Research Chair in Artificial Intelligence and Digital Health for Health Behaviour Change

Drs. Simon Bacon and Éric Granger have been awarded a double research chair in artificial intelligence and digital health for health behaviour change. This innovative project is funded by Québec’s Fonds de recherche Santé and seeks to develop automatic emotion recognition algorithms to help health programs better personalize program content to address ambivalence, an important factor in helping people stay committed to health behaviour change interventions. This chair is also shared with Concordia University, ÉTS, and CIUSSS-NIM.

UPCOMING EVENT

One-Day Symposium – October 18, 2024

Incorporating Affective Computing to Enhance Health Assessments and Interventions

One-Day Hybrid Event – Open to students, academics, researchers, industry, and the general public

Expert talks from the fields of computer science and health science
Presentations from Montreal-area research groups working in the fields of health behaviour change and affective computing.
Poster session and an industry showcase

École de technologie supérieure (ÉTS) – 1100 Rue Notre-Dame Ouest, Montréal (Pavillon A)

Visit the event website

PAST INITIATIVES

Summer School 2023 – July 3 to 7

Concordia University – Montreal, Canada
Leveraging Artificial Intelligence to Optimise Behavioural Health Interventions and Assessments
Creating synergies between computer and health sciences

Chronic diseases like cancer and heart disease are major problems in Canada and the rest of the world.

These diseases, often called non-communicable diseases (NCDs), are largely caused by poor health behaviours, like physical inactivity and a bad diet. Changing behaviours is hard. Interventions that work on a person’s ambivalence, “I want to exercise more, but I don’t have the time,” tend to really help with changing a behaviour.

Online, or eHealth, behaviour change interventions are becoming very popular but, at the moment, they can’t measure ambivalence, meaning that they are not as successful as they could be. Part of the problem is that people often express ambivalence in non-verbal ways, such as a look of confusion, a shrug of the shoulders. New advances in artificial intelligence and the fact that most digital devices have cameras and microphones provides us with a great opportunity to be able to measure ambivalence in eHealth.

“The ability to identify when an individual has moments of ambivalence can allow us to adapt the content of the health program and provide a more personalised effective intervention.”

—Dr. Simon Bacon

Most automatic expression recognition (AER) systems include image-based recognition algorithms for a limited set of discrete emotion classes (e.g., happiness, sadness, anger) but to our knowledge there are no existing algorithms that can yet recognise discrete or continuous levels of ambivalence.

Using an existing eHealth behaviour change program (ACCELERATION), this new project will record videos of people doing tasks that create ambivalence in people. Using AI techniques for automatic expression recognition (AER), we will develop a way to recognise ambivalence.

Such videos are challenging because many different people will be recorded in the real-world conditions (people might be moving, not completely facing the camera, etc.) and will combine both visual and audio data. This project will develop ways to overcome these problems in order to accurately measure ambivalence. We will then be able to improve ACCELERATION in such a way as to adapt to an individual’s specific needs. These solutions also have potential to be used in other nonhealthcare areas as well.

“Applying affective computing in real-world healthcare applications is a relatively new field of research that requires a cross-disciplinary approach at the frontiers of machine learning, multimedia signal processing, and behavioural psychology.”

—Dr. Éric Granger

This collaboration will allow for the construction of the first affective computing system capable of reliably recognising human ambivalence and use this to create a more efficient and personalised cost-effective, eHealth behaviour change intervention, which, ultimately, could contribute to helping improve health and reduce the burden of non-communicable diseases.

This project will train ten students with the ability to understand both the behaviour change and AI aspects of the problem, and will also continue to work with patients, healthcare works, and partners in academia and industry. This program will lead to an eHealth behaviour intervention, which will be able to adapt better to individuals and be accessible to many people.

Dr. Simon Bacon

Simon Bacon is the CIHR-SPOR Chair in Innovative, Patient-Oriented, Behavioural Clinical Trials, FRQS Chair in Behavioural Medicine, and a Professor in the Department of Health, Kinesiology, and Applied Physiology at Concordia. Dr. Bacon is also the co-director of the MBMC, a CIUSSS-NIM research group. He is internationally recognised for his work on developing novel behaviour change interventions. Importantly, he has a long history of capturing multimodal physiological data and is the co-primary investigator of the ACCELERATION program.

Dr. Eric Granger

Eric Granger is the ETS Industrial Research Co-Chair on Embedded Neural Networks for Intelligent Connected Buildings (Distech Inc.), a Professor in the Department of Systems Engineering, and Director of LIVIA. His research expertise includes machine learning, pattern recognition, and computer vision, with applications in affective computing, biometrics, medical imaging, and video analytics/surveillance. His contributions on the development of deep learning (DL) models for video-based face recognition has led to several collaborations with governmental and industrial partners like CBSA, Nuvoola, Ericsson, and Genetec Inc. Dr. Granger is an associate editor for Elsevier Pattern Recognition, Springer Nature Computer Science, and the EURASIP Journal on Image and Video Processing.