Much has been written in recent years about the significant challenges facing Canada’s stressed and cash-strapped health care system. Could artificial intelligence (AI) provide some much-needed solutions?
Entrepreneurs across the country are developing new platforms and devices harnessing AI, aimed at providing better, more efficient and more accessible health care experiences for patients. Here are three Canadian companies developing AI solutions they hope will revolutionize the way medical care is delivered:
Better predicting patient outcomes
When patients are admitted to hospital, they will either improve with treatment or decline. But predicting which way a patient will go can be a challenge, leaving care teams unsure of how much support a patient needs to thrive. Tech company Signal 1, in partnership with Unity Health Toronto, is leveraging AI to make that guesswork more informed.
St. Michael’s Hospital (part of Unity Health Toronto) developed their own AI solution, CHARTwatch, in November 2022. It uses a machine learning algorithm to predict which patients are at a higher risk of deteriorating during their hospital stay. That allows staff to make preparations for moving patients into the ICU sooner, improving outcomes and reducing stress. Signal 1 partnered with Unity Health to market and distribute CHARTwatch to other health care organizations.
“It’s a fantastic AI problem,” says Signal 1 co-founder and COO Mara Lederman. “We look at all the data that’s coming in on this patient: their lab results, their vitals, their age and how many days they’ve been here, and we can constantly generate a prediction of how at-risk this patient is for getting worse.”
Signal 1 and Unity Health have also co-developed a complementary solution to identify which patients are improving and ready to be discharged soon.
“Now we’re trying to predict who is going to be clinically stable in the next two days, so that all the activities required to get them out of the hospital can begin now,” says Dr. Lederman, a business professor turned tech entrepreneur. “The transition planning, like calling the family, getting home support, finding a rehab bed, whatever it may be, it all takes time. We’re shortening those excess days at the end of a patient’s stay.”
Dr. Lederman’s Signal 1 co-founder is AI pioneer Tomi Poutanen, founder of AI research lab Layer 6 (which has since been sold to TD Bank Group). The two joined forces to figure out how to use the AI tech stack Mr. Poutanen developed to improve the health care industry.
“Our health care system is in crisis,” Dr. Lederman says. “We simply don’t have the human and physical resources to meet the health care demands of our population. The only way to solve this is through innovation, and AI is able to make our existing resources more productive, delivering health care in a lower-cost environment.”
More efficient physician note-taking
When Vancouver-based software engineer Vaibhav Malhotra dislocated his shoulder during a snowboarding accident in the winter of 2023, he went through a series of medical appointments with doctors and specialists to treat his injury.
“I saw how fragmented the health care documentation system is,” Mr. Malhotra says of the experience. “There were errors in documentation or missing information that the new practitioner wasn’t aware of.”
He also noticed how the burden of note-taking negatively affected both the patient experience and doctors’ workloads. “The physician isn’t making eye contact with the patient [or] giving their full concentration,” he says. “[And] physicians are completing documentation at the end of the day, when they should be winding down and spending time with their families.”
Mr. Malhotra teamed up with his former software engineering classmate Bhimesh Chauhan to explore how they could automate note-taking for health care providers. The result is Scrubs Co-Pilot, an ambient AI transcription tool.
“We’re utilizing a language learning model optimized for the medical industry,” Mr. Chauhan says. “It is able to identify and extract relevant actionable notes that doctors can then review and input into the patient’s electronic health record.” As the tool is used by a particular health professional, it learns more about that practitioner’s work and the terms specific to their specialization.
Testing on the Scrubs Co-Pilot transcription tool began in the spring of 2024 with a small number of doctors and dentists in the U.S. Mr. Chauhan and Mr. Malhotra estimate that using Scrubs Co-Pilot saves doctors two or three hours of note-taking time a day.
“This can also be expanded into generating reference letters [or] a plan of action for patients to refer to,” Mr. Malhotra says. “The possibilities are endless.”
Ultrasounds for more patients
In 2016, Clarius Mobile Health developed a wireless, handheld ultrasound device with the aim of making imaging accessible to more patients by “miniaturizing” ultrasound tools. When Ohad Arazi joined the Vancouver-based medtech company as president and CEO four years later, he saw an opportunity to improve the product with AI.
“To put it in the hands of many more medical professionals, we needed to make it much easier to use,” says Mr. Arazi, who is based in Vancouver.
Ultrasound typically requires specialized training to capture the right images and perform measurements to diagnose a patient’s condition, but Clarius’s AI tools can help clinicians identify and analyze anatomical structures, says Mr. Arazi.
“Very similar to how you use portrait mode on your camera is how Clarius automatically detects the anatomy and guides you to get a high-quality image,” he says. This ease of use allows more health care professionals to conduct ultrasound imaging.
“It’s making [ultrasounds] available in many new care settings, such as nurses in the developing world or Indigenous communities, EMTs, family doctors providing rural medicine and surgeons performing safer procedures in their offices.” Mr. Arazi says over a hundred locations in rural B.C. already use the Clarius device, which means those patients don’t need to fly out to a larger town to get medical imaging done.
Traditional ultrasounds are still required for complex procedures, like echocardiograms, he adds. But he sees a future where AI-enhanced software becomes advanced enough for patients to use the devices at home to monitor conditions like congestive heart failure.
“If you could arm patients in their own home and get an accurate measurement that is shared electronically with their care team, they can decide whether they need to come into the ER or whether they can be cared for in their home or community.”
