Compare ML models that predict COVID-19 from a Chest X-Ray

Home / Portfolio / Predict COVID-19 from a Chest X-Ray

Upload a DICOM file of an AP frontal chest x-ray (recommended). Alternatively, a JPEG or PNG screenshot of the chest x-ray can also be uploaded. Take a picture of a DICOM image or upload a DICOM file of an AP frontal chest x-ray (recommended). Alternatively, a JPEG or PNG screenshot of the chest x-ray can be uploaded.

Keras classification model

Model which was trained on VGG16 in Keras Framework
PyTorch Binary classification model

Model which was built using Conv2d Layers in PyTorch Framework
Pytorch ResNet50

Model which was built using ResNet50 Architecture in Pytorch Framework
COVID-Net Geographic severity model

Outputs the SARS-CoV-2 severity scores for geographic extent. This model predicts the geographic severity. Geographic severity is based on the geographic extent score for right and left lung. For each lung: 0 = no involvement; 1 = <25%; 2=25-50%; 3=50-75%; 4=>75% involvement, resulting in scores from 0 to 8.
PyTorch classification model

Model which was trained on ResNet18 in PyTorch Framework
PyTorch Multi-class classification model

Model which was built using Conv2d Layers in PyTorch Framework
Binary Classification model

Model which was built using the DenseNet121 model in Keras Framework.
COVID-Net Opacity severity model

Outputs the SARS-CoV-2 severity scores for opacity extent. This model predicts the opacity severity. Opacity severity is based on the opacity extent score for right and left lung. For each lung: 0 = no opacity; 1 = ground glass opacity; 2 =consolidation; 3 = white-out, resulting in scores from 0 to 6.
Version - 1.2

Disclaimer: This is not for medical purposes and is for informational purposes only. No images are saved on our servers. Models are run as serverless functions!

X

To add this Web App to your homescreen click and then "Add to Home Screen".