AI Software Detects COVID-19 Patients at Early Stage as Well as Quantifies Extent of Lung Lesions

A Belgian initiative for AI lung scan analysis which supports radiologists in the assessment of CT images of the lungs of COVID-19 patients has grown into a multicentre European project.

The Belgian initiative icovid was set up in March by UZ Brussel (Jette, Belgium), KU Leuven (Leuven, Belgium), icometrix (Leuven, Belgium) and ETRO, an imec research group of Vrije Universiteit Brussel (VUB; Brussels, Belgium). The aim was to investigate how to deploy lung scans in the COVID pandemic and what AI software would be needed to do so. As a result, the AI tool icolung was born.

icolung was certified in April and is currently being used by over 75 hospitals worldwide and has analyzed over 35,000 lung CT scans. The icovid project now builds on the development of icolung and is committed to scaling up. The partners will also collaborate to incorporate other clinical information into the models in addition to CT images. icolung is being offered free of charge to hospitals across Europe. While vaccination will eventually normalize daily life, the virus will still circulate and other coronaviruses may emerge. In local outbreaks, it will be important to identify COVID-19 patients early.

icovid will continue to respond to the rapidly evolving needs of the COVID-19 pandemic. For example, it is estimated that about 2% of COVID-19 patients have symptoms lasting longer than three months, called Long COVID or Chronic COVID Syndrome (CCS). With the improved software, the consortium will also better predict the long-term effects of COVID-19, which are today still largely unknown. The ability to identify patients who may develop CCS opens up possibilities for hospitals to better adapt care and share the burden.

“What started as a local project is now being rolled out in 800 hospitals in Europe and supported by excellent research centers all over Europe,” said Professor Jef Vandemeulebroucke of ETRO. “With icolung, we can detect COVID-19 patients at an early stage and quantify the extent of lung lesions. Meanwhile, we are further improving the AI software to identify lung damage as COVID-19 even more quickly, and to determine the further care path of the patient faster and better through prognostic models.”

Related Links:
UZ Brussel
KU Leuven
icometrix
Vrije Universiteit Brussel