Projets novateurs réalisés

Machine Learning for Particle Detection

The overall goal of the project is too improve the precision of CMS, a detector used in particle physics experiment. The project will focus on two aspects: improving photon and electron detection and improving jet detection (a shower of particles). In the detector, multiple particles are detected at the same time. Photons and electrons are detected in the same section of the detector, making it hard to distinguish their signal. As a solution, a new machine learning algorithm has been created to replace the current algorithm. The intern is to evaluate the performance of this new algorithm and compare it with the older one. It is expected that the algorithm will be more successful than the previous one. If this is the case, we will then use our improved photon energy readings as a reference to compare signals from particle jets which should have the same energy. This will allow to determine a correction factor for the jets energy readings.

Epictrode – Customizable Printed Medical Tattoo Electrodes for Bio-signal Acquisition

The proposed research project is focused on customizable medical tattoo electrodes used on patients in different clinical settings. These electrodes are placed in specific places on the body and are connected to medical signal acquisition systems to collect different medical signals. The signals can include ECG (heart) signals and EMG (muscle) signals, and these signals can be used to diagnose or simply monitor the respective systems (cardiac and skeletomuscular) in the body. This project is part of the Lab2Market program. Therefore, the main goal is to gain insights into the market for the developed medical tattoo technology, which will help translate the technology from the academic lab to market.

Automated Video Content Chaptering via Machine Learning

The research will focus on building a deep learning model to analyze video and audio streams of input video material, essentially creating something like a table of contents, for example:
* Minutes 0-3: Introduction. Our topic is binomial coefficients.
* Minutes 3-4: Problem source.
* Minutes 4-6: Details of the mathematics. Calculations of values.
* Minutes 6-7: Conclusions. Applications of binomial coefficients. Further reading.

Video chaptering involves deducing cues, possibly from the transcript (word choices, such as “to begin”, or “in conclusion”), the audio (clapping, pauses), or the video itself (movement, scene cuts). Such video analysis is a very large challenge, because of the huge data volumes, and so the internship will be undertaken in concert with existing video-chaptering research being conducted in the Vision & Image Processing group at the University of Waterloo. The key goal will be the construction of small, efficient networks, to see which key features emerge as having significance in the chaptering process. The intern will gain expertise in both machine learning and statistical analysis, and be expected to contribute to research in both areas.

Post-Wildfire Stand-Type Shifts in the Western Boreal

We tend to think of wildfire as resetting the biological clock back to zero; forest types always come back to whatever was there before it burned. However, new evidence reveals that this is not always the case and fire can also cause changes to forest types. For example, very hot fires can kill seeds stored in the soil, while cool fires leave behind a significant amount of stored seed in both underground and in seed cones in surviving tree crowns. Fire can thus also function as a vegetation reset button in response to environmental conditions such as droughts or changes in wildfire frequency. Understanding the details of how, why, and where such changes have occurred in the past are necessary for us to be able to predict where the boreal forest might be heading in the future under climate change scenarios. For forest managers, this knowledge will also provide some guidance on how to manage post-harvesting forest types in service of creating healthier, more resilient, and potentially more productive landscape ecosystems.

Implementing a principles-focused evaluation at the New Canadians Health Centre (NCHC)

This project will support the New Canadians Health Centre (NCHC) in implementing an internal evaluation to assess its impact on refugee communities. As a health centre that is grounded in five principles, the goal of the project is to assess whether the principles are being upheld, and how they influence refugees experience with healthcare and enhance equity and social inclusion. The NCHC will be able to use the evaluation findings to inform its practices and services to better serve refugee communities. It will support the NCHC in its continued development and in sharing evidence on refugee health within a supportive learning environment.

Functional Brain-Computer Interface (f BCI)

As brain computer interface (BCI) is an emerging technology, this project attempts to understand the functionality of this technology from the user’s perspective. To achieve this goal, a BCI-based application is developed, and users’ ideas are collected from pre- and post-experience surveys. The previous study dealt with functionality of BCI in technical level and this project is more about the user experience.

Healing the Land, Healing Ourselves: Indigenous Land Stewardship and Educational Resurgence – Youth and Schools Program

This project will explore the strengths of one existing Indigenous-led school-based ecocultural restoration program, PAPAKEN HAUTW, and support the program in dreaming and designing of a larger and more sustainable youth land stewardship/land-based learning program within the tribal school where it is currently housed and beyond, to the community and Greater Victoria School Board-SD61.

