Projets novateurs réalisés

Using AI for Chronic Pain Assessment and Malingering Detection

Pain is an unpleasant feeling and a significant health issue. A 2021 report by the Canadian Pain Task Force indicates that one in five people in Canada live with chronic pain, many of whom report it adversely affecting some or most daily activities. Chronic pain also has a tremendous impact on the Canadian economy. Additionally, the lack of effective and specialized pain services, long waiting times, and the scarcity of trained pain management professionals exacerbates the hopelessness for people who live with pain. This reduces productivity due to unemployment and absenteeism among individuals living with activity-limiting chronic pain. Hence, pain assessment is essential to providing proper patient care and assessing its efficacy under clinical settings. To assess pain, physicians typically rely on patients’ statements, which are subjective, or imaging technologies, which are unreliable as detected abnormalities do not always indicate pain and are not always present in cases of chronic pain. The Karmy Clinic specializes in assessing and managing patients with chronic non-malignant pain. For nearly two decades, Karmy Clinics have performed medicolegal assessments for several clients including lawyers and medicolegal assessment centres. Karmy Clinics currently process about 40 of such assessments per month representing over $100M in assessments, legal processing fees, and claim payments each year. Leveraging AI could improve the accuracy and efficiency of pain assessment and detection of malingering. This is the central focus of this project.

Find My Bed: An AI Navigator for Babies Requiring Immediate Hospital Transfers

Every year, ~15,000 newborns in Canada who are critically ill are admitted to 31 Level 3 neonatal intensive care units (NICU). These newborns are at a high risk of death and illness. Over 65% of these newborns are not followed before birth in a hospital with a Level 3 NICU and need to be transferred either prior to birth (ie the pregnant mother) or after birth to a hospital with a Level 3 NICU. This transfer process is time-sensitive and requires the doctor to find a bed in the unit. This process can take up to 35 minutes, and there is a chance that the decision made may not be the optimal one.

To improve the process, we propose to develop Artificial Intelligence driven Decision Support Application that will help identify the most appropriate hospital for the transfer. This will be done through collaboration with experts and stakeholders. The project includes experts in neonatal health, nursing, process optimization, and national networks, as well as stakeholders such as the Ministry of Health. The expected outcome is to improve the efficiency of decision-making and better distribute admissions of critically ill newborns, which can have a significant impact on the overall system.

Refonte et élargissement du champ d’application de l’outil «Évaluation des coûts économiques des bris d’infrastructures souterraines»

Les infrastructures souterraines sont essentielles pour les services publics. Toutefois, en 2022, le Canada a enregistré une moyenne de 42,4 bris d’infrastructures souterraines causés par des tiers par jour de travail, posant des risques pour la sécurité et générant des coûts importants. Les coûts directs de réparation sont identifiables, mais les impacts sociaux et environnementaux indirects, tels que les interventions d’urgence, les perturbations de circulation, et les interruptions de service, restent difficiles à évaluer et sont rarement pris en compte dans les décisions de travaux ou de prévention.
Face à cette problématique, le CIRANO a développé en 2015 un outil pour le Canadian Common Ground Alliance (CCGA) permettant l’analyse socio-économique des dommages aux infrastructures pour quatre provinces. Aujourd’hui, une mise à jour s’impose pour tenir compte de l’évolution du contexte économique, rafraîchir les données et hypothèses, et adapter l’outil aux spécificités d’autres provinces.
Cet outil est important et pertinent car il comble un vide en fournissant une évaluation des conséquences socio-économiques des dommages, essentielle pour élaborer des mesures de prévention efficaces et orienter les décideurs dans la mise en place de politiques adéquates. Cela renforce ainsi la gestion globale des infrastructures souterraines.

Perceived impact of Peer Advisory Conversation participation on early career and newly licensed Ontario veterinarians

The College of Veterinarians of Ontario (CVO) is implementing a new initiative to assist new licensees, including early career veterinarians and newly licensed Ontario veterinarians. This initiative involves peer advisors appointed by the CVO to work collaboratively with new licensees through a program called the Peer Advisory Conversation (PAC). Working with Dr. Gohar, Dr. Brunt will conduct interviews with new licensees and focus groups with the peer advisors who mentor the new licensees. Interviews and focus groups will be facilitated by Dr. Brunt, audio recorded and these recordings will be transcribed verbatim by an external company. Transcripts will be inductively analyzed by applied thematic analysis. This analytical method borrows from numerous traditional qualitative methodologies and was selected as it is in alignment with the applied nature of the project where credibility of findings to an external audience is paramount.

Machine Learning developer and Product interns working within cross-functional teams to develop and commercialize AI-powered solutions in both the Public and Private sectors (1)

The project comprises internships in a variety of technical and business roles within our organization. Within the organization, roles include associate machine learning developer, product management associates, associate business solutions consultants, communications associates, and finance associates. Within GovLab specifically, roles include associate machine learning developer, associate business solutions consultant, and project delivery associate. The difference is mainly that in GovLab, there is a focus on public sector problems, whereas in AltaML overall, we work across all sectors.

From farm to fork: Refining polyhydroxyalkanoate bioplastics from upcycled dairy waste

The attractive physicochemical, thermal, and mechanical properties of PHAs, combined with their biodegradability and biocompatibility, make them exciting materials that can support an environmentally sustainable circular bioeconomy. In this research project, we plan to develop an extraction and purification process for recovering PHAs from bacteria. The recovered PHAs will be further characterized with a focus on developing sustainable bioplastic materials.

