With our world in lockdown, people everywhere are working on ways to safely open up our economies. But until there’s an effective vaccine, we won’t be able to return to normal. In the absence of a vaccine, improved identification and treatment of COVID-19 is our best path forward. And, right here in Edmonton. We have a team that is getting closer to finding a better way to identify the virus.
Bio-Stream Diagnostics is a brand-new company led by John Murphy. John is a long-standing member of Alberta’s tech community and downplays his background and experience despite his work with a variety of startups. His first company, built alongside his brothers, was Shana Corporation. Shana provided electronic forms management, which was groundbreaking in the late ‘80s. Shana was sold in 2003 to Filenet, which eventually was sold to IBM.
Since 2003, John has been focusing on technology to improve optical spectrometry. He’s been kept busy recently running Bio-Stream and Stream.ML. Stream.ML specializes in deep learning and allows users to use machine learning to improve image detection, for instance, identifying specific bacteria in soil. It was this work by Stream.ML that sparked the idea for Bio-Stream.
Bio-Stream is in the very first stages of research focused on Raman spectroscopy and identifying COVID-19. To get a little nerdy, Raman spectroscopy is a technique typically used to determine the vibrational modes of molecules and most commonly used in chemistry to provide a structural fingerprint by which molecules can be identified.
Raman spectroscopy has been around since the 20’s and is used heavily for food safety, pharmaceutical purposes and for military identification of explosives or chemicals. In this instance, the goal for John’s team is to use Raman spectroscopy and machine learning (neural nets), to identify the signal of the COVID-19 virus.
When a person is tested for the virus now, it is done using PCR (polymerase chain reaction) testing which in Alberta takes approximately 36 hours. A leading edge PCR test at the White House has it down to 20 minutes. The cost for these tests is significant at $50 per test or more and between time and cost, it isn’t scalable enough to get our economy going.
If Bio-stream can use Raman spectrometry and neural nets to detect COVID (SARS-CoV-2) at a molecular level, there is the potential to have test results in 30 seconds at lower costs and high accuracy. That 30-second test could be used to identify more people at a cost of potentially one dollar. It would open the doors for wider testing at hospitals, nursing homes, or even office towers and public events. Additional benefit? It could increase the detection of asymptomatic carriers as we open up.
John’s team is taking a very collaborative approach on moving this project forward. Working with BioMark Diagnostics, the University of British Columbia (UBC), the Vancouver General Hospital, and experts around the world, John’s goal is to get this project moving quickly.
The first step is already underway with researchers at UBC already gathering the data they need to identify the virus—and make it so that the neural nets can help with the quick identification based on that data. From there, they need 500 samples that consistently identify the virus before they can get Health Canada to approve it as a Class 3 device.
John is hopeful for a fall launch of the initial solution but with it being so early in the process, he knows there will be problems to solve and hurdles to jump. Which is why they are hoping to get as many people as possible to follow their progress. Whether it be scientists who are experts, investors, or people looking to support with their skills, the more people that can help this move forward the better.
You can sign up (like I did) at bio-stream.ca/how-to-help to see how you can get involved. You never know—you might just know the right person who can help Bio-Stream Diagnostics through the next step.
Check out our full interview to learn more: