Innovative Projects Realized

Pragmatic Automated Code Transformation Leveraging Large Language Models

With recent advancements in Large Language Models (LLMs) designed for coding, Artificial Intelligence for Software Engineering (AI4SE) remains pivotal in enhancing both developer productivity and software quality. However, to excel at more specialized SE tasks like code transformation, companies still need to optimize LLMs to their specific context to avoid LLMs’ tendency to “hallucinate”, i.e., make up certain predictions.

The goal of this partnership is to develop advanced LLM-based methods that enable the translation of code across various software frameworks, focusing on detailed method-level and extensive project-aware file-level transformations. Such code transformation enhances the adaptability of complex software systems to changes in both the functional requirements and technological landscape (e.g., moving away from a legacy framework), ensuring that these systems remain efficient, scalable, and up-to-date with the latest industry standards and practices.

Monark Leadership Training Engagement

Many organizations feel unsatisfied with the results of leadership training programs. To tackle this problem, this research project explores new ways to make leader training more engaging and effective. Drawing on theories about human motivation and self-views, the project tests three different intentions to improve leader training. The first study investigates whether focusing on personal enjoyment and fulfillment can make leaders more engaged in their training, and ultimately more effective. The second study examines how emphasizing a person’s identity as a leader affects their training engagement and effectiveness. Lastly, the third study investigates whether caring about others and their well-being can boost training engagement compared to focusing on personal gain. By testing these approaches, this research hopes to find better ways to enhance leader training and make it more effective. The results will be shared through a scientific article and a white paper, reaching both academic researchers and Monark’s clients.

Former des personnes du troisième âge à la gouvernance et au service pastoral dans des Églises en mutations

Au Québec, dans les milieux pastoraux de l’Église catholique, les responsabilités pastorales sont assumées par des clercs et par des personnes bénévoles le plus souvent du 3e âge. On constate que ces dernières ont peu ou n’ont pas bénéficié par le passé d’une formation initiale et continue, leur permettant d’assumer pleinement leurs tâches dans un contexte social et religieux où la foi ne va plus de soi. Les enjeux d’une formation théologique à leur intention sont multiples: originalité de l’apprentissage au 3e et 4e âges, transformations culturelles et religieuses contemporaines, mutations des milieux ecclésiaux, etc. En explorant les perspectives de l’apprentissage pastoral et théologique chez les personnes du 3e âge, cette étude souhaite élaborer de nouveaux cadres pour la formation initiale et continue adaptés aux contextes pluriels actuels.

Les avantages attendus de cette étude pour l’organisme partenaire sont triples. Premièrement, l’identification de nouveaux cadres de formation permettant le renouvellement des pratiques de formation. Deuxièmement, l’acquisition de nouvelles connaissances scientifiques et probantes selon les normes universitaires. Troisièmement, le rayonnement de l’OSBL auprès de ses partenaires.

Investigation of the Fluid Flow and Heat Transfer in Fluidized Bed Combustor and Fluidized Bed Cooler During the Thermal Sand Reclamation Process to Improve its Performance

Foundries play a crucial role in recycling metals However, they face challenges, especially with sand, which is essential for making molds. During casting, sand gets contaminated from binder materials, making it impossible to reuse and causing environmental issues when disposed of. Sand reclamation, like thermal sand reclamation (TSR), helps by cleaning the sand through burning these contaminants. Using computer simulations, this project aims to study the TSR system designed by Gudgeon Thermfire International Inc. We aim to optimize heat and fluid flow with computational fluid dynamics (CFD). Initially, we’ll create a model to study fluid dynamics and then add heat transfer. We’ll analyze factors like gas velocity and fluid height to make the process more efficient. By simulating gas-solid fluidized bed dynamics with ANSYS Fluent software, we aim to enhance foundry sustainability by improving sand reclamation efficiency. This project uses advanced modeling, simulation, and optimization to achieve this goal.

Massive and Novel Electrostrictive Row-Column Ultrasound Arrays and Electronics

Imagine a cell phone camera with only 32×32 (1024 or 1K) pixels. The image quality would be very poor. This is the current regime where 3D ultrasound imaging is at. Going to Mega-Pixel sensors would create a quantum jump
in 2D and 3D image quality. However, doing so is challenging since in ultrasound MHz range signals must be recorded from each element or pixel and many wires or signal channels are needed which is currently technically
infeasible. A 1000×1000 array would require 1 million channels which would cost nearly $1B and is not practical. To address this challenge, CliniSonix has a new aperture encodable row-column array and novel readout schemes
which require only row- and column addressing. Moreover, we have learned how to use such arrays to make images that outperform current technologies. Our approach will lead to arrays approaching Mega-Pixel sensors,
which will lead to unprecedented image clarity, resolution, and field of view. Moreover, our approach enables significant improvements in tracking subtle motions and should lead to ~50x improved bloodflow sensitivity
compared to current ultrasound technology. To achieve these objectives we will require new arrays and new electronics. MITACS interns will develop these new arrays and new electronics.

Exploring Progressive Solutions: Innovating within the Regional Cannabis Sector

Prior to cannabis legalization the regional cannabis industry in the Central Kootenay Boundary Region was flourishing with an estimated 2,500 small grey-market producers who were an integral part of the socioeconomic fabric. Legalization offered the potential to transition the regional cannabis economy from an illicit to a legal sector, bringing with it potential for economic development. However, due to uncertainty and volatility the regional cannabis economy has dramatically changed bringing concern over losing unique Kootenay genetics, expertise, knowledge, and resources if a viable path forward cannot be found. The purpose of this project is to co-create, pilot, and evaluate innovative opportunities for the rural cannabis sector in the Central Kootenay Boundary using the smart specialization framework. The Mitacs student interns will be critical to the delivery and implementation of the pilot projects; they will support ongoing outreach with project stakeholders, administer various surveys for feedback, support the organization of sub-committees for each pilot, help create validated methods to assess the pilots, and summarize and export findings to other jurisdictions and sectors. These activities will support the regional cannabis sector, provide work-integrated learning experiences, and supply data for evidence based-decision making, which is essential for the partner organization, KAST, to help move the sector forward.

