Projets novateurs réalisés

Urban Fire Risk Assessment

Predictive Fire Incident Risk Scoring System aims to develop a comprehensive risk assessment tool tailored to urban neighborhoods. Utilizing a combination of Census data, historical fire incident records, building structures, and expert knowledge, the project seeks to create a predictive risk scoring system capable of quantifying the overall fire incident risk level for each neighborhood. By analyzing diverse datasets and employing machine learning algorithms, the system aims to provide valuable insights into urban fire risks, enabling proactive measures to enhance public safety. Collaboration with Darkhorse Emergency presents an opportunity to expand expertise in public safety assessment and offer innovative solutions. Integration of the resulting model into Darkhorse’s data analysis software tools will enhance the company’s market presence and benefit fire agencies across North America.

Quantum Computing to enhance Machine Learning Accuracy

in this project, we are exploring the combination of quantum computing with machine learning to ensure a quantum-enhanced machine learning. We are focusing on the improvement of the accuracy and reliability of machine learning algorithms. While quantum improvements in supervised, unsupervised learning, and reinforcement learning have been reported, distributed models of machine learning such as Federated learning have received much less attention. Within our project, we tackle the problem of quantum enhancements in federated learning as well as in a set of specific use cases, and we will propose a systematic scheme for providing accuracy and reliability improvements.

Sea Cucumber as a source of active ingredient for cosmeceutics

The research project involves studying how extracts from sea cucumbers can be used in skin care products. Sea cucumbers are marine animals known for containing natural substances that might be beneficial for skin health, like slowing down skin aging or lightening skin tone. The project has two main goals: first, to make sure these sea cucumber extracts are safe for skin use and to find out what good they can do for the skin; and second, to figure out how to mix these extracts into skin care products effectively.

Pharmacology of GBA1 Chaperones

Mutations in the enzyme glucocerebrosidase (GBA1) are the most common genetic risk factor for development of Parkinson’s disease (PD). PD is characterized by the buildup of abnormal protein deposits in the brain, followed by progressive loss of neurons and behavioural symptoms. Numerous studies have noted a correlation between reduced GBA1 activity and increased levels of these abnormal protein deposits in the brain, but the relationship remains poorly understood. The aim of this project is to investigate whether augmenting levels of GBA1 can prevent the accumulation of abnormal proteins in the brain. The success of this project may lead to a better understanding of the causes of PD and could ultimately result in new strategies to develop disease-modifying therapies for Parkinson’s disease.

Integration of a somatotopic non invasive proprioceptive feedback strategy for robotic hand prosthesis users

The aim of my PhD research project is to provide hand proprioception perception to users of robotic hand prostheses. To do so, we developed a system to convey information about the configuration and movement of artificial limbs via evoked sensations that target specific regions of the phantom hand, while involving no perceptible delay or surgical intervention.

This research internship represents an invaluable opportunity for us to assess the effect of introducing the developed system on users’ technological embodiment of a robotic prosthesis. It is expected that the introduction of this type of proprioceptive feedback will alleviate some of the problems associated with the use of robotic hand prostheses, and thus promote the integration of these devices by the user. The expertise and research interests of Plasticity Lab’s researchers, led by Professor Makin, make them ideal partners with whom to collaborate in this endeavor.

PetDrifts Market Launch

PetDifts is launching smart collars to the market. This project includes refining product application to get ready for customers, spread the word out using digital marketing and partner with retail pet shops.

Electrodialysis Process for Produced Water Treatment and Purification

Arrowhead is partnering with the SAIT Centre for CERCUTS research group to develop a novel process that takes produced/waste water (PW) for the production of clean, fresh water and hydrogen using new generation Electrodialysis (ED) technology. This proposed research project aims to design, build and test a process skid incorporating a pretreatment system to remove solids and excess organics from produced/waste water, and an ED system to remove salts and dissolved solids while producing hydrogen as a byproduct. The key objective is to develop and test a pilot scale of New ED method with collection and monitoring of the produced green hydrogen from produced/waste water with the added benefits of producing clean fresh water and reducing the amount of produced/waste water sent to disposal by injection. The results of this project are expected to set a foundation for a scalable process, equipment specification parameters and operation conditions for the technology to be deployed at Arrowhead’s sites. An additional but equally important objective of this project is bringing technology development and training to Indigenous communities, incorporating First Nations companies and individuals in the process.

Developing LLM Chatbot to Bridge from Lab to Analysis

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

Investigate machine learning algorithms to detect anomalies in computing infrastructures in real-time

Metafor is developing a new class of IT system management solution to monitor computer and application activities, and alert when anomalous behavior occurs. Current commercial tools for anomaly detection use simple statistical rules and thresholds to detect anomalies. These methods are failing for today’s dynamic cloud environment where change is constant. As a result, IT operators are flooded with false alerts; become overwhelmed with alert fatigue and learn to ignore the alerts. This research internship will help to identify and create appropriate anomaly detection algorithms for the IT environment and extend and improve the functionality of Metafor’s current suite of algorithms.

CRITAC : Modèles d’estimation de pose pour les arts du cirque

Les compagnies de cirque au Canada innovent en intégrant des technologies numériques comme la capture de
mouvement dans leurs spectacles, créant ainsi des expériences virtuelles et hybrides. Toutefois, les systèmes de
capture actuels avec marqueurs ne sont pas adaptés aux performances acrobatiques. Le projet vise donc à tester
différents modèles d’estimation de mouvement pour trouver ceux qui conviennent le mieux aux mouvements
uniques des artistes de cirque. Cette technologie pourrait ouvrir de nouvelles possibilités créatives pour
l’enseignement, la recherche et le divertissement, permettant aux compagnies et artistes québécois de se
démarquer par leur innovation dans le domaine du numérique.

Calcul des performances d’un fuselage intégré hybride en régime transsonique

En 2015, la NASA, Boeing et Pratt & Whitney ont entrepris des études expérimentales et numériques pour calculer les performances d’un fuselage intégré hybride (HWB), désigné N2A-EXTE 0009H1, pour tous les régimes de vol, dont celles en transsonique. Le but de ces études était de vérifier un concept d’intégration moteur avec cette cellule d’avion non-conventionnelle. Le but du stage est de reconstituer dans un modèle CAO (Catia V5) la maquette utilisée pour ces études antérieures en se basant sur les documents publiés par ces trois organisations. Le modèle doit être construit en partie sous forme paramétrique de façon à que le modèle soit prêt à une mise à jour automatique selon les choix de paramètres décidés par un utilisateur(trice) ou une boucle automatique d’optimisation de formes aérodynamiques autour des moteurs. Une fois le modèle CAO reconstitué, des simulations numériques de l’écoulement seront obtenues à l’aide d’un solveur fluide (StarCCM+ ou autre), suite à la discrétisation du volume fluide autour du modèle de l’avion à l’aide d’un générateur de maillage (Pointwise). L’objectif est de valider par itération que le modèle est fonctionnel côté aérodynamique, c’est-à-dire débarrassé de toute anomalie dans l’écoulement d’air autour de l’avion.

Large Language Models for Improving Quantum Communications and Networking

The aim of this research project is to explore the use of Large Language Models (LLMs) in quantum technologies for computer network system scenarios. The hypothesis is that we can easily comprehend and utilize quantum algorithms to propose solutions for networks by leveraging Large Language Models technologies in specific tasks. Moreover, another interesting question that we want to find the answer to is if we can safely transmit qbits of information between quantum network systems using the (QKD) paradigm with Large Language Models helping.