3-Minute Thesis Competition

Overall Winners

Faith Akinyemi - Second Place, People's Choice
MSc in Applied Computer Science and Society
Supervisors: Dr. Michael Beck and Dr. Christopher Bidinosti

From Data to Dinner: Predicting harvests before they happen
My research uses field images to predict crop yield, leveraging machine learning techniques to extract patterns and features correlating yield. These features include plant health indicators, growth stages, or canopy coverage.

I am particularly interested in using these features to develop models that improve the accuracy of yield prediction, helping farmers make data-driven decisions. My approach considers temporal changes in the crop, capturing how its characteristics evolve.

My work contributes to precision agriculture, a field that seeks to optimize resource use, increase productivity, and promote sustainability in farming. My research has the potential to transform traditional agricultural practices by integrating advanced AI methods.

Noah Lewicki
MSc in Bioscience, Technology, and Public Policy
Supervisor: Dr. Sanoji Wijenayake

Ready to Rumble: Milk Fights Liver Disease
Obesity during pregnancy can have long-term health effects on offspring, increasing their risk of conditions like non-alcoholic fatty liver disease (NAFLD). NAFLD is the most common liver disease in children, and is characterized by excess fat buildup in the liver, leading to inflammation, liver damage and liver failure. Breast/chest feeding helps counteract the effects of obesity, but what about for NAFLD? Human milk contains biological nanovesicles called milk-derived extracellular vesicles (MEVs). MEVs positively influence metabolism and can be anti-inflammatory. My study explores how MEVs impact NAFLD risk in offspring with gestational obesity. I hypothesize that MEVs will provide protection against NAFLD and reduce chronic liver inflammation and fat buildup in offspring. Understanding MEVs’ role could shape policies promoting breastfeeding and the enhancement of infant formulas with MEVs, providing a new approach to improve long-term health outcomes for children.

Hardi Matholia
MSc in Applied Computer Science and Society
Supervisor: Dr. Mary Adedayo

Guardians of the Database Galaxy: A forensic analysis of MongoDB
This study looks at how to keep data safe in MongoDB, a type of database used by many businesses to store large amounts of information. As more companies use MongoDB, it becomes a target for hackers who may try to steal or delete important data. While there has been a lot of research into protecting traditional databases, there is less focus on databases like MongoDB. This study explores ways to detect and stop harmful activities in MongoDB, as well as how to recover deleted data. By analyzing the database's logs, we can track and prevent unauthorized actions. The goal is to create a tool that helps protect databases from attacks like data theft or loss, and ensures data is recoverable if something goes wrong. This tool will help businesses protect their data and recover it when necessary.

Ranjini Mukherjee
MSc in Bioscience, Technology, and Public Policy
Supervisor: Dr. Jean-Pierre Desforges

Polar Bears and Pollutants: Using Cells to Improve Risk Assessments
Polar bears (Ursus maritimus) are apex marine predators in the Arctic, exposed to high levels of persistent organic pollutants (POPs) through biomagnification. While previous studies have detected legacy and emerging contaminants in polar bears, their biological effects remain unclear due to ecological and biological confounders. This study improves chemical risk assessment using in vitro methods with primary polar bear cells to evaluate species-specific toxicity of priority Arctic contaminants. It employs New Approach Methodologies (NAMs) through in vitro dose-response experiments to assess individual POPs and Chemicals of Emerging Arctic Concern (CEACs) across key physiological systems, including immune, endocrine, reproductive, and hepatic function. Given Arctic Indigenous communities’ reliance on traditional diets, they are particularly vulnerable to these pollutants. This research will enhance understanding of POP and CEAC toxicity, informing safer chemical management strategies to protect Arctic wildlife and human health.

Wild Cards

Dhulmy Bandara
MSc in Environmental and Social Change
Supervisor: Dr. Srimathie Indraratne

Fixing Soil Heavy Metals - A Play by Plants and Modified Organic Materials
Heavy metal contamination in boreal forest soil particularly by Cadmium (Cd), Copper (Cu), Lead (Pb), and Zinc (Zn) is an environmental issue associated with mining. Heavy metal contaminated soil causes food chain contamination, detrimental effects on humans, contamination of natural waters and impairment of plant growth. Chemical immobilization combined with phytostabilization is a promising remediation strategy of heavy metal contaminated soil. In this technique, various kinds of amendments are added to soil which immobilize heavy metals whereas an established vegetation cover stabilizes heavy metals within the rhizosphere zone. This project will assess the effectiveness of modified biochar as amendments in immobilizing Cd, Cu, Pb, Zn in acidic boreal forest soils with different levels of concentrations. Additionally, it will evaluate the phytostabilization potential of native Canadian grass species to reduce mobility and bioavailability of these heavy metals contributing to development of effective remediation measures in multi-metal contaminated boreal forest ecosystems.

Tabitha Evans
MA in Environmental and Social Change
Supervisor: Dr. Ryan Bullock

Transdisciplinary Research Teams Co-Produce Essential Climate Knowledge and Solutions
Transdisciplinary research approaches to climate change mitigation are being used more often given their strengths in collaboration, knowledge integration and collective decision making. Such approaches warrant more attention to understand how diverse teams produce knowledge and practice problem-solving. My thesis research explores the strengths and challenges of transdisciplinary research to offer future avenues for team collaboration and policy decision–making processes.

Rubeena Gosal - First Place
MSc in Bioscience, Technology, and Public Policy
Supervisor: Dr. Tabitha Wood

Molecular Puzzles, Home Depot Buckets, and the Quest for a Fertility Agonist
Infertility affects millions worldwide, yet treatment options remain limited, particularly for those with polycystic ovary syndrome (PCOS)—the leading cause of anovulatory infertility. At the root of the issue is a hormone imbalance that disrupts follicle-stimulating hormone (FSH) signaling, preventing ovulation.

My research focuses on restoring this signal by identifying small molecules that can activate the FSH receptor (FSHR). Using computational modelling, I screen thousands of compounds through molecular docking—searching for the right fit like a missing puzzle piece. But proteins aren’t static, and molecular dynamics simulations allow us to test whether these molecules actually stay bound. Once promising candidates are found, synthetic chemistry takes over, using reactions like the Truce-Smiles rearrangement to reshape and refine molecules into viable therapeutics.

By bridging computer modeling with chemistry, my work aims to develop new treatments for infertility—offering solutions where “just keep trying” has never been good enough.

Malcolm Reimer
MSc Bioscience, Technology, and Public Policy
Supervisor: Dr. Craig Willis

Environmental emergence cues for hibernating bats
In the cool depths of a limestone cave, temperature, humidity, and darkness are constant — ideal conditions for hibernators to save energy over winter. Endangered little brown bats (Myotis lucifugus) may hibernate for up to eight months, emerging in spring with minimal stored fat. Exiting the cave on warm, calm days with higher insect activity could provide an opportunity to forage and recover from hibernation. But without weather cues from the outside world, how might hibernating bats anticipate good conditions for emergence? Atmospheric pressure changes, which precede warm and cold fronts, are sensed by many animals, and little brown bats appear to synchronize activity during hibernation with pressure patterns as spring approaches. Using infrared cameras and radio telemetry, my research monitors the activity of bats throughout their hibernation at a Manitoba cave to reveal how air pressure and weather influence their emergence timing and behaviour.

AM Heat

Malcolm Reimer - Wild Card

MSc Bioscience, Technology, and Public Policy

Supervisor: Dr. Craig Willis

Environmental emergence cues for hibernating bats
In the cool depths of a limestone cave, temperature, humidity, and darkness are constant — ideal conditions for hibernators to save energy over winter. Endangered little brown bats (Myotis lucifugus) may hibernate for up to eight months, emerging in spring with minimal stored fat. Exiting the cave on warm, calm days with higher insect activity could provide an opportunity to forage and recover from hibernation. But without weather cues from the outside world, how might hibernating bats anticipate good conditions for emergence? Atmospheric pressure changes, which precede warm and cold fronts, are sensed by many animals, and little brown bats appear to synchronize activity during hibernation with pressure patterns as spring approaches. Using infrared cameras and radio telemetry, my research monitors the activity of bats throughout their hibernation at a Manitoba cave to reveal how air pressure and weather influence their emergence timing and behaviour.

Leslie Kirkness

Master’s in Development Practice: Indigenous Development

Media, a social determinant of health for Indigenous people
The media focuses on the negative stereotypes of Indigenous people and reports incorrect information with little to no accountability for the trauma they inflict on the Indigenous population. There is a lack of correcting misleading information to the general population, which can lead to influencing healthcare workers who assume Indigenous people are seeking medication and not being truthful about the ailments they suffer. Media is a contributing factor that creates unnecessary barriers for Indigenous people accessing healthcare; media causes cumulative effects of harm, which is a detriment to the health of Indigenous people.

