This text is part of the special Mental Health notebook
Kids Help Phone has just launched a partnership with the Vector Institute for Artificial Intelligence to optimize its online services across the country.
“Use of our services has more than doubled since the start of the pandemic and continues to increase,” reports Darren Mastropaolo, vice-president of innovation and data at Kids Help Phone (JJE), the pan-Canadian service. which offers bilingual professional intervention services, resources and online support to young people in matters of emotional and mental health at all times. For him, innovation, including artificial intelligence, is therefore essential to meet demand. “Thanks to AI, we can ensure that we respond to the most urgent needs first,” he emphasizes, at a time when his organization has just joined forces with the Vector Institute, a leader in the field, in order to develop and improve its services.
As Darren Mastropaolo reminds us, JJE is based on a human-to-human relationship first and foremost. “We are turning to artificial intelligence and machine learning to assist our staff,” he says. As for the Vector Institute, its vice-president for health, Roxana Sultan, affirms that there is no question of replacing people. “Young people will not only have to deal with a chatbot, but the goal of our partnership with JJE is to increase the number of services provided while maintaining the human side,” she says. In fact, this alliance will allow respondents to be more precise in the assistance provided and to support existing systems.
Identify and act quickly
“The project we are currently working on aims to make it easier and faster for front-line staff to identify the problems of the person on the other end of the line,” explains Joanna Yu, from the health data department. from the Vector Institute. “We want to reduce the cognitive load on staff who must quickly find out what clinical problem the young person is facing in order to connect them with available local resources,” she adds.
By creating a tool using AI and machine learning that helps determine the problem for which a young person is contacting JJE — such as suicidal ideation, depressive and anxious thoughts, etc. —, the forecasts, and even the predictions, will thus be optimized. “Advisors will be able to put the appropriate protocol in place as quickly as possible and offer personalized service to young people in need,” believes Roxana Sultan.
“In general, all of this is correlated to certain words or certain phrases used by young people when they communicate with JJE advisors,” she explains. To do this, what is exploited from a technological point of view is natural language processing. “It is a way of analyzing language to better understand the context of words. With the data we have collected so far, we can establish links between the use of certain terms and potential danger,” adds Roxana Sultan. For her colleague at the Vector Institute, since language is constantly evolving, the challenge is not the least. “We can develop a particular taxonomy in real time to provide even more support, but ultimately it’s always the human who will make the final decision,” says Joanna Yu.
The constraint of anonymity
“Respecting confidentiality is essential to Kids Help Phone,” says Darren Mastropaolo. He continues: “Respondents and counselors will not share the information they obtain outside of Kids Help Phone unless they are concerned about the safety of young people. » This is intended to be reassuring. “All data shared publicly is anonymous and aggregated, meaning it is an overview of conversations that does not allow individuals to be identified,” he explains. And Roxana Sultan assures: “We work very hard to secure everyone’s information, so that no one can be identifiable. »
Furthermore, the anonymity factor is a real challenge for the Vector Institute. “When we research health data, we try to create artificial intelligence or machine learning that finds solutions. But where it gets complicated, in our partnership with Jeunesse, j’oreille, is that we cannot know if a person has already called several times,” explains Roxana Sultan.
If, in a hospital, it is easy to follow the trajectory of patient data, even if it is confidential, this is not the case for JJE. “Young people can provide additional information, but it is optional,” mentions Joanna Yu. The development of the tools therefore depends on data collected on a voluntary basis. “There is an additional challenge in mental health, because there is not really objective biological data, such as blood markers. It’s more complex to do early detection,” she continues. Regardless, JJE and the Vector Institute are on a mission to prove that AI can have a positive effect on mental health, in a very practical way.
Need help ?
This content was produced by the Special Publications team at Duty, relating to marketing. The writing of the Duty did not take part.
