A European initiative for

Automated diagnosis and quantitative analysis of COVID-19 on imaging


Imaging COVID-19 AI initiative is a multicenter European project to enhance computed tomography (CT) in the diagnosis of COVID-19 by using artificial intelligence. The project group will create a deep learning model for automated detection and classification of COVID-19 on CT scans, and for assessing disease severity in patients by quantification of lung involvement.

Many different hospitals and institutions across Europe will collaborate to rapidly develop an artificial intelligence solution in this time-sensitive research project. The AI model will be made freely available to all participants for clinical validation.


There is an increasing interest in the role of imaging for diagnosis of COVID-19. The infection causes a wide variety of imaging findings on CT scans, most typically ground-glass opacities and consolidations in the periphery of the lungs.


The sensitivity of chest CT to diagnose COVID-19 has been reported as high and can predate a positive viral laboratory test. Therefore, in endemic areas where the healthcare system is under pressure, hospitals with a high volume of admissions are using CT for rapid triage of patients with possible COVID-19 infection.

Chest CT has an important role in the assessment of COVID-19 patients with severe and worsening respiratory symptoms. Based on imaging, it can be evaluated how severely the lungs are affected, and how the patient’s disease is evolving, which is helpful in making treatment decisions.

There is an increasing awareness that lung abnormalities caused by COVID-19 can be found unexpectedly in CT examinations performed for other clinical indications, e.g. abdominal CT scans for bowel problems, or in patients without respiratory complaints.


To be able to cope with the growing number of COVID-19 patients, it is necessary to find ways to help doctors in their fight against the virus. In these dramatic circumstances the value of AI comes into play, by reducing the burden on clinicians. While a manual read of a CT scan can take up to 15 minutes, AI can analyze the images in 10 seconds.

Therefore, automated image analysis with artificial intelligence techniques has the potential to optimize the role of CT in the assessment of COVID-19 by allowing accurate and fast diagnosis of infection in a large number of patients. AI has the potential to support clinical decision making and improve workflow efficiency.

The project

The project group will train an AI algorithm with anonymized imaging data* from multiple hospitals and institutions in affected countries across Europe. The aim of the project is to automate the diagnosis of COVID-19 on CT scans, and to quantify disease burden in the lungs of infected patients. The developed AI model will be made freely accessible to all participating hospitals and institutions.

The project is supported by the European Society of Medical Imaging Informatics (EuSoMII). The research collaboration will be coordinated by the Netherlands Cancer Institute, who are experienced in applying artificial intelligence to medical imaging.

*The project is GDPR compliant. The results which can be derived from the AI model, cannot be related to any data subject.

Within less than a week, the researchers were able to obtain an agreement with more than 20 partners from many European countries, both academic and non-academic hospitals, willing to share anonymized data and to support us in preparing these data for training the algorithm. The participating hospitals are located in the most affected areas of Italy and Spain, and also in Germany, Belgium, the Netherlands and the UK. Not only the speed by which this collaborative network was established, but also the amount of anonymized data that now can be collected from European patients, is really unprecedented.

The project is being supported by Quibim (Spain) and Robovision (Belgium) who will provide a platform and their extensive experience for data preparation,  algorithm training and deployment of the deep learning model. The participating radiologists will collaborate by annotating the large imaging dataset on a cloud labelling tool powered by ImFusion.

The project will result in a joint scientific publication. The model and implementation platform will be made freely available as a research solution to the participating hospitals. The model will be clinically validated in real world practice.

Multicenter European collaboration

Map showing the European hospitals collaborating in the Imaging COVID-19 AI initiative (as far as registered on March 20th, 2020).

The research project is now restricted to EU countries due to regulations.

Call us

Send us an email

Proudly working together with
In collaboration with