Identification et caractérisation des propriétés viscoélastiques d’un point trigger localisé dans le muscle gastrocnémien à partir d’une technique d’élastographie par ondes de cisaillement

La douleur d’origine myofasciale (DMF) serait la principale cause des douleurs musculosquelettiques dans le monde et affecterait entre 30 et 85% des individus au courant de leur vie. Or, cette pathologie est actuellement très difficile à diagnostiquer car il existe aucune méthode d’évaluation objective permettant de quantifier les changements que l’on retrouve dans les muscles des personnes qui en souffrent. La DMF est caractérisée par la présence de petit nodules (nœuds) douloureux dans les muscle que l’on nomme les points gâchettes. À ce jour, l’identification de ces nodules s’effectue par palpation, une méthode d’évaluation très subjective. Il existe un besoin criant de valider un outil de mesure qui pourrait caractériser et quantifier les points gâchettes afin de supporter le diagnostic de la DMF et préciser les interventions qui visent cette condition de douleur chronique.

L’échographie par onde de cisaillement pourrait s’avérer être une solution à ce problème. Cette méthode de mesure non invasive et sans douleur a fait ses preuves pour d’autres pathologies.

Le but de cette étude est de vérifier si il est possible d’identifier, de quantifier et de caractériser un point gâchette localisé dans un muscle du mollet à l’aide de l’échographie par onde de cisaillement.

AI-Driven Insights from Knowledge Graphs via Integration with Large Language Models

In the current digital world, users are constantly sending companies signals of their experiences and preferences in various shapes and forms, including likes/dislikes of a particular offer/product, reviews on platforms and social media, and contacting customer service. However, despite the abundance of signals, E-commerce companies are unable to rapidly gain meaningful and actionable insights from those signals to react to the customer experience in a timely manner.

This project aims to address this challenge by leveraging knowledge graphs and Large Language Models (LLMs) to build an AI-based insights engine with a feedback loop to address the challenge of leveraging centralized customer signals/data from multiple sources to generate actionable insights that inform various product initiatives, campaigns and experiments across the organization to improve customer experience and generate business value.

The overarching goal is to provide a tool with an interface that allows business users to set their objectives or priorities and either automatically or manually prompt for hypothesis recommendations related to their product or campaign and receive insights for a complete set of hypotheses while accounting for overlapping initiatives within the company, enabling rapid experimentation.

Projet PISE- Pratiques d’Intégration Saines des Écrans- volet écoles primaires

Ce projet propose d’étudier les usages appropriés des écrans en classe à l’intention des enseignants qui doivent composer avec une présence de plus en plus importante des écrans à usage individuel. Il est développé en parallèle avec autre projet de recherche portant sur les répercussions des usages de l’écran à la maison que nous menons en partenariat avec le FRQSC et la Fédération des comités de parents (FCPQ). Les résultats émanant de ces deux projets de recherche nous permettrons d’avoir un portrait complet de l’usage des écrans en milieu familial et en milieu scolaire.
Nous proposons de 1) Sonder les usages que les enseignants et le personnel éducatif font des écrans et de 2) Analyser ces pratiques aux regards des recommandations canadiennes et québécoises sur l’usage des écrans.
Les conclusions de ce sondage et de cette analyse de pratiques permettront de situer les pratiques enseignantes aux regards des recommandations et de développer un guide de pratiques sur l’utilisation saine des écrans en classe.

Research Proposal: Understanding and Mitigating the Risks of Generative AI in Propaganda and Informational Warfare

THIS IS A GENERIC TEXT PUT IN PLACE AS THERE WAS NO PROJECT OVERVIEW

FireBozz – Wildfire Prevention Sprinkler Nozzle System Development

As climate change intensifies, the frequency and intensity of wildfires have surged, demanding innovative solutions to mitigate their destructive impact. In response to this urgent need, FireBozz seeks to develop a specialized sprinkler head tailored specifically for wildfire suppression, aiming to revolutionize firefighting strategies and enhance community resilience. FireBozz, a leading provider of rapidly deployable wildfire fighting equipment, recognizes the inadequacy of traditional agricultural sprinkler systems in effectively combating wildfires. To address this gap, FireBozz envisions the design of a wildfire-specific sprinkler head targeted at effective water usage and dispersion to vulnerable areas to protect assets and prevent the spread of fires. By collaborating with applied research, iterative design, and field testing, FireBozz aims to pioneer a groundbreaking solution that safeguards communities and protects assets from the ravages of wildfires.

Graphene protected silicon alloy-based Li-ion batteries

The proposed research aims to advance energy storage technology by improving the cyclability of silicon-based anodes in lithium-ion batteries. Silicon anodes offer high theoretical capacity but face issues such as volume expansion and capacity fade. The project will explore substituting pure silicon with a silicon-carbon alloy core, hypothesized to enhance cyclability due to its amorphous structure and oxidation resistance. The research involves synthesizing silicon-carbon alloy particles, integrating them into shell/core structures, optimizing electrode casting, and testing electrochemical performance. Successful outcomes could significantly boost the energy density and commercial viability of silicon-based anodes.