed Fibro-CoSANet, a novel end-to-end multi-modal learning-based approach, to predict the FVC decline. Fibro-CoSANet utilized CT images and demographic information in convolutional neural network frameworks with a stacked attention layer. Extensive experiments on the OSIC Pulmonary Fibrosis Progression Dataset demonstrated the superiority of our proposed Fibro-CoSANet by achieving the new state-of-the-art modified Laplace Log-Likelihood score of -6.68. This network may benefit research areas concerned with designing networks to improve the prognostic accuracy of IPF.
https://github.com/zabir-nabil/fibro-cosanet
ct-scans deep-learning efficient-networks efficientnet fvc lung-disease lungs lungs-ct-scan mri pulmonary pytorch
Last synced: 6 months ago
JSON representation
Idiopathic pulmonary fibrosis (IPF) is a restrictive interstitial lung disease that causes lung function decline by lung tissue scarring. Although lung function decline is assessed by the forced vital capacity (FVC), determining the accurate progression of IPF remains a challenge. To address this challenge, we proposed Fibro-CoSANet, a novel end-to-end multi-modal learning-based approach, to predict the FVC decline. Fibro-CoSANet utilized CT images and demographic information in convolutional neural network frameworks with a stacked attention layer. Extensive experiments on the OSIC Pulmonary Fibrosis Progression Dataset demonstrated the superiority of our proposed Fibro-CoSANet by achieving the new state-of-the-art modified Laplace Log-Likelihood score of -6.68. This network may benefit research areas concerned with designing networks to improve the prognostic accuracy of IPF.
- Host: GitHub
- URL: https://github.com/zabir-nabil/fibro-cosanet
- Owner: zabir-nabil
- License: mit
- Created: 2020-10-07T18:03:41.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2021-04-14T09:12:31.000Z (about 4 years ago)
- Last Synced: 2023-10-20T23:58:39.127Z (over 1 year ago)
- Topics: ct-scans, deep-learning, efficient-networks, efficientnet, fvc, lung-disease, lungs, lungs-ct-scan, mri, pulmonary, pytorch
- Language: Python
- Homepage:
- Size: 384 MB
- Stars: 13
- Watchers: 2
- Forks: 3
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Fibro-CoSANet: Pulmonary Fibrosis Prognosis Prediction using a Convolutional Self Attention Network
### Installation
1. `git clone https://github.com/zabir-nabil/osic-pulmonary-fibrosis-progression.git`
2. `cd osic-pulmonary-fibrosis-progression`
3. Install Anaconda [Anaconda](https://www.anaconda.com/products/individual)
4. `conda create -n pulmo python==3.7.5`
5. `conda activate pulmo`
6. `conda install -c intel mkl_fft` (opt.)
7. `conda install -c intel mkl_random` (opt.)
8. `conda install -c anaconda mkl-service` (opt.)
9. `pip install -r requirements.txt`
### Download Dataset
1. Download the kaggle.json from Kaggle account. [Kaggle authentication](https://www.kaggle.com/docs/api)
2. Keep the kaggle.json file inside data_download folder.
3. `sudo mkdir /root/.kaggle`
4. `sudo cp kaggle.json /root/.kaggle/`
5. `sudo apt install unzip` if not installed already
* `cd data_download; python dataset_download.py; mv osic-pulmonary-fibrosis-progression.zip ../../; unzip ../../osic-pulmonary-fibrosis-progression.zip -d ../../; cd ../; python train_slopes.py`
### Training
1. Set the training hyperparameters in `config.py`
2. Slope Prediction
* To train **slopes model** run `python train_slopes.py`
* trained model weights and results will be saved inside `hyp.results_dir`
3. Quantile Regression
* To train **qreg model** run `python train_qreg.py`
* trained model weights and results will be saved inside `hyp.results_dir`
* Volume calculation: https://www.kaggle.com/furcifer/q-regression-with-ct-tabular-features-pytorch
### Arxiv pre-print
https://arxiv.org/abs/2104.05889