This research will examine how media perception can influence healthcare policy and Indigenous people's experiences when seeking healthcare. It will conduct a peer-reviewed literature review to find common themes within reporting and review Indigenous peoples' experiences through exploring case studies throughout Canada. Lastly, it will identify gaps and provide recommendations.

Sujay Rittikar

MSc in Applied Computer Science and Society

Supervisor: Dr. Sheela Ramanna

Multilingual Propaganda Extraction using LLMs
News is increasingly being communicated through social media, allowing readers to explore various topics at their own pace. While the information is easily and instantly accessible, it attracts manipulation and disinformation through propaganda techniques such as misleading persuasion and claims. These can have an adverse effect on society in terms of biased opinions about a person or an organization. Using the advancements in the Large Language Models (LLMs), our study focuses on establishing an end-to-end multilingual propaganda extraction system in 22 languages. This research aims to extract the claims and narratives a news article intends to spread and the roles of persons or organizations involved in its context.

Tabitha Evans - Wild Card

MA in Environmental and Social Change

Supervisor: Dr. Ryan Bullock

Transdisciplinary Research Teams Co-Produce Essential Climate Knowledge and Solutions
Transdisciplinary research approaches to climate change mitigation are being used more often given their strengths in collaboration, knowledge integration and collective decision making. Such approaches warrant more attention to understand how diverse teams produce knowledge and practice problem-solving. My thesis research explores the strengths and challenges of transdisciplinary research to offer future avenues for team collaboration and policy decision–making processes.

Ranjini Mukherjee - Overall Winner

MSc in Bioscience, Technology, and Public Policy

Supervisor: Dr. Jean-Pierre Desforges

Polar Bears and Pollutants: Using Cells to Improve Risk Assessments
Polar bears (Ursus maritimus) are apex marine predators in the Arctic, exposed to high levels of persistent organic pollutants (POPs) through biomagnification. While previous studies have detected legacy and emerging contaminants in polar bears, their biological effects remain unclear due to ecological and biological confounders. This study improves chemical risk assessment using in vitro methods with primary polar bear cells to evaluate species-specific toxicity of priority Arctic contaminants. It employs New Approach Methodologies (NAMs) through in vitro dose-response experiments to assess individual POPs and Chemicals of Emerging Arctic Concern (CEACs) across key physiological systems, including immune, endocrine, reproductive, and hepatic function. Given Arctic Indigenous communities’ reliance on traditional diets, they are particularly vulnerable to these pollutants. This research will enhance understanding of POP and CEAC toxicity, informing safer chemical management strategies to protect Arctic wildlife and human health.

Zachariah Code

Joint Master of Arts in History

Supervisor: Dr. Alexander Freund

Left Behind and Forgotten: The History of German-Canadian Families and Internment During the Second World War
This thesis addresses the lives of German-Canadian individuals who were affected by internment under the Defense of Canada Regulations during the Second World War. The primary focus of this research is to identify and analyze the experiences of German-Canadian internees from before their internment to eventual release, which has so far been neglected in the historiography. Additionally, through the analysis of individual internees’ experiences, a secondary aim of this research is to examine the lives of internees’ family members and explore what their experiences were outside the camp. What were the experiences for German-Canadian wives and children while husbands were taken away to remote internment camps? Using previously unanalyzed collections of personal and business correspondence, this research hopes to provide a more substantial narrative of German-Canadian internees and their families' experiences during the Second World War.

Britney Champagne

Master's in Development Practice: Indigenous Development

When Hunger Speaks, Learning is Silenced: Fighting for Food Security for Indigenous Student
Food insecurity is a major barrier to academic success for many First Nations, Inuit, and Métis students in Canada, both on and off reserve. Limited access to nutritious food affects more than health, it hinders concentration, cognitive development, and overall academic performance. Research shows food insecurity is deeply connected to cultural practices, socio-economic status, and educational opportunities (Shafiee et al., 2022). Students who experience hunger struggle with focus, participation, and social engagement, while also facing health risks like weight fluctuations and malnutrition (Jyoti et al., 2005). School-based nutrition programs have shown promise in improving both dietary needs and academic outcomes, underlining the need for culturally relevant interventions (Gillies et al., 2020). Addressing food insecurity requires holistic, community-driven solutions that empower Indigenous families and respect cultural values (Fieldhouse & Thompson, 2012). Strengthening culturally relevant food security is essential not only for better health but for fostering cultural pride, well-being, and academic success.

