Projets novateurs réalisés

L2M – UTI Diagnostics Market Research

Urinary tract infections (UTIs) are one of the most common bacterial infections globally, affecting millions of people each year. The inappropriate use of antibiotics for UTIs is a significant driver of antibiotic resistance, posing a grave threat to public health. Complicated UTIs, characterized by treatment failure, recurrent infections, and increased mortality, disproportionately affect vulnerable populations, including older adults and individuals with neurogenic bladder conditions.

This project seeks to address these critical issues by conducting comprehensive research and developing innovative solutions. The project’s key objectives are to uncover the challenges experienced by healthcare providers, patients, and laboratory professionals in the UTI diagnostic process and to analyze the healthcare infrastructure, regulatory framework, and access disparities in North America concerning UTI diagnosis and treatment.

Évaluation statistique et prévision de la performance de gestionnaires d’actif

Ce projet porte sur l’évaluation de la performance de gestionnaires de fonds de placement. Plus spécifiquement, dans ce projet on cherche à développer des outils qui permettront à la compagnie de déterminer, à partir de données sur ses rendements passés et sur le rendement d’indices de marché appropriés, si un gestionnaire d’actif a été significativement plus performant dans ses placements que l’indice de marché en question. L’objectif principal de ce projet est qu’à la fin du stage, l’entreprise ait à sa disposition un outil simple (par exemple un fascicule composé de tableaux) qui lui permettra de déterminer le rendement moyen requis sur une période visée pour que l’on puisse déclarer qu’un gestionnaire est significativement meilleur qu’un indice de marché visé. Cet outil sera utile à la compagnie, car elle pourra l’utiliser pour mesurer de façon objective la performance des gestionnaires qui offrent leurs services pour gérer ses actifs.

L2M – A handheld bioprinting platform for treating severe skin wounds

Our project focuses on developing a bioprinting technology for treating severe burns. We aim to reduce scarring and enhance skin regeneration, which can significantly improve the lives of burn victims. This technology has the potential to make a big difference in the medical field, but it’s a challenging journey, especially due to the strict regulations and financial requirements. We’re partnering with L2M to gather feedback from experts, refine our business strategy, and make the right industry connections. In return, L2M will benefit from our progress in the medical field and our engagement with various experts, potentially creating opportunities for knowledge exchange and collaboration.

The influence of the wood morphology on the stability of a plasma process at atmospheric pressure for fire-resistance

Wood is an eco-friendly material but is very susceptible to fire, therefore, treatments using organosilicon products can be carried out to make the wood more resistant to fire. The way Laroche’s group has carried out these treatments is through the application of plasma to the wood surface and subsequent application of the organosilicon product. In this context, understanding the surface changes in wood morphology is an essential step to understanding the effectiveness of the treatment and promoting its industrial application. This project aims to address the need for new advanced solutions to reduce the risk of flame spread in wooden buildings and modern timber-framed structures. Additionally, this project will facilitate the establishment of a research, education, and industry initiative at the intersection of plasma technology and wood science. The combination of engineering and scientific expertise employed in this project will offer a fresh perspective on addressing this global challenge. This work will empower students from both centers to create value, innovate, and become the leaders of tomorrow. The consolidation of this long-term partnership will also offer the possibility to continue this strategic research and will provide new solutions for the Canadian and Brazilian economy.

L2M- Unobtrusive Monitoring of Heart Failure via Ballistocardiograph

In Canada, heart failure (HF) has a 30-day re-admission rate of 17.6%. While patient care at home via follow-ups and self-management may reduce this rate, factors including physical fatigue and cognitive frailty delay follow-up visits and impair the patient’s ability to self-manage. An implantable solution such as CardioMEMS is available to monitor HF, however, its invasiveness and high cost (>$19,000) make the solution largely inaccessible. This project will study the use of smart home sensors to monitor HF patients and detect signs of decompensation autonomously during daily activities such as sleeping. At the core of the technology is a bio-signal called ballistocardiograph (BCG), which is a recoil force generated by the heart as it ejects blood. Our research has shown that BCG signals differentiated the physiologic state of 29 younger healthy adults and 26 older adults with moderate HF. Technical advances have also enhanced BCG signal acquisition from a single-axis to three-dimensional. We will evaluate data acquired from BCG sensors embedded in a conventional bed compared to contemporaneously acquired invasive heart chamber pressure measurements in a sample of HF patients. The system will be tested on HF patients with a range of health states, from ambulatory outpatients to hospitalized inpatients. The technology’s ability to detect the improvement after the treatment will be assessed, the success of which will validate the technology’s effectiveness as a HF monitoring tool. We will also evaluate and project the economic and health equity impact of the technology.

