X-COV is a research project of the Nuclear Physics Group of the Universidad Complutense de Madrid (GFN-UCM). Our aim is to help doctors by providing radiologists and other medical and diagnostic imaging professionals with tools for the automatic analysis of chest X-rays with Artificial Intelligence (AI), for the evaluation and monitoring of COVID-19 patients.
Below you will find all the detailed information about the project and access to its references, open database and demo.
Our aim is to help doctors. To this end, we develop and make available to radiologists and other medical and diagnostic imaging professionals automatic image analysis tools with Artificial Intelligence (AI).
In this project we focus on the COVID-19 pandemic and the interesting uses of X-rays to combat it.
The tools of the X-COV Project carry out an evaluation and follow-up of potential COVID-19 patients, based on their chest X-rays and basic information such as age or number of days with symptoms. The analysis, assessment and indicators offered by them can help doctors to have a more complete view of each case.
In this way, we try to facilitate and expedite their important decision-making.
This tool is of special interest in patients with moderate symptoms, or in those cases in which certain medical tests are inconclusive or give false negatives.
The software allows to obtain an immediate analysis, thanks to the use of Artificial Intelligence and Deep Learning techniques. To create the model, we have used machine learning tools such as Keras, TensorFlow, and Tensorflow.js. In the training of the model, open data available in multiple public repositories of chest radiographs and the data and images provided by the collaborating hospitals of the project have been used.
We follow two lines of action.
1.- Online tool:
- Open and freely accessible.
- For Research Purposes: Not Certified for Clinical Use.
- It allows access to the tool to doctors and researchers around the world, promoting and supporting research on COVID-19.
- It serves as a testing platform for the new features that are implemented as the tool improves.
- It facilitates the dissemination of the complete project.
2.- Version for clinical use:
- Exclusive use for medical imaging professionals.
- It requires certification with CE (Conformité Européenne) and / or FDA (Food and Drug Administration) marking of medical devices.
- It will be adapted to be integrated in the PACS (Picture Archiving and Communication System) of hospitals and with some radiology machines from specific suppliers.
- Winner in the category “Health and Social Issues” and master winner of the #CodeTheCurve hackathon organized by UNESCO, IBM and SAP (04/2020).
- Certification of finalist project of the #VenceAlVirus hackathon, promoted by the Community of Madrid (04/2020).
- First Prize CompluEmprende 2020.
San Carlos Clinical Hospital of Madrid, University Hospital of Fuenlabrada, University Hospital October 12th, Rúber Juan Bravo Hospital, Complutense University of Madrid, CompluEmprende Program, Hackathon Madrid #VenceAlVirus, Department of Science, Universities and Innovation of the Community of Madrid , Madri+D, CSIC, DASEL, Hackathon UNESCO #CodeTheCurve, SAP, IBM, Google Cloud, Amazon AWS and Repsol-Sinopec.