Skip to content

mlmed/covid-severity

Repository files navigation

Predicting COVID-19 Pneumonia Severity on Chest X-ray with Deep Learning

🛑 NOT FOR MEDICAL USE 🛑

There are now two datasets:

  1. Scored severity for the COVID-19 Image Data Collection Dataset is here: Pneumonia severity scores for 94 images
  2. Stonybrook Radiographic Assessment of Lung Opacity (RALO) dataset is here: Pneumonia severity scores for 2373 images

The scores are explained as follows for the RALO dataset:

  • geographic_extent_mean: The extent of lung involvement by ground glass opacity or consolidation for each lung (right lung and left lung separately) was scored as: 0 = no involvement; 1 = <25% involvement; 2 = 25-50% involvement; 3 = 50-75% involvement; 4 = >75% involvement. The total extent score ranged from 0 to 8 (right lung and left lung together).
  • opacity_mean: The degree of opacity for each lung (right lung and left lung separately) was scored as: 0 = no opacity; 1 = ground glass opacity; 2 = consolidation; 3 = mix of consolidation and ground glass opacity (>50% consolidation); 4 = white-out. The total opacity score ranged from 0 to 8 (right lung and left lung together). NOTE: The total opacity score ranged from 0 to 6 for the COVID-19 Image Data Collection Dataset so scalling (like opacity/6*8) will align the two datasets.

Data is here: Pneumonia severity scores for 94 images

License: CC BY-SA Creative Commons Attribution-ShareAlike

These are from the follow paper: Cohen, Joseph Paul, et al. Predicting COVID-19 Pneumonia Severity on Chest X-Ray with Deep Learning. Cureus Medical Journal, 10.7759/cureus.9448, http://arxiv.org/abs/2005.11856.

@article{Cohen2020Severity,
title = {Predicting COVID-19 Pneumonia Severity on Chest X-ray with Deep Learning},
author = {Cohen, Joseph Paul and Dao, Lan and Morrison, Paul and Roth, Karsten and Bengio, Yoshua and Shen, Beiyi and Abbasi, Almas and Hoshmand-Kochi, Mahsa and Ghassemi, Marzyeh and Li, Haifang and Duong, Tim Q},
journal = {Cureus Medical Journal},
doi = {10.7759/cureus.9448}
url = {https://www.cureus.com/articles/35692-predicting-covid-19-pneumonia-severity-on-chest-x-ray-with-deep-learning},
year = {2020}
}

To run the CLI:

# basic command line predictions
$ python predict_severity.py 2966893D-5DDF-4B68-9E2B-4979D5956C8E.jpeg
geographic_extent (0-8): 5.978744940174467
opacity (0-6): 4.169582852893416

# or to output a saliency map:
$ python predict_severity.py 01E392EE-69F9-4E33-BFCE-E5C968654078.jpeg  -saliency_path heatmap.jpg
Image Saliency map Predictions
geographic_extent (0-8): 5.979
opacity (0-6): 4.17
geographic_extent (0-8): 6.293
opacity (0-6): 4.367
geographic_extent (0-8): 3.067
opacity (0-6): 2.335
geographic_extent (0-8): 0.9483
opacity (0-6): 1.0

About

Predicting COVID-19 Pneumonia Severity on Chest X-ray with Deep Learning

Resources

License

Stars

Watchers

Forks