Modèles pour la prévision de la durée d’une police d’assurance

Dans un marché très compétitif comme celui de l’assurance, estimer la durée de l’association potentielle entre un nouveau client et la compagnie est un enjeu majeur. Ceci permet à l’entreprise de cibler de futurs clients potentiels qui seront plus susceptibles d’augmenter sa rentabilité en répartissant les coûts dus à l’émission du contrat sur une plus longue période. Une telle estimation permet aussi à la compagnie d’utiliser des incitatifs afin de retenir des clients acquis lorsqu’ils deviennent à risque de terminer ou de rompre leur contrat. Dans ce projet, nous visons à développer des méthodes qui permettront à la compagnie de déterminer avec précision, et ce dès le début d’un contrat, si un client est plus susceptible de quitter en cours de police ou de ne pas renouveler à échéance, et de modéliser le risque de quitter la compagnie en fonction de la durée de la police.

Development and application of a GIS-based lake assessment andmanagement system

A GIS-based water quality assessment approach will be developed in this study, and it is based on a comparison analysis using USA, Canadian and Chinese water quality index models. The method is developed within Geographic information system?GIS? framework. It has the functionalities to manage a large amount of georeferenced input data, to run different water quality index models for a real case study, to compare and verify the water quality assessment results, and to visualize the final results within the GIS platform. The proposed approach will be applied to real lake in China supported by the Globalink Program. Water quality assessment results based on widely used index models (US-OWQI, Canada-CWQI, and China-CNWQI) will be verified with each other for the final assessment conclusions. The study results will be expected to extend the research on new effective lake assessment and management technologies.

Enhancing Fine-Tuning Techniques of Large Language Models (LLMs) in the Insurance Industry

The landscape of the insurance industry has undergone rapid transformation in recent decades. A driving force behind this transformation is the exponential growth of data volumes, necessitating effective processing to unlock their full potential. This surge has led to the development of advanced processing technologies, much like how AI is harnessed to transform the insurance sector, leading to improvements in customer experiences, risk assessment, operations, and fraud detection. Insurance companies can leverage large language models to employ big data analytics, gathering and analyzing structured and unstructured data from diverse sources. This process enhances risk assessment accuracy and enables personalized offerings. By comprehending customer behavior, insurers create customized policies and targeted marketing, elevating customer relationships and loyalty. The adeptness of big data in swiftly spotting unusual patterns aids fraud detection and loss prevention. Predictive analytics anticipates customer needs, enhancing service quality and satisfaction. Additionally, big data optimizes claims processing, leading to quicker response times. The proposed research is aimed at developing automated RL-based innovative fine-tuning strategies tailored to address the above distinctive challenges and the one within the insurance industry.

Exploring co-location and social presence in augmented reality

Creative research to investigate what types of Augmented Reality experiences cause users to feel co-presence. A set of prototypes will be created that explore different ways that users can experience presence in virtual worlds at the same time as other users. A publicly accessible artwork in Victoria, BC, will be created based on learnings from the prototypes.

L2M-A Novel Titanium-Enriched Bioactive Glass to Enhance Enamel Mineralization, Whiteness, and Hardness.

This project is aimed at supporting the translation of research innovation from the laboratory into the market. Our proposed innovation involves the use of a newly improved biomaterial in toothpastes to help transform oral health and hygiene. The proposed toothpaste will be effective in addressing tooth caries and discoloration, and simultaneously increase enamel hardness to prevent future caries. The proposed toothpaste poses zero concerns about damage to the teeth and oral environment. With Our toothpaste, a simple brushing step is all that is required to obtain a healthier enamel and brighter smile. The project is beneficial to the partner organization in meeting one of its core objectives: supporting the commercialization of post-secondary research.

A research on Liang Shuming’s reciprocal centeredness

Liang Shuming’s”reciprocal centeredness”shows us a new way of gaining rights, he suggests that the realization of one’s personal
rights will be naturally entitled via his fulfillment of necessary obligations.It’s likely that people lead a meaningful life among
harmonious interactions rather than through the “outward demand” of rights,though the status he assigned to rights needs to be
reexamined.Liang’s pre-modern narratives induce him to neglect the potential tension between rights and obligations.He also
simplified a community as a mono-opinion entity, ignoring the potential conflicts between individuals in a group. So the two tasks I
hope to accomplish are to modernize the idea of ” reciprocal centeredness ” and to optimize relevant narratives in the relationship
between group and individuals.

Mastering Multi-modal Decision-making with World-Models

The quest for building an artificial intelligence (A.I.) that is general and able to solve multiple tasks, as humans do, has experienced a growing interest in the last few years. Current A.I. systems can communicate with humans and provide sensible answers and ideas in the form of text. However, if we expect A.I. systems to be pervasively used in real-world applications, we need to provide them with the ability to sense, interpret, reason, and act in the world. This project aims to bridge the gap between A.I. language systems and reality, by employing world models, a recent idea that has been applied in many decision-making tasks, such as video games, simulations, and robotics. By successfully combining language systems with world models, the project aims to build generalist A.I. systems, which have the potential to be adopted for solving numerous real-world tasks, including visual question answering and web-browser based tasks, two challenges that are being extensively researched at ServiceNow.