CEPS Annual Research Report 2022-23

Research by the Numbers

Our stories

We welcome you to explore just some of the innovative research our college is generating. Faculty and students from Chemistry, Computer Science, Engineering, Mathematics & Statistics, and Physics are contributing to our knowledge of the world around us and advancing science to address critical issues. From reducing our reliance on plastics to protecting groundwater and biodiversity, and educating the next generation of scientists, CEPS is at the forefront of scientific research.

A new life for waste plastics and coffee grounds

Today, about 400 million tonnes of plastic are produced annually, with roughly 80 per cent entering landfills or the natural environment. The primary contributor: single use plastics. Researchers have been investigating ways to turn recycled and waste plastics into new more sustainable products, such as biocomposites – a material made from combining plastics and a natural resource (like natural fibers).

In Canada, researchers are attempting to find new uses for an estimated 300,000 metric tonnes of used coffee grounds generated each year. Dr. Manjusri Misra, a professor in the School of Engineering, set out to develop sustainable biocomposites from coffee grounds. The coffee grounds were made into biocarbon and then combined with recycled plastics (water bottles and bale wrap) to produce a biocomposite. The evaluation of the biocomposites strength, durability and flexibility showed that the incorporation of biocarbon produced a biocomposite that demonstrated balanced properties for rigid packaging and 3D printing applications.

Finding solutions to skin cancer

Each year, over 9,000 Canadians will be diagnosed with and 1,200 will die from melanoma, the deadliest form of skin cancer. The main cause of melanoma is Ultraviolet radiation (UVR). UVR damages DNA and creates reactive oxygen species (ROS), which damage skin cells and cause cancer. Researchers are developing cancer prevention strategies, like augmenting sunscreen, to reduce the formation of ROS.

Chemistry professor Dr. Michael Denk, and international collaborator Dr. Robert Lennon from Penn State University, explored the use of folate derivatives to reduce ROS before they can cause skin cancer. Folate derivatives naturally occur in foods, such as dark green leafy vegetables, and are essential for cell division and DNA creation. They exposed melanoma cells to different concentrations of folate derivatives and then subjected them to ROS through UVR and hydrogen peroxide. They found that folate derivatives inhibited the formation of ROS, which is expected to decrease melanoma formation.

MG360 travels to Sweden

The Morwick G360 Groundwater Research Institute has long standing relationships with Sweden through Lund University and the Swedish Geological Survey (SGU), through collaborative research efforts at multiple field sites, where MG360 have deployed various components of the DFN-M field approach for site characterization. In 2020, Dr. Beth Parker was awarded the Tage Erlander Visiting Professorship which was accompanied by an eight month sabbatical at Lund University in Lund, Sweden, with the first half being in 2022 and the remaining half during the spring of 2023. During the second half of her sabbatical, Dr. Parker taught a 2.5-week intensive PhD-course in Hydrogeological Conceptual Models for Groundwater Use and Protection: Characterization methods and scaling considerations, in the Department of Geology at Lund University, from March 15-31st, 2023. This course was designed to lean into hands-on education and incorporate lessons through lectures, field-work demos and real-world experience, case studies and student-led presentations.

Donation supports AI and biodiversity

Working in Dr. Graham Taylor’s lab, graduate students were afforded the chance to work on several projects thanks to a gift of $50,000 USD from NVIDIA, a global software company based in California and specializing in artificial intelligence (AI) research. One such project, called BIOSCAN-1M Insect Dataset released images with DNA barcodes to get ML researchers interested in biodiversity. The Team also trained and released baseline ML models for order- and family-level classification.

The BIOSCAN-1M Insect Dataset is a novel resource with potential to transform our understanding of global biodiversity. Comprising a million images with taxonomic labels, DNA barcodes and DNA-based species proxies, it allows for image-based taxonomic assessment, offering a reliable proxy for identifying known and unknown species and analyzing their ecosystem interactions. This dataset will be instrumental in developing artificial intelligence (AI) and machine learning models that can handle large volumes of complex biodiversity data. Its unique characteristic, a long-tailed class-imbalance distribution, poses a compelling challenge to the machine learning community, encouraging the development of innovative solutions.

Supporting people with colour blindness

ColourIconizer representing red, orange, yellow, blue, purple and pink. Described under the heading ColourIconizer full text.

There are an estimated 300 million people worldwide who have colour vision deficiency (CVD) (commonly known as colour blindness). Our society colour codes the physical world, which can lead to accessibility issues for those with CVD. To help, researchers have been exploring the use of colour patterns, though these are often unintuitive.