Noah Lewicki - Overall Winner

MSc in Bioscience, Technology, and Public Policy

Supervisor: Dr. Sanoji Wijenayake

Ready to Rumble: Milk Fights Liver Disease
Obesity during pregnancy can have long-term health effects on offspring, increasing their risk of conditions like non-alcoholic fatty liver disease (NAFLD). NAFLD is the most common liver disease in children, and is characterized by excess fat buildup in the liver, leading to inflammation, liver damage and liver failure. Breast/chest feeding helps counteract the effects of obesity, but what about for NAFLD? Human milk contains biological nanovesicles called milk-derived extracellular vesicles (MEVs). MEVs positively influence metabolism and can be anti-inflammatory. My study explores how MEVs impact NAFLD risk in offspring with gestational obesity. I hypothesize that MEVs will provide protection against NAFLD and reduce chronic liver inflammation and fat buildup in offspring. Understanding MEVs’ role could shape policies promoting breastfeeding and the enhancement of infant formulas with MEVs, providing a new approach to improve long-term health outcomes for children.

PM Heat

Manjot Singh Sran

MSc in Applied Computer Science and Society

Supervisor: Dr. Sheela Ramanna

Advancing Emotion Recognition with Artificial Intelligence
What makes humans so adept at reading emotions? It’s not just what we hear or see—it’s the combination of multiple senses. Machines relying on only one sense often struggle to interpret cues behind emotions. That’s where our research comes in.

We investigate how combining audio and visual cues can enable machines to better understand emotions. Using transformer models—Wav2Vec2 for audio and VideoMAE for video—we extract detailed emotional patterns from speech and facial expressions. We evaluated three fusion strategies: early, mid, and late fusion. Late fusion, which blends outputs from separate models, achieved remarkable accuracy on two popular audio-visual datasets—86.84% on CREMA-D and 95.39% on RAVDESS—far surpassing traditional methods.

This work in emotion-aware AI has tremendous potential to transform applications in healthcare, education, and human-computer interaction. By teaching machines to interpret emotions more like humans do, we are creating a future where technology truly connects with us.

Brent Wennekes

Master’s in Development Practice: Indigenous Development

Supervisor: Dr. Melanie O’Gorman

Capitalizing Indigenous Entrepreneurs
Access to capital remains a persistent challenge for Indigenous entrepreneurs in Canada, limiting opportunities for business growth and economic self-determination. Despite perceptions that ample funding exists, much of it is inaccessible due to systemic barriers embedded in colonial financial structures. This study, conducted in partnership with an Indigenous-led cooperative, employed a qualitative approach to explore these barriers and identify solutions. Through 135 interviews and surveys with Indigenous business owners and both Indigenous and non-Indigenous capital providers, the research highlighted three key challenges: the misalignment of traditional lending models with Indigenous economic systems, systemic bias within financial institutions, and the lack of Indigenous-led capital providers. However, Indigenous respondents also identified solutions, advocating for alternative financing models such as revenue-based financing, community-backed lending, and Indigenous-led investment funds. Addressing these barriers requires reshaping financial systems to align with Indigenous governance, economic realities, and self-determination.

Dhulmy Bandara - Wild Card

MSc in Environmental and Social Change

Supervisor: Dr. Srimathie Indraratne

Fixing Soil Heavy Metals - A Play by Plants and Modified Organic Materials
Heavy metal contamination in boreal forest soil particularly by Cadmium (Cd), Copper (Cu), Lead (Pb), and Zinc (Zn) is an environmental issue associated with mining. Heavy metal contaminated soil causes food chain contamination, detrimental effects on humans, contamination of natural waters and impairment of plant growth. Chemical immobilization combined with phytostabilization is a promising remediation strategy of heavy metal contaminated soil. In this technique, various kinds of amendments are added to soil which immobilize heavy metals whereas an established vegetation cover stabilizes heavy metals within the rhizosphere zone. This project will assess the effectiveness of modified biochar as amendments in immobilizing Cd, Cu, Pb, Zn in acidic boreal forest soils with different levels of concentrations. Additionally, it will evaluate the phytostabilization potential of native Canadian grass species to reduce mobility and bioavailability of these heavy metals contributing to development of effective remediation measures in multi-metal contaminated boreal forest ecosystems.

