600 patients are expected to be recruited in the next six months. Study also includes development of an artificial intelligence model
WATERLOO, Ontario– Two novel technologies developed in Canada have begun testing this week at the Toronto-based Princess Margaret Cancer Centre at University Health Network (UHN)). KA Imaging, a University of Waterloo spin-off company, has partnered with a team of researchers from the University of Toronto and UHN in a study for the early detection of COVID-19.
“We want to develop and evaluate two novel imaging techniques, specifically dual-energy X-ray and thoracic tomograms, for early and reliable detection of COVID-19 manifestation in the lung,” explained Dr. Patrik Rogalla, principal investigator of the study.
The images from both procedures will be compared with the ones obtained on a CT scan, which is currently the standard of care for confirming pneumonia. The data will allow researchers to build an artificial intelligence model for automated and nearly instant image interpretation.
“Early triaging is an essential countermeasure to prevent the spread of COVID-19 and we are confident that our technology can make a significant contribution,” said Dr. Karim S. Karim, CTO of KA Imaging.
KA Imaging’s RevealTM, a patented portable dual-energy X-ray solution, will be one of the two novel modalities to be evaluated.
“Dual-energy X-ray isn’t new, but what we did with Reveal cannot be accomplished by anyone else in the world,” explained Karim.
With one shot, Reveal produces three images: one that is a regular digital radiography, another that highlights just the bones, and a third one that removes the bones to show only the soft tissue.
“All three images are sharp and free of motion artifacts, which are like streaks or defects that are frequently seen in other dual-energy technologies. Also, unlike other dual energy solutions or CT machines, Reveal is portable so it can be taken to the patient, minimizing the risk of spread of COVID-19 infection,” commented Dr. Karim.
The other technology to be tested, Thoracic Tomogram (TT), has been developed by Dr. Rogalla’s research group. It is a novel image post-processing technique that simplifies interpretation. The technique will be applied to exploit the complementariness between CT scans and X-ray images.
“What we propose is an innovative, multimodal, deeplearning architecture for COVID-19, with a promise to gain higher accuracy of diagnosing COVID-19 than methods based on one single image type,” said Dr. Rogalla.
The goal is to recruit 600 people with suspected pulmonary infection to participate in the study over the next six months.
About KA Imaging
A spin-off from the University of Waterloo, KA Imaging is a Canadian company that specializes in developing innovative X-ray imaging technologies and systems, providing solutions to the medical, veterinary and NDT markets. The company has successfully developed a line of innovative X-ray imaging products in the areas of phase contrast micro-computed tomography, ultra-high spatial resolution X-ray detectors and large area digital, dual-energy X-ray material separation detectors.