This challenge aims to develop a multiclass classification algorithm capable of detecting COVID-19 in Chest X-ray images. The dataset contains 3 image classes: COVID-19, Pneumonia and Normal (healthy) (See example images below). With 20,000+ images, the participants can train their algorithms to solve this challenge. A test set will be released and will be used to benchmark the obtained results.
challenge link: https://cxr-covid19.grand-challenge.org/CXR-COVID19/
dataset available at: https://drive.google.com/drive/folders/1T2O6Z8FkZiZv0Krwb44PsjZ6tdS4-PDm?usp=sharing
VGG16 with 16 batch size: best model link (174 MB): https://drive.google.com/file/d/1MsLsNDj5CfKNwI0V1eNGU0JS-nliY_Ve/view?usp=sharing