Eye contact and earlier diagnosis: How AI is transforming front-line health care in B.C.

There's no shortage of headlines about artificial intelligence being used in all sorts of areas.

British Columbia’s medical community is buzzing with enthusiasm and ideas, personal anecdotes and concerns, as the adoption of artificial intelligence becomes increasingly mainstream – and valuable.

While a handful of family doctors have been using tools like AI scribes for years, Dr. Inderveer Mahal began relying on Heidi Health for summaries of her patient interactions last year, and is considered an early adopter of the technology.

“We’re often busy typing while also speaking to our patients, and it is so nice to be able to look at a patient, look at their body language, be focused on how I communicate versus also trying to type and document the visits,” she explained.

There are currently requirements to notify patients that an app is listening and generating a synopsis of a visit, but that can be something as basic as a sign that a patient may not notice. Mahal makes sure that she mentions it herself at the start of an appointment.

And while she has to review the generated summary to ensure it’s accurate and make corrections, Mahal said she saves hours per week, meaning “less screen time, less mental fatigue, less administrative tasks.”

A second set of eyes for diagnosis

The University of British Columbia now has a special hub for staff, students and clinicians working at the intersection of artificial intelligence, research, and health care, putting the university at the forefront of this new frontier.

“A few years ago, as the AI revolution took hold, there were some fears among some clinicians that they could replaced,” said UBC associate professor of biomedical engineering Roger Tam.

“Now, in many cases, it’s used basically as what they call a second reader, so the AI provides an opinion, but the radiologists are the clinician is still in the driver’s seat and they are the ones who still make the primary call.”

Tam explained that while radiologists are highly skilled at identifying cancers from medical scans, for example, machine learning algorithms are trained on thousands of images, which allows them to detect some serious illnesses before the patient shows any of the typical signs.

“These diseases can be asymptomatic for a long time,” he said. “(AI) is able to see things that humans can’t, that’s why the two work so well together.”

New medical school will incorporate AI

The founding dean of Simon Fraser University’s coming medical school in Surrey has been dabbling with artificial intelligence tools for a decade and expects that scribe summaries will be the norm within the next five years in family doctors’ offices.

That’s why Dr. David Price is already planning for incoming medical students to incorporate artificial intelligence technologies from day one.

“It’s going to be a core part of the curriculum, absolutely, and it’s going to be really embedded through everything that we do,” he said. “I’m sure every medical school is helping their students in their residence and their fellowship to understand how to use these tools responsibly.”

Price believes that aside from being a significant time-saver for clerical tasks, artificial intelligence software can be a powerful tool for researching symptoms and treatments – as long as it’s drawing from reputable, verified, reliable sources.

“So many times, a recommendation comes up and it may or may not be appropriate for you and in your particular life circumstances in your particular value set, your desires, for your own health,” he said. “We need to understand those nuances so that at the end of the day, (the treatment) is a decision between the patient and their physician or their clinician.”

This is the first part in a CTV Vancouver series taking a deep dive into the use of artificial intelligence in health care.