covid19net

COVID19 CT Scan Visual Recognition Project. The research paper on COVID19 diagnosis by chest CT scans is now published. You can find it here: https://doi.org/10.1016/j.ibmed.2021.100027

https://github.com/josehernandezsc/covid19net

Science Score: 57.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
    Found CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
    Found 2 DOI reference(s) in README
  • Academic publication links
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (9.5%) to scientific vocabulary
Last synced: 6 months ago · JSON representation ·

Repository

COVID19 CT Scan Visual Recognition Project. The research paper on COVID19 diagnosis by chest CT scans is now published. You can find it here: https://doi.org/10.1016/j.ibmed.2021.100027

Basic Info
  • Host: GitHub
  • Owner: josehernandezsc
  • Language: Python
  • Default Branch: master
  • Homepage:
  • Size: 24.4 KB
Statistics
  • Stars: 3
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 0
Created over 5 years ago · Last pushed over 4 years ago
Metadata Files
Readme Citation

README.md

COVID19Net

COVID19 CT Scan Visual Recognition Project.

The research paper on COVID19 diagnosis by chest CT scans is now published. You can find it here: https://doi.org/10.1016/j.ibmed.2021.100027

This repository consists of three .py files:

Model_Training.py shows the code developed by this repository's author for his paper titled "An Ensemble Approach for Multi-Stage Transfer Learning Models for COVID-19 Detection from Chest CT Scans". Preprint soon available. Model training is shown from the data preprocessing to the final ensemble model evaluation. Proposed model is identified as MultCovNet.

Results_Evaluation.py shows the evaluation metrics for the baseline ensemble method and the final ensemble model. This script uses the validation and test set scores and true labels obtained as CSVs from Model_Training.py script.

Grad-CAM.py constructs the influence representation image for a series of randomly picked COVID19 CT Scan images under the proposed model.

This work relies on the COVID19 dataset collected by:

Zhao, Jinyu and Zhang, Yichen and He, Xuehai and Xie, Pengtao
COVID-CT-Dataset: a CT scan dataset about COVID-19
arXiv preprint arXiv:2003.13865
2020

Requirements

  1. PyTorch
  2. Torchvision
  3. tqdm
  4. Pillow
  5. Pandas
  6. Numpy
  7. Scikit-image
  8. Scikit-learn
  9. pytorch-gradcam

Environment

Development is based on Python 3.6.6 and was run on a NVIDIA K80 GPU with 12GB of internal memory.

The final model parameters dictionary will be made available soon.

For instructions on dataset handling and loading please refer to https://github.com/UCSD-AI4H/COVID-CT

Owner

  • Name: Jose Hernandez Santa Cruz
  • Login: josehernandezsc
  • Kind: user
  • Location: Edinburgh, Scotland
  • Company: The University of Edinburgh

Data scientist with a deep inclination for neural networks (no pun intended). Guitar player, obsessed learner and heavy reader.

Citation (CITATION.cff)

# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!

cff-version: 1.2.0
title: >-
  An ensemble approach for multi-stage transfer
  learning models for COVID-19 detection from chest
  CT scans
message: Please cite this software using these metadata.
type: software
authors:
  - given-names: Jose Francisco
    family-names: Hernandez Santa Cruz
    email: jose.hernandez.sc@gmail.com
    orcid: 'https://orcid.org/0000-0002-7033-4191'

GitHub Events

Total
Last Year