X-COVNets

This repository aims to provide a library of open-source models for computer-aided diagnosis of pneumonia-like lung diseases. Using deep transfer learning through the Xception architecture, Deep Convolutional Neural Network Models have been trained on the COVID-19 Chest X-Ray Database to classify COVID-19, pneumonia, and healthy patients.

X-CovNet Specifications

• Image Scale: 0-1

• Image Size: 224x224x3

• Class Mode: Categorical

• Interpolation: Bicubic

• External Validation: ✅

Collaborating Universities and Research Centers

• University of Computer Sciences, Havana, Cuba (UCI)
• Center for Medical Informatics (CESIM)
• Center for Computational Mathematics Studies (CEMC).

Authors

• Jorge Félix Martínez Pazos
• Arturo Orellana García
• Jorge Gulín Gonzáles
• David Batard Lorenzo