Innovative