Étude de cas : Le rôle de l’intermédiaire GoExport dans l’internationalisation des PME québécoises au Brésil

Le projet intitulé « Étude de cas : Le rôle de l’intermédiaire GoExport dans l’internationalisation des PME québécoises au Brésil » a pour but de déterminer comment des agents intermédiaires comme l’entreprise montréalaise GoExport permettent à des PME québécoises de surmonter les distances qu’elles peuvent rencontrer lors de leur processus d’internationalisation au Brésil. Ces distances sont d’ordre culturel, économique, administratif, technologique et géographique, et peuvent être extrêmement difficiles à surmonter pour des PME ayant peu de ressources à investir dans leurs exportations. Ce projet permettra donc à l’entreprise partenaire GoExport de justifier sa pertinence dans ce processus d’internationalisation auprès de futurs clients.

Governing Networks, Forced Migration and Precarious Housing in the City

The “local turn” of migration policies has been more pronounced since the “refugee crisis” of 2015. Its accompanying camps, emergency shelters, and buildings occupation have put cities under the microscope, underlying the multiple interactions between homelessness and forced migration, including precarious housing and hidden forms of homelessness. Several types of networks are engaged in tackling homelessness and precarious amongst migrant populations. While social, family or ethnic networks might alleviate the risks for more visible forms of homelessness, network members often lack resources to deal with housing and living costs in general (Hermans et al., 2020). In some cases, in particular unaccompanied minors, little or no support network is available, and many depend on the advocacy of neighbours’ collectives close to encampments (Coutant, 2018). Finally, cities themselves are engaged in local, national and transnational city networks aiming at sharing practices, leveraging resources and strategizing advocacy (Spencer, 2022). This project aims to look at exploring the construction and mobilization of these networks as well as analyze their impacts.

Impact of videos about UN SDGs on learning Ukrainian

The project “My UKRainian World” aims to create, pilot test and revise educational video resources for learners and speakers of Ukrainian, whose first or primary language is not Ukrainian. It will attempt to bring Ukrainian language video resources on par with other international languages.

The current phase of the project consists of: an analysis of the feedback data collected from beginner level Ukrainian learners, revision of already created videos during the GRI term, analysis of the structure of voice recordings of the videos in order to obtain datasets for simple Ukrainian language usage, researching and implementing at least one tool to use the acquired dataset in order to create Ukrainian language extension for digital products and preparation for knowledge mobilization.

The intern will be enrolled into all the above objectives of the project during his visit to University of Alberta.

Artificial intelligence for the prediction of variables in health

Today, thanks to microelectronics, it is possible to find technology that continuously collects data such as motion kinematics or physiological data. This type of technology is commonly referred to as wearable and therefore the scope of this project is to find methodologies based on artificial intelligence for the creation of models that can predict a variable of interest. These predictive models are of interest in sports and in general in human health that will allow to follow up the physical condition of a person, estimate the appearance of an injury or determine variables of interest without the need to invest in high technology, among other advantages. The objectives are: develop a software interface that allows the capture of kinematic, physiologic and performance variables from wearable sensors, conceive a software based on machine learning that allows the prediction of kinematic, physiologic and performance variables from data collected and evaluate the robustness and precision of different machine learning techniques.

Performance Analysis of Northern Infrastructure Affected by Climate Change

In this project, considering the rate-dependent behavior of permafrost, a multiphysics-based hazard ??assessment tool will be developed to investigate the life-cycle performance of infrastructure subject ?to ?different seasonal climate projections and climate change scenarios. For this purpose, first, an ?advanced thermo-elasto-viscoplastic (TEVP) constitutive model is being developed based ?on the ?framework of critical state soil mechanics to thoroughly capture the thermal creep ?behavior of ?permafrost. The TEVP constitutive model can thoroughly capture ?the thermal creep behavior of ?permafrost and the rate-dependent deformation of degrading ?permafrost. Afterward, the ?constitutive model will be implemented into a Finite Element (FE) ?software called DISROC to develop a ?multi-dimensional THM numerical platform that can be used as a robust hazard ?assessment tool in the ?life-cycle performance assessment and climate-resiliency analysis of ?northern infrastructure (physical ?asset) ?in permafrost regions.?