Innovative Projects Realized

Chercher et identifier les “policiers” dans la population québécoise, fin du 19e – début du 20e siècle

L’histoire de la police peut s’écrire en s’intéressant à l’identité et au profil sociologique des policiers, pour comprendre de quelle manière ils reflètent la société, comment ils y évoluent et s’y intègrent, mais aussi comment ils font « profession ». Cette approche humaine souligne les enjeux corporatistes, institutionnels de l’exercice de la police et montre que son action dépend également de l’agentivité de ses agents,. Il existe au Québec des données permettant de développer cette approche. L’Infrastructure intégrée des microdonnées historiques de la population du Québec (IMPQ, https://impq.cieq.ca/) rend en effet accessible un vaste ensemble de microdonnées saisies dans les actes de l’état civil québécois (1621-1923) et les recensements nominatifs canadiens (1852-1921). Elle regroupe des données sur 12 millions d’individus. Le stage proposé entend interroger ces données pour y trouver et mieux connaître les “agents de l’ordre” au Québec entre la fin du 19e siècle et 1921. Le stagiaire pourra ainsi interroger le profil des agents, leurs origines et logiques familiales, leurs conditions sociales, les enjeux qui différencient le policier urbain ou rural. Il articulera des réflexions sur le groupe professionnel et sur certains de ses membres, dont le profil est révélateur de réalités plus générales.

MLOPs – Toolchain Update

The management of the lifecycle of artificial intelligence (AI) models—commonly referred to as MLOps – has become a major concern for organizations aiming to deploy AI at scale. The MLOps process holds great promise for accelerating AI development and demonstrating its robustness and reliability before deploying it on critical systems such as drones. The expected outcomes of this project are numerous: increased productivity for AI experts, faster deployment of innovative solutions, reduced development and production time for AI algorithms by shortening delays between AI research teams and software development teams, and improved competitiveness for companies (such as Thales) through optimized MLOps process management. Furthermore, the success of this project will not only benefit Thales but also strengthen Canada’s position as a leader in innovation and technological development in key areas such as artificial intelligence. Additional academic collaborations are planned for the development of specific tools that will be essential and should be integrated into an MLOps chain associated with AI for critical systems.

L2M – AgrUS (Agricultural UltraSound)

When farmers are selling their cattle for meat, the quality of the meat can make a large difference in profits. Farmers and producers see between 30-45% of the retail price, so ensuring that they sell quality beef makes a difference for their bottom line. So how do they know when their cattle are ready for market? Most of the time, it comes down to time and weight, and producers only find out quality when cattle are slaughtered. That’s where our ultrasound-based project can be a difference maker. Using ultrasound, we can see into the muscles of live cattle. By adapting cutting edge medical ultrasound algorithms, we can inform producers of the meat quality before slaughter. For producers, their bottom line can improve in two ways: 1) saving on costs (up to $15/day/head of cattle), 2) upgrading their beef quality (e.g. from AA to AAA). With around 100 million livestock cattle across the US and Canada, our project can significantly improve profitability for beef producers.

Accessible Pressure-Driven Microfluidic Sorting of Deformable Particles for Biomedical Research

This project focuses on developing a pressure-driven microfluidic system capable of separating cells or cell-like particles based on their mechanical properties, without the use of chemical labels. By combining controlled fluid flow with digital image analysis, the system will quantify particle deformation and sorting behavior under well-defined conditions. The project aims to improve the robustness and real-time performance of deformability-based microfluidic sorting, supporting biomedical research on heterogeneous cell populations and applications such as cancer diagnostics.

Experimental Evaluation of Phase-Change-Material-Integrated Concrete for Passive Thermal Energy Storage in Buildings

This project explores how concrete can be enhanced to help buildings better manage temperature and reduce energy use. The material to be used in this project can absorb heat when temperatures rise and release it when temperatures drop, helping to keep indoor spaces more comfortable without relying as heavily on heating or cooling systems. The visiting PhD student from Japan will work at the University of Windsor to design, test, and evaluate these temperature-regulating concrete materials using laboratory experiments. This project will strengthen collaboration between the University of Windsor and Utsunomiya University, provide advanced research training for the student, and generate new knowledge that supports more energy-efficient and resilient building materials.

Model-informed precision dosing of mycophenolic acid in pediatric kidney transplant recipients

Mycophenolic acid (MPA) is a key anti-rejection drug for most children who receive a kidney transplant, yet many have large swings in MPA blood levels during the first 6 months after surgery—even with consistent dosing. This variability can contribute to serious side effects such as low white blood cell counts (higher infection risk) and may increase the risk of rejection or graft loss. Nearly one-third of pediatric recipients discontinue MPA within 2 years because it is difficult to tolerate, and current monitoring approaches are largely adapted from adult studies and may not reflect how children process the drug or what exposure levels trigger toxicity. We will study children at multiple time points within 6 months after transplant through a multi-centre Canadian cohort (Vancouver, Calgary, Edmonton, Toronto). We will identify patient factors (e.g., age, sex, kidney function, ethnicity, and time since transplant) that drive MPA exposure, define relationships between MPA levels and side effects, and build a predictive tool to support individualized dosing and proactive toxicity prevention. This work aims to reduce adverse events, improve graft survival and quality of life, and provide the institution with validated precision-dosing tools and multicentre data that strengthen clinical leadership, research capacity, and quality-improvement outcomes.

