Urgences-sante to use artificial intelligence to optimize paramedic care

Ambulances are parked in downtown Montreal on Saturday, Nov. 19, 2022. (Joe Lofaro/CTV News)

Urgences-santé and Airudi, a Quebec company specializing in the design of AI solutions, announced on Thursday a partnership to optimize pre-hospital care using artificial intelligence.

"The aim is to provide the Urgences-santé team with intelligent dashboards that will give them access to real-time data and help them make the best decisions," Amanda Arciero, co-founder and vice-president of operations at Airudi, told The Canadian Press.

Urgences-santé provides paramedical care to 2.5 million people in Montreal and Laval annually. The company receives around 1,000 calls a day, representing around 36 per cent of ambulance requests in Quebec.

This demand is expected to increase by 10 per cent in Montreal and Laval, putting additional pressure on the workload of paramedics.

Urgences-santé believes that its partnership with Airudi will enable it to better predict demand, leading to better planning and management of the paramedic workforce.

To this end, the MODUS platform will analyze historical data, real-time information and other factors that can impact demand, such as local demographics, weather and road conditions.

A sophisticated machine learning algorithm will integrate these various elements to develop a predictive model of call volumes, optimizing workforce planning and proposing optimal deployment scenarios to meet the expected level of service, according to a press release.

But it will first and foremost be a matter of predictions, stressed Arciero.

"The comparison we sometimes use is that of the weather, so the user must have a tolerance threshold," she said. "Weather forecasts aren't always perfect, but it's still a good indicator and you can tell whether you're bringing an umbrella tomorrow or not. So it's really the same dynamic that we're putting in place through the use of our products."

With this solution, Urgences-santé seeks to equip itself with a decision-making aid that will "ensure an immediate or virtually immediate response to the population" in the event of need, said its Deputy Director General - Executive, Mathieu Campbell.

"Urgences-santé is present at the time (the event) occurs or based on trends over the last few years," he said. "But what we're trying to do here is plan ahead to match supply with demand."

Urgences-santé obviously has the capacity to respond to urgent calls, said Campbell, "but it's difficult to know whether we're going to see a marked increase in calls in the next hour or in the next few days."

"Of course, if there's an ice storm, we'll be able to tell and then we'll structure ourselves," he said. "But beyond that, there are other factors, such as demographics, that really influence response capacity, and that's where we think this solution will enable us to get ahead of the game ... and optimise our response."

The employees currently in place are experienced people who are able, based on data such as current conditions or trends over the last few years, to "manage their workforce well," but they also rely heavily on what Campbell amiably calls "their gut feeling."

"But we're in a social context where there will be a lot of retirements and, therefore, a lot of loss of knowledge," he added. "There are also trends that are going to change and that we hadn't seen before."

Arciero and Campbell also agree that artificial intelligence is "not at all" replacing human expertise but complementing it.

"Current employees have access to a wealth of information, but they have to process and analyze it, and there's a limit to what they can do in terms of day-to-day management," said Campbell.

"The dashboard will support decision-making,' Campbell said. "It doesn't replace the human element; it supports it in its operations. The solution will provide a complete picture so that the right decisions can be made.

One of the aims is to strike the right balance between the number of ambulances available to respond to emergencies that can't wait and other priorities," he added.

"This partnership is arousing the curiosity of several other ambulance services, including those in Alberta and Nova Scotia. We could also consider adapting the solution to the hospital environment, for example, to anticipate emergency room traffic," Arciero said.

"It could be possible, subject to a number of conditions, including the availability of data," she explained. "It's often a challenge for certain organizations to be mature when it comes to data availability. But yes, it's certainly something that could be transferred to the health-care sector."

MODUS will be rolled out progressively over the coming months, with a full roll-out planned for the second quarter of 2026.

This report by The Canadian Press was first published in French on Sept. 12, 2024.