School of Computer Science professor Dr. David Flatla has developed two new colour patterns to assist people with CVD more easily distinguish between colours. Their first pattern, ColourIconizer, uses analogies to represent colours – a flamingo represents pink. Their second pattern, ColourMix, assigns shapes, or a combination of shapes, to the colours – X represents red.

They compared their two new colour patterns to the previously published colour pattern by enlisting volunteers with CVD from Reddit and Facebook. They found that ColourMix provided the best overall user performance and ColourIconizer helped the most when users were asked to select and sort colours.

ColourIconizer image full text

ColourIconizer is the first pattern developed for the research and uses detailed patterns that provide a direct analogy for the colour they represent. For example, as seen in the image in a series of colour blocks with icons (from top left), red is represented by a stop sign. Orange is represented by the corresponding fruit (an organge). The colour yellow is represented by a star, blue by a water droplet, purple by a bunch of grapes and pink by a flamingo.

Meet PhD candidate Kristina Kupferschmidt

Kristina Kupferschmidt is pursuing her PhD in the School of Engineering and the Vector Institute for Artificial Intelligence. Under Dr. Graham Taylor's guidance, she tackles the "translation gap" in AI for real-world healthcare challenges, exploring AI applications in mental health and medical imaging.

She plays a crucial role in the Ontario Vehicle Network (OVIN) project led by The Centre for Advancing Responsible and Ethical Artificial Intelligence (CARE-AI). The OVIN project is an accessible teaching course to help inform secondary students about the potential for AI to be applied within the rapidly changing automotive sector in Ontario. The course focuses on highlighting non-traditional uses of AI including in business, design, and supply chain management. 

Her impactful AI research and education reshape the future of AI for society's benefit. Outside of her PhD, she acts as a Scientist-in-Residence for NEXT-AI, where she advises nine start-up ventures on their AI strategies.

What can black holes tell us?

When it comes to testing the limits of what we know about the nature of the universe, nothing quite compares to black holes. These objects have extremely large masses packed into a very tiny space, exerting a gravitational pull so strong that space and time warp around them. The extreme conditions that black holes exist in make them useful for testing the limits of our current physics theories, such as Einstein’s theory of general relativity. 

Physics professor Dr. Huan Yang tested how well our current theories that describe strong gravitational objects, like black holes, hold up in extreme cases.  By applying a small change to the equations describing a stable black hole, Yang could see how the spectrum the theory predicts responds. The black hole spectrum was found to remain stable when small changes were applied to current theories, validating their ability to accurately analyze data in this extreme scenario.

Meet undergraduate student Heather Robertson

Heather Robertson, a Physics student, worked with Prof. Vlad Ladizhanksy during her Undergraduate Student Research Award term. Using the U of G’s solid-state nuclear magnetic resonance (SSNMR) spectroscopy to characterize alpha-synuclein fibrils, Roberston’s project looked at one implicator of Parkinson’s disease.  

Alpha-synuclein is a protein produced naturally in the body, however, when it becomes misfolded into a fibril form it can develop aggregates in the brain called Lewy bodies. Research has shown Lewy bodies can cause a number of synucleinopathies including Parkinson's Disease. 

Alpha-synuclein has a specified sequence of amino acids but when a protein folds, the relative location of the amino acids to each other will change. The SSNMR helps detect this. By comparing our experimental measurements of the spins for each amino acid with previously published sets of spin assignments, we can determine where the folding is occurring in our fibrils and how the structure is changing. 
 

Linamar partnership creates data-driven innovation

The Master of Data Science (MDS) program has established a collaboration with industry leader Linamar, resulting in impactful internships and propelling data-driven innovation forward.

MDS student Mohsen Kasiri worked on the first MITACS internship project with Linamar aimed at optimizing machine usage and productivity across Linamar's global factories. The project yielded valuable insights, including correlations between machine downtime and specific indicators. These findings not only sparked discussions on improved data governance within Linamar but also paved the way for subsequent projects, such as an evolving project led by Dr. Sheng Yang.

This ongoing partnership, and MDS’s partnerships with Cooperators, Gartner, Inc. and Geosyntec, exemplifies the MDS program's commitment to empowering students and delivering innovative solutions that shape every industry’s data-driven future.

MDS is establishing itself as a leader in data science education, bridging academia and industry to drive data science innovation and create leaders of tomorrow.