Leuven, Belgium – April 23, 2020 – icometrix, VUB, UZ Brussel, KU Leuven, UZ Leuven, UZ Antwerpen and imec, together with hospitals and organizations worldwide, joined forces in a global initiative to leverage artificial intelligence (AI) in the fight against COVID-19. The multinational collaboration resulted in the development of an AI algorithm, icolung, which received CE-marking for clinical use in Europe. icolung is the first CE-marked AI solution for CT resulting from a COVID-collaboration and offers fast and objective quantification of lung pathology on chest CT scans in admitted COVID-19 patients.
Evaluating the type, pattern, and extent of lung pathology on chest CT can help in the assessment, triage, and follow-up of COVID-19 patients. Currently, the health care system worldwide is overwhelmed by the surge of admitted COVID-19 patients, leading to an excessive workload and a shortage of resources and intensive care units (ICU) beds. Triage can help alleviate the increasing burden on the ICU and allocate resources. icolung has the potential of further decreasing workload in clinical practice by providing a fully automated assessment of the total and lobular disease burden.
“The smooth 24/7 collaboration between icometrix, the academic groups of the VUB and KU Leuven, the university hospitals UZ Brussel, UZ Leuven, UZA and many other hospitals, has ensured that we developed clinically usable AI software that meets all privacy rules and security issues, in an incredibly short period of time.” says Prof. Paul Suetens, coordinator of the consortium.
"We have been testing the first versions of icolung, and believe that quantitative CT data is important," says Prof. Johan de Mey, head of radiology of the University Hospital of Brussels. "Thanks to the cloud, improvements of the AI algorithm are instantly available to us, which is essential in this rapidly evolving pandemic."
“We are proud to launch a solution that can be integrated in less than 30 minutes and has a big impact on patients. Collaboration is fundamental, and we are honored to dedicate our resources and AI expertise to this global AI initiative against COVID-19,” says Dirk Smeets, Ph.D., CTO at icometrix.
icolung is a cloud-based AI software, developed by researchers of ETRO, an imec research group of VUB, and ESAT-PSI of KU Leuven, to quantify disease burden in COVID-19 patients on non-contrast chest CT. icolung quantifies total and regional lesion burden and returns a concise report and annotated images directly into the hospital PACS, within 10 min. icolung is the result of an ongoing collaboration, further improvements and developments of the icolung algorithm are automatically deployed to all users via the cloud. The icolung software seamlessly integrates into the hospital PACS and is currently offered pro bono.