Fast approximate solutions to large and sparse systems of equations via convex optimization

Many effective Convex Optimization techniques face a common bottleneck: the resolution of large, sparse systems of equations. Despite the important advances in the state-of-art at theoretical and algorithm development, a significant challenge persists in the tendency to overlook the commercial viability of these algorithms.
Recent strides in computer hardware, driven by advancements in AI and cryptocurrency, have made small clusters of GPUs or FPGAs viable for addressing high-value optimization problems encountered by Kinaxis customers. If these hardware configurations can significantly reduce computational time, they offer a promising solution.
The collaboration between uOttawa, Kinaxis, and MITACS aims to bridge the gap between academic research and industry in high-performance computing. This partnership focuses on developing efficient, sustainable, and scalable optimization methods and computational algorithms tailored to real-world challenges convex optimization.

Bone targeted EP4 Agonist as muscle anabolics In muscular dystrophy

The overarching project goal is to validate the compound’s use as a management strategy for muscular dystrophy, opening the path to its clinical testing in a rare disease and expanding its commercialization potential beyond bone-related conditions.

Amelioration of marginal soil for cultivation of Camelina sativa

There is growing interest in biofuel as a low-carbon alternative to fossil fuels, which could help reduce greenhouse gas emissions, hence climate change impacts from transport. Marginal land presents an opportunity for the production of bioenergy feedstocks while minimizing competition with food crops for productive agricultural land. Examples of marginal soils are those that have any combination of low organic matter content, poor drainage, and high salinity, sodicity, pH and clay content, which make them less suitable for conventional food crops. Some biofuel crops, such as camelina (Camelina sativa), have been shown to grow well in marginal soils with minimal fertilizer inputs. Therefore, amelioration of marginal soils has potential to enhance the yields of such crops. The overall objective of this greenhouse experiment is to investigate the potential of soil amendments to restore the productivity of a marginal soil for cultivation of camelina as a biofuel crop. Treatments will include biochar, biostimulants, and gypsum. Camelina will be grown to physiological maturity in the potted soil and assessed for total biomass and seed yields, oil content, fatty acids, and protein content. Treatment effects on nitrous oxide emission and ammonia volatilization (from urea application) will also be tested. Results will assist the partner organization formulate strategies for improved biofuel crop production on marginal land.

Assessing threats to salmon and Tla-o-qui-aht priorities for restoration in the ha?ukmin (Kennedy) Watershed

The ha?ukmin (Kennedy) Watershed is vital to Tla-o-qui-aht First Nation as a source of food and culture. The goal of this research, led by a Tla-o-qui-aht member and under Tla-o-qui-aht guidance, is to support the Nation’s fisheries and restoration decision-making for ha?ukmin. Specifically, the research will investigate two factors: the diet of Peamouth Chub (who may compete with salmon for food) in ha?ukmin and the Tla-o-qui-aht community’s priorities for their salmon hatchery. Diets will be assessed by sampling stomach contents from ha?ukmin chubs throughout a year, while Tla-o-qui-aht values for the hatchery will be investigated with a series of interviews. This project will be a partnership between the UBC Centre for Indigenous Fisheries and Ha’oom Fisheries Society. Ha’oom, an organization that supports the rights-based fisheries for five Nations including Tla-o-qui-aht, benefits from this work through its potential to increase salmon returns and support Tla-o-qui-aht’s relationship to the salmon and waters.

Advanced Matrix Computation Methods for EMT Simulations

Computer simulation of transient events in an electric power system requires advanced modeling and computational tools. Introduction of renewable resources, such as wind and solar power, has caused fundamental changes to our power systems, thereby rendering many of our existing simulation tools inadequate. This research aims to solve some of the underlying shortcomings of power systems computer simulation tools through advanced computational methods and modern computing hardware so that simulations of large, complex systems is possible with efficiency and speed.

MALDI mass spectrometry technologies for investigating tumor metabolism

Cancer treatments that stimulate the immune system, called immunotherapies, have shown promise in fighting cancer. However, the success of immunotherapy depends on many factors, including tumor’s metabolism and the surrounding environment—the tumor microenvironment (TME). The TME is a complex mix of different cell types, including those that support the tumor’s growth and others that fight it. Cancer cells change the way they use energy and build new molecules, which creates a unique metabolic signature. This metabolic signature can affect how well immune cells called tumor-infiltrating lymphocytes (TILs) can attack the tumor. Tumors that have TILs, called ‘hot’ tumors, usually have better outcomes for patients than tumors without TILs, called ‘cold’ tumors. By understanding the metabolic differences between ‘hot’ and ‘cold’ TMEs, we can develop more effective immunotherapies. To study the metabolic profiles of ‘hot’ and ‘cold’ TMEs, we need highly sensitive techniques like mass spectrometry (MS). One powerful MS technique is matrix-assisted laser desorption/ionization (MALDI) imaging (MALDI-MSI), which can create detailed maps of the distribution of metabolites within the TME. The goal of this proposed research project is to improve the sensitivity of MALDI-MSI so that it can be used to detect biologically important metabolites in ‘hot’ or ‘cold’ TME.