Projets novateurs réalisés

Disinfection of wastewater effluent in low resource and humanitarian contexts

Wastewater is one of the primary point-source contaminants polluting freshwater sources, including shallow groundwater sources. Over 80% of wastewater worldwide is neither collected nor treated and more than 70% of sewered wastewater from human activities is discharged without any form of pollution control, increasing the risk of waterborne diseases. This project, in collaboration with Centre for Affordable Water and Sanitation Technology (CAWST) will address a critical public health issue by evaluating potential onsite wastewater disinfection solutions integrating UV light-emitting diode (LED) technology for disinfecting wastewater. Experimental trials using UV LED reactors that are manufactured in Canada, will be conducted at the University of British Columbia to evaluate disinfection performance and validate the predictions of a computational tool. The study will analyze UV disinfection dose-response curves, inactivation rates, and water quality parameters to assess the efficacy of UV LED technology for inactivating E. coli and other contaminants. Partner organization, CAWST, with expertise in water, sanitation and hygiene options using simple, affordable technologies, will work with the intern on the experimental design, engineering considerations, testing protocols, results interpretation and data analysis. Together, we aim to assess whether this novel disinfection solution is appropriate for sanitation in a disaster relief and humanitarian contexts.

Development of algorithms to improve microflow analysis

Our primary mandate at API is to advance basic research and innovation to commercialization by providing access to world-class industry expertise, services, and infrastructure. Our activities focuses on engaging and supporting drug discovery and development initiatives, ensuring compliance with regulatory standards and driving innovation and commercialization through collaborative research and clinical studies. The investigation of extracellular vesicles (EV) as the next frontier for diagnostic evaluation is underway across the globe. There is a myriad of techniques available to interrogate extracellular vesicles, but the most versatile by far is EV flow cytometry (evFC). This technique can assess millions of potential cell fragments from microlitres of biological material such as plasma, urine or cerebrospinal fluid. From these investigations, we can begin to unravel potential new biomarkers when intact tumor cells may simply be too rare to monitor. Similarly, we can monitor patient response to medical treatments without requiring excessive amounts of material. In addition, we can detect and monitor viral and bacterial infections due to the sub-micron resolution of the technique.
However, with any technique that enhances resolution, detection of the appropriate signals becomes critical. Identifying the true positive signal from the “negative” background has traditionally been a subjective protocol with cell-based flow cytometry. This process, called “gating” involves selecting a subset of events from all events collected during a flow cytometry experiment for further analysis or data presentation. This includes general clean-up of the data, such as removing dead and dying cells or events consisting of multiple cells, as well as isolating your target cell population using their characteristic size, granularity, and expression of various cell markers. Proper gating, along with smart panel-building, can make your data easier to interpret and more publication-ready. The development of machine-learning based approaches for dynamic gating and instrument calibration (refractive index and size) will not only enhance the analysis of small particle evFC data, but also significantly increase the throughput capability of analyses of large datasets from months to weeks or even days.

Intelligent Survey Technologies for Education: A Research, Design and Development to Enhance Usability, Data Collection, and Adaptive Logic

Xello is a leading college and career readiness platform designed to engage K-12 students in career exploration, academic planning, and skill development. It provides educators with data-driven insights to support student success, helping school districts make informed decisions.
To maintain its market leadership and enhance user experience, Xello seeks to improve its survey platform. The current system has limitations in usability, accessibility, and data management, impacting the effectiveness of surveys in collecting meaningful insights. Key challenges include the lack of safe survey deletion, limited printable survey functionality, restricted survey distribution to alumni, and the absence of file attachment support for responses. Additionally, complex survey branching logic needs improvement for better customization and engagement.
The Xello team aims to collaborate with WIMTACH’s applied research team to conduct an in-depth analysis of these challenges, explore innovative solutions, and develop an enhanced survey management system. This partnership will leverage research-driven methodologies and technological expertise to create a more efficient, user-friendly, and data-rich platform.
This project will drive innovation by optimizing survey management, streamlining data collection, and integrating advanced features such as multimedia attachments and dynamic survey logic. By modernizing these tools, Xello will enhance the efficiency of educators, improve student engagement, and expand survey capabilities for long-term data-driven decision-making. The anticipated benefits include increased adoption of Xello’s platform, improved educator workflows, and stronger data insights that support student success, ultimately reinforcing Xello’s position as a premier educational technology provider.

XEOS : Détection par IA de défauts sur des photographies aériennes

XEOS : Détection par IA de défauts sur des photographies aériennes

Principales activités du partenaire :
XEOS Imagerie est une entreprise spécialisée en photographie aérienne, relevés lidar et cartographie par intelligence artificielle. Elle a développé une grande expertise en cartographie par intelligence artificielle notamment à partir de nuages de points lidar en 3 dimensions et de photographie aérienne.

Problématique et avantages escomptés du projet :
La validation des reconstructions 3D des bâtiments est facilitée par l’examen de photographies aériennes dans les mêmes zones. Celles-ci permettent de mettre en contexte les surfaces des bâtiments et autres structures détectées. Elles permettent aussi d’identifier des problèmes ayant potentiellement des conséquences sur les reconstructions 3D et leur interprétation.
La présence de défauts d’acquisition dans la photographie aérienne réduit la qualité d’interprétation ou rend la photo inutilisable. La détection manuelle de ces défauts dans des milliers d’images prend du temps et est laborieuse. Il serait utile d’automatiser et d’optimiser ce processus. Plusieurs approches s’offrent à nous.
La définition du problème de la détection d’objets se divise essentiellement en deux parties distinctes : où se trouvent les objets dans une image donnée (localisation des objets) et à quelle catégorie appartient chaque objet (classification des objets). Par conséquent, les pipelines des modèles traditionnels de détection d’objets peuvent être divisés en trois étapes : sélection de la région informative, extraction des caractéristiques et classification. Les réseaux de neurones permettent d’effectuer ces trois étapes en même temps.
XEOS Imagerie possède déjà à l’interne une grande expertise en programmation en intelligence artificielle mais désire s’entourer de plusieurs stagiaires afin d’augmenter sa capacité de développement de produits.

Deformable Image Registration for Multimodal Radiotherapy Treatments in Gynaecological Cancers: External Beam and Brachytherapy

The Radiation Medicine Program at the Princess Margaret Cancer Centre delivers curative radiation therapy (RT) to cancer patients, including those with gynecological cancers, using a combination of external beam radiation therapy (EBRT) and brachytherapy (BT). A key clinical challenge is the accurate accumulation of radiation dose delivered across these modalities. Current methods rely on deformable image registration (DIR), which is hindered by large uncertainties due to anatomical changes caused by the BT applicator, variable bladder and rectum filling, and tumor shrinkage during treatment (Fu et al., 2023). These limitations reduce the accuracy of longitudinal dose accumulation and compromise treatment effectiveness. This project addresses that challenge by developing deep generative models to remove BT applicators from MR images, enabling accurate DIR and dose mapping between EBRT and BT sessions. The partner organization will benefit clinically by improving treatment precision and enabling better-informed re-irradiation strategies. Socially, the project supports safer, more effective cancer therapy, while economically, it may reduce planning errors and treatment complications, ultimately improving healthcare resource efficiency. By advancing AI-driven radiotherapy tools, this project also enhances the partner’s leadership in integrating machine learning into clinical cancer care.

Thales : Briefing de Situation Intelligent

Thales : Briefing de Situation Intelligent
Principales activités du partenaire :
Thales Canada conçoit et met en oeuvre des solutions reposant sur des hautes technologies. L’entreprise offre des capacités de pointe dans les secteurs de l’aviation civile, de la défense, de l’identité et de la sécurité numériques.
Problématique et avantages escomptés du projet :
Ce projet vise la création d’une capacité de génération de rapports structurés à partir de notes non structurées, en utilisant des LLM (Large Language Model). Plus précisément, la solution développée prendra en entrée (x) des traces écrites et désorganisées, transcrites de source audio, notes papier et/ou digitales, afin de générer un rapport structuré (y), spécifique au domaine. Typiquement, de tels rapports représenteront de façon concise et organisée le contexte, les événements clés et les conclusions (recommandations, actions, etc.) associés.
Cette capacité permet d’abord de faciliter la création de rapports requis dans plusieurs domaines et diminue la quantité d’information perdue ou oubliée. Cette capacité représente un composant dans une suite de capacités GenAI d’aide à la décision dans les domaines critiques.

Monitoring hair surface chemical modifications using atomic force microscopy

Since the beginning of recorded time, humans have been developing ways to make themselves more beautiful or
otherwise change their appearance. The hair-care industry itself has a huge global economic power: its estimated total
value is $47B annually. However, beauty does not come without a price: methods currently being used for hair
colouring and styling damage hair greatly. More importantly, they involve treatments that have negative effects on
human and environmental health.
To reduce the toxicity of hair-care treatments, SLI Beauty is developing new hair-surface chemical modification
techniques. I have partnered with them to bring my expertise at biophysical characterization to assess the success of
these surface modifications, and to develop new treatment modalities. With dedicated time spent onsite in the SLI
Beauty labs, I have the opportunity to bring my critical skills to help develop marketable products for this expanding
company.

Detecting Vulnerabilities in Generative AI

This project, a collaboration between 3Tenets Consulting Inc. and Dr. Wenjing Zhang from the University of Guelph, seeks to address emerging security and privacy vulnerabilities associated with the use of Large Language Models (LLMs) in enterprise environments. The initiative will focus on developing a prototype Privacy Leakage Assessment (PLA) Toolkit to evaluate and mitigate risks such as data extraction, membership inference, and prompt leakage attacks. Through systematic assessment, exploratory defense testing, and technical documentation, the project will provide 3Tenets with a preliminary framework to enhance its AI security offerings. This work supports the partner organization’s strategic goal of delivering advanced, privacy-aware cybersecurity solutions for clients adopting LLM-driven business applications.

Human-Centered Design and Regulatory Strategy for a Digital Heart Failure Self-Management Tool

This project involves partnering with a medical device company focused on supporting older adults with heart failure through an innovative self-management software. Key activities include developing comprehensive documentation to support the company’s quality management system, contributing to a usability study to evaluate the software’s real-world impact, and refining the design to ensure it meets both user needs and regulatory standards. By strengthening the software’s development and testing processes, the project aims to enhance usability, safety, and compliance. This will help the company meet important regulatory milestones while contributing to a more reliable and accessible solution for older adults managing heart failure at home.

Algorithme de vision artificielle pour la navigation chirurgicale en laparoscopie

Scopia aide les chirurgiens à réaliser des chirurgies minimalement invasives guidées par caméra (endoscope, laparoscope, etc.). Notre solution ajoute des couches d’intelligence aux images pour améliorer la visualisation, la navigation, le diagnostic et l’intervention. Le stage s’inscrit dans un projet plus large de développement de navigation chirurgicale en laparoscopie. Le but du projet est de superposer, en temps réel pendant la chirurgie, l’anatomie non exposée sur des images de laparoscopie.

Le camp de l’AJBQ : modéliser et évaluer les effets d’un camp estival visant la confiance comme communicateur·rice

Ce projet de recherche, mené en partenariat avec l’Association des Jeunes Bègues du Québec (AJBQ), vise à évaluer l’impact de leur camp estival visant à augmenter la confiance comme communicateur·rice chez les jeunes qui bégaient. L’objectif principal est de comprendre comment cette nouvelle approche influence le bien-être des participants et de créer un cadre d’évaluation que l’AJBQ pourra utiliser de manière autonome à l’avenir. Une évaluation réaliste sera mise en place pour mieux comprendre les mécanismes d’actions du camp à l’aide à des entrevues, de l’observation participative et des questionnaires. Ces informations permettront non seulement d’ajuster les activités du camp pour répondre aux besoins des jeunes, mais aussi d’assurer que cette initiative continue d’apporter des bénéfices durables. Pour l’AJBQ, ce projet offre une opportunité précieuse d’améliorer et de pérenniser son approche, tout en renforçant son rôle de leader dans le soutien des personnes qui bégaient.

Homo/Hetero Hybrid Dyadic Systems for Artificial Photosynthesis

Climate change is one of the main concerns of our society and is closely linked to the large consumption of fossil fuels and their associated carbon emissions. An appealing alternative is the production of hydrogen from water, powered by sunlight. Our project aims to develop a first family of efficient hybrid dyadic catalysts for this purpose. These catalysts will combine a molecular light-absorbing unit, known as photosensitiser (PS), anchored to the surface of a metallic nanoparticle (NP), which will act as the catalyst. A first-of-its-kind hybrid dyadic system was recently developed in a collaborative work between the University of Montreal and the Autonomous University of Barcelona. In this project, we will address the issues found in the first-generation catalysts. For this, we will design tailored PSs with improved properties, which we expect to increase the hydrogen production. In addition, we will work towards a second-generation of hybrid dyadic systems able to perform overall water splitting upon sunlight irradiation. In this way, both institutions will work together in expanding our knowledge of this unexplored field of research, which will enable the rational design of new efficient hybrid materials.