We’re on the edge of a healthcare revolution, says Bryant University Assistant Professor of Information Systems and Analytics ML Tlachac, Ph.D. — and data science is leading the way. Amidst a global mental health crisis, new advances in how we see ourselves, and the tools we use to look, could be a lifesaving gamechanger.
“It’s incredible to watch history unfold — and to be able to help shape it,” Tlachac notes.
Tlachac’s research focuses on affective computing, which spans the intersection of human-computer interaction, machine learning, and mental health — and has led to a flurry of key publications in prestigious journals including IEEE Journal of Biomedical and Health Informatics and Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, many written in collaboration with psychologists, computer scientists, and other cross-disciplinary experts.
“This is a new field that has the potential to do so much good — and I see so much research that needs to be done,” Tlachac notes.
Tlachac’s most recent research includes a review of datasets utilized to assess depression using smartphone use information, published in IEEE Transactions on Affective Computing. An analysis of perceptions of AI chatbots within the Bryant community, conducted alongside other Bryant researchers and collected as part of a multi-center collaboration led by Ithaka S+R, was just accepted for publication by the Journal of Computer Information Systems.
“We are using sophisticated models, but there still needs to be a skeptical human in the loop to interpret them, and to question them.”
Powerful artificial intelligence and machine learning tools assist Tlachac’s efforts by helping to pluck out and discover key mental signifiers across a range of sources, from studying text logs to examining audio. “By using artificial intelligence, we’re able to find patterns in the data that we as humans would not necessarily be able to find — or even ever think of,” Tlachac says.
This work is also aided by the rise of wearable technology: smartphones and watches, fitness trackers, and other personal devices that generate vast quantities of information about our health, our habits, and how we live. “It's opened up so many avenues and opportunities for researchers,” Tlachac notes.
This new information, coupled with new ways of examining it, could lead to a sea change in how we approach health issues, particularly mental health. “You’re able to look at a person in comparison to the population as a whole. But you can also look at the person in relative comparison to themselves over time,” Tlachac says. “Right now, a lot of our health standards are set at the population level rather than the individual level. And I see a lot of this research as helping to move it towards more of an individualized, personalized medicine.”
The right signifiers could be used to pre-program wearables to alert care teams when certain criteria are met — or inform users of issues before they become crises. “If it notices you're socially isolating yourself, it could call a pre-programmed care team for you, or provide context-sensitive interventions such as suggesting: ‘Maybe this is a good time to take a few deep breaths or write down three things you're grateful for or go on a short walk,’” Tlachac offers. “It could help empower people to take more control of their mental health wellbeing.”
There are still mysteries to solve though. Every person is different, meaning that every person’s data is different; generalizations can be dangerous, Tlachac notes, which motivates the need for personalized medicines or interventions.
And as powerful and sophisticated as AI has become, it still requires a data scientist at the helm.
“We are using sophisticated models, but there still needs to be a skeptical human in the loop to interpret them, and to question them,” says Tlachac.