The ETS of the University of Quebec hosts two chairs on artificial intelligence in health

0
The ETS of the University of Quebec hosts two chairs on artificial intelligence in health

The École de technologie supérieure (ÉTS), part of the Université du Québec network, has announced the upcoming arrival of two new research chairs specializing in AI applied to health. The Fonds de recherche du Québec – santé (FRQS) is funding these chairs to the tune of C$1.5 million over three years. The purpose of this chair program is to train qualified personnel in AI so that they can work in a field related to health and AI.

The holders of these two research chairs will be Éric Granger, a professor in the Department of Systems Engineering, and Rita Noumeir, a professor in the Department of Electrical Engineering at ÉTS, which they will co-direct with their colleagues at the Université de Montréal and Concordia University. François Gagnon, director general of ÉTS, said of the arrival of these two chairs:

“ÉTS is the only university to have been awarded two research chairs under this program, which initially planned to fund only one for all of Quebec. This double award shows that our researchers have acquired very specific expertise in data science. Their scientific contribution will undoubtedly strengthen the international influence of this strategic centre. ÉTS is also about engineering for health, for a healthier future.

Rita Noumeir spoke around the use of FRQS funding to set up these two chairs:

“The funding we received will be almost entirely in the form of scholarships to students who will be able to develop their expertise in the field of artificial intelligence applied to health.”

First chair: on the development and validation of clinical decision support systems using AI

In this chair, Rita Noumeir and Professor Philippe Jouvet, attached to the University of Montreal and the Sainte-Justine UHC Research Centre, are seeking to address two issues:

  • Using data to improve care: ICUs contain a large amount of data that could be used to design AI algorithms and models to improve care. This data is collected from patients who consent to its use for research purposes, and will be overseen by an ethics committee.
  • Data processing to support decision making: several types of data exist in the medical world. It may be laboratory tests, physiological signals, radiological images or medical notes. All of this content will be subjected to new data processing methods to support decision making in healthcare.

By taking these two issues into account, they want to develop a powerful algorithm that can assess a patient’s condition and level of distress in real time. The researchers also want the model to be able to reduce readmission rates to intensive care and manage the flow of patients between care units. The goal of this solution will be to help health care professionals and managers make faster decisions in a care setting.

Second Chair: on AI and digital health for changing health behaviours

In the context of this second chair, a problem was raised: “How can we help people follow a treatment plan or adopt healthier habits when they use an online health service without human intervention?”. To answer this question, Éric Granger and Simon Bacon, professor of behavioural psychology at Concordia University and researcher at the CIUSSS du Nord de l’île de Montréal Research Centre, will work together.

The research team hopes to develop an algorithm that will be able to interpret the non-verbal language of users, which provides subtle clues about a person’s ambivalence, which is not often taken into account during an online intervention. This would allow care services to adapt their actions to suit the user’s emotional state. According to the experts, there are two main areas of work to focus on:

  • Researchers will analyze a large amount of multimodal data from videos. Deep learning models will assign an emotional state to a combination of visual or audio data including voice intonation, gesture, posture or facial expression.
  • They will seek to improve deep neural networks (DNNs) so that they can more easily recognize expressions. According to HTA, DNNs tend to degrade when there is a small amount of data and a diversity of sources.

All of this research will aim to develop tailored interventions that will impact on health behaviors that may represent the increase in chronic non-communicable diseases.

Translated from L’ETS de l’Université du Quebec accueille deux chaires autour de l’intelligence artificielle dans la santé