Tarik Aziz

MA in Applied Economics

How Does Corruption Affect Innovation in OECD Countries? A Panel Cointegration Analysis
This study investigates the impact of corruption on innovation across 20 OECD countries from 2002 to 2021. Control variables such as educational attainment, median age of the population, and GDP per capita are included in a multivariate framework. R&D expenditures are used as a proxy for innovation, while control of corruption index is employed to measure corruption. Additionally, regulatory quality and voice and accountability indicators are utilized. Second-generation techniques are applied to account for heterogeneous panel data characteristics. Augmented Mean Group (AMG) and Mean Group (MG) estimators are used for long-run relationship estimations, and the Panel-Corrected Standard Error (PCSE) estimator is applied for robustness checks. The results indicate that corruption negatively affects innovation in the selected countries over the long run. Regulatory quality enhances innovation, while limited voice and accountability support R&D expenditure. Scientific and technological research and innovation are essential components for attaining sustainable development objectives.

Hardi Matholia - Overall Winner

MSc in Applied Computer Science and Society

Supervisor: Dr. Mary Adedayo

Guardians of the Database Galaxy: A forensic analysis of MongoDB
This study looks at how to keep data safe in MongoDB, a type of database used by many businesses to store large amounts of information. As more companies use MongoDB, it becomes a target for hackers who may try to steal or delete important data. While there has been a lot of research into protecting traditional databases, there is less focus on databases like MongoDB. This study explores ways to detect and stop harmful activities in MongoDB, as well as how to recover deleted data. By analyzing the database's logs, we can track and prevent unauthorized actions. The goal is to create a tool that helps protect databases from attacks like data theft or loss, and ensures data is recoverable if something goes wrong. This tool will help businesses protect their data and recover it when necessary.

Rubeena Gosal - Wild Card

MSc in Bioscience, Technology, and Public Policy

Supervisor: Dr. Tabitha Wood

Molecular Puzzles, Home Depot Buckets, and the Quest for a Fertility Agonist
Infertility affects millions worldwide, yet treatment options remain limited, particularly for those with polycystic ovary syndrome (PCOS)—the leading cause of anovulatory infertility. At the root of the issue is a hormone imbalance that disrupts follicle-stimulating hormone (FSH) signaling, preventing ovulation.

My research focuses on restoring this signal by identifying small molecules that can activate the FSH receptor (FSHR). Using computational modelling, I screen thousands of compounds through molecular docking—searching for the right fit like a missing puzzle piece. But proteins aren’t static, and molecular dynamics simulations allow us to test whether these molecules actually stay bound. Once promising candidates are found, synthetic chemistry takes over, using reactions like the Truce-Smiles rearrangement to reshape and refine molecules into viable therapeutics.

By bridging computer modeling with chemistry, my work aims to develop new treatments for infertility—offering solutions where “just keep trying” has never been good enough.

Lisa (Dumas) Neufeld

Joint MA in Peace and Conflict Studies

Supervisor: Dr. Jessica Senehi

Carrying the Multitude: Personal Transformation as a Pathway to Reconciliation
Can personal transformation lead to reconciliation? First, we must explore what is being transformed, and what reconciliation means in this case. Next, we’ll wander the path in between, where we’ll explore transformation of story and identity as a movement toward harmonizing the self and, maybe, society. This work is rooted in Grounded Autoethnography, weaving together disparate parts and pieces through life-storying, poetry, and photography, dancing with Justice Murray Sinclair’s four eternal questions: Who am I? Where do I come from? Why am I here? Where am I going? Underpinning this work is the understanding that people may begin to reconcile between themselves, but if they remain alienated within themselves, the work is incomplete.

Faith Akinyemi - Overall Winner

MSc in Applied Computer Science and Society

Supervisors: Dr. Michael Beck and Dr. Christopher Bidinosti

From Data to Dinner: Predicting harvests before they happen
My research uses field images to predict crop yield, leveraging machine learning techniques to extract patterns and features correlating yield. These features include plant health indicators, growth stages, or canopy coverage.

I am particularly interested in using these features to develop models that improve the accuracy of yield prediction, helping farmers make data-driven decisions. My approach considers temporal changes in the crop, capturing how its characteristics evolve.

My work contributes to precision agriculture, a field that seeks to optimize resource use, increase productivity, and promote sustainability in farming. My research has the potential to transform traditional agricultural practices by integrating advanced AI methods.

All participants in the semi-finals receive a certificate of participation and a small gift.

The overall winners from each heat receive $200.

The wildcards from each heat receive $100.