This folder includes a database of chest X-ray patients of posteroanterior (PA), anteroposterior (AP) and supine views. We have included all the Covid-19 diagnosticated patients databases we have found up to date, other examples of pneumonia from diseases different to Covid-19, and healthy patients. Our objective is to create a global and open database that may help to the diagnosis of the Covid-19 based on chest X-ray images.

We have also included a folder after crop and reshape the images to 224x224 pixels in order to train a Neural Network capable to diagnose the positive Covid-19, other pnuemonias, and healthy patients.

Contents:
 -/Images_original -> Includes the images downloaded from the original databases.
 -/small_images -> Includes the reshaped images to 224x224 pixels.
 -Metadata_COVID19IA.csv -> List of image file names, labels, and general information for each image.

We have identified the images with different labels that are availables in the Metadata_COVID19IA.csv.
 ·Status: 0=healthy patient; 1=Covid-19; 2=Other pneumonias; 3=Undefined.
 ·devices, markers: 0 = no devices or markers present in the image; 1=devices (wires, electrodes, pacemakers, etc.) or markers (arrows, letters, etc.) present in the image.
 ·Cropped lung (only applicable to the copped images data set): 0 = no visible parts of the lung have been cropped; 1 = parts of the lungs may be eliminating during the cropping process.

Besides this fundamental labels, we have maintained additional information from other databases (mainly those from https://github.com/ieee8023/covid-chestxray-dataset):
 ·patientID: Numerical identifier for patients with images at different stages of the disease.
 ·offset: Number of days since the start of symptoms or hospitalization for each image. If a report indicates "after a few days", then 5 days is assumed. This is very important to have when there are multiple images for the same patient to track progression.
 ·Sex: M=male; F=female; blank=unknown.
 ·Age: Age in years.
 ·finding: Type of pneumonia
 ·survival: Y=Yes; N=No; blank=unknown.
 ·intubated: Y=Yes; N=No; blank=unknown.
 ·temperature: Temperature of the patient in Celsius.
 ·pO2 saturation: partial pressure of oxygen saturation in %.
 ·leukocyte count / neutrophil count / lymphocyte count: counts in 10^3/uL.
 ·view: PA=posteroanterior; AP=anteroposterior; AP supine=supine anteroposterior.
 ·Modality: x-Ray.
 ·date / location
 ·DOI: Digital object identifier (DOI) of the research article.
 ·url: URL of the paper or website where the image came from.
 ·repository: URL of the repository website where the image came from.
 ·license: License of the image such as CC BY-NC-SA. Blank if unknown.
·clinical notes / other notes.

REFERENCES:
 -https://github.com/ieee8023/covid-chestxray-dataset
 -https://www.kaggle.com/andrewmvd/convid19-x-rays/
 -https://covid19.espacio-seram.com/index.php
 -https://www.bsti.org.uk/training-and-education/covid-19-bsti-imaging-database/
 -https://stanfordmlgroup.github.io/competitions/chexpert/