Quantification des bénéfices du mesurage des boues et des écumes des fosses septiques résidentielles

Les installations septiques résidentielles desservent environ un million de résidences non raccordées à un réseau d’égout au Québec. La bonne gestion de ces installations, notamment la vidange des fosses septiques, est garante de leur bon fonctionnement et d’une protection adéquate de l’environnement. Bien que plusieurs gouvernements municipaux (villes et MRC) assurent la gestion du service de vidange des fosses septiques, moins de 10% de ces municipalités effectue une campagne de mesures des boues et écumes et adaptent la fréquence des vidanges aux résultats de ces campagnes de mesures. Or, cette adaptation de la fréquence des vidanges à l’accumulation réelle de boues et écumes peut présenter des avantages non négligeables : économies pour la municipalité et pour les citoyens, prévention du relargage des contaminants, diminution des GES émis par les passages du camion de vidange, etc. La MRC de Rimouski-Neigette souhaite mettre en place un service de mesures des boues et écumes, et veut quantifier les bénéfices potentiels de cette pratique sur son territoire. En exploitant des données existantes provenant d’autres MRC et en effectuant un portrait de l’utilisation actuelle des fosses septiques sur le territoire, ce projet permettra de quantifier les bénéfices financiers et environnementaux de cette pratique.

Make (insert country) Great Again? The New Right and Countries without Narratives of Historical Greatness: The Case of Colombia.

The various manifestations of the new right observed across the globe frequently exhibit a nostalgia for a selective and romanticized version of their nations’ past. Moreover, the manner in which new-right actors frame the past offers crucial insights into the actions they advocate in the present and the future they aim to construct. However, there are countries –such as Colombia– where new-right actors do not typically invoke narratives of a “glorious” past. This raises several questions: How do new-right actors in countries like Colombia engage with representations of the past? What role does the past play for these actors in contexts that do not have shared narratives of historical greatness? Are their backward-looking orientations similarly romantic or do they articulate more complex and ambivalent views of history? To answer these questions, I will employ original data and draw on the expertise in memory studies of my host supervisor at the Universidad Tecnológica de Pereira and the expertise in democratic erosion of my home supervisor at the Université de Montréal. The publication of this research, the collaborative exploration of this significant topic, and our participation in international conferences will benefit both participating institutions and help enhance the reach of our work.

SyNOF, symbolic regression for reduced density matrix functionals

This project will develop a computational framework that uses machine learning to automatically discover mathematical formulas for describing the electron arrangements of molecules. The electronic structure problem is of great interest in quantum chemistry, where cost-effective methods that capture electron correlation are essential for industrial and technological applications. Current machine learning methods often act as black boxes, making it difficult to understand why they produce certain results. In contrast, this project discovers interpretable equations that researchers can read, understand, and verify. The framework also offers a natural connection to quantum computing: near-term quantum devices can provide electron distribution data that would be used to discover formulas capturing correlation effects beyond the reach of classical computers. The collaboration brings together McMaster University’s expertise in machine learning with CSIC’s knowledge of electronic structure theory, producing open-source software tools, benchmark datasets, and trained researchers with skills spanning computational chemistry, artificial intelligence, and quantum simulation.

Simulation Studies of a Future Xenon Rare Event Search Experiment

This project will contribute to the development of a leading experiment in physics seeking to observe dark matter particles and a hypothetical nuclear decay process. The experiment is called XLZD. Through the development and execution of sophisticated simulation codes, the performance of XLZD will be evaluated through the calculation of several figures of merit. The calculations will provide insight into how to enhance the design of XLZD which is still in its early planning stages. This project will benefit the host researchers who are piloting the design of the experiment, and the home researchers who will join and contribute to the collaboration through their own research and development projects.

SEE Gender: Gender Inclusive Social Ecological Entrepreneurship for Small-Scale Fisheries Vulnerability to Viability Transition – A Study Based in the Indian Sundarbans

The “SEE-Gender” research initiative is part of the V2V global partnership project, aiming to explore the potential of gender-inclusive, socially and ecologically oriented entrepreneurship in facilitating the transition of small-scale fisheries from vulnerability to viability in the Indian Sundarbans mangrove delta. The primary objective of this research is to characterize social-ecological entrepreneurship within small-scale fisheries and to examine the relationship between gendered interactions and viability attributes ( increased resilience, enhanced access to capital, and improved well-being). This objective will be achieved through an interpretive case study approach, utilizing field interviews, key informant interviews, and direct observations at the field sites in the Indian Sundarbans. This project is a collaborative effort involving the University of Waterloo and ICAR-CIFRI. For both institutions, this project not only strengthens their partnership but also enriches their knowledge and skill sets, trains students in global research, and paves the way for shared publications, policy ideas, and funding opportunities, positioning both institutions as leaders in equitable and sustainable coastal solutions.

Production Planning at NWC to Support Inventory Management at its Warehouses

This project addresses the complex production and inventory challenges at NWC’s assembly facility, which manages 150 SKUs with 26 requiring specific setups. With finite production capacity and storage constraints, the current system struggles to maintain optimal stock levels.

The primary objective is to develop a dynamic production planning model using Reinforcement Learning (RL). This AI-driven solution will intelligently balance production schedules against capacity limits, warehouse space, and pre-defined target inventory levels.

The key deliverable is an automated system that generates optimized production plans. Success will be measured by the model’s ability to consistently meet target stock levels, minimize excess inventory and waste, and ensure production aligns closely with demand. This project will empower NWC with a robust, data-driven tool to enhance operational efficiency and reduce costs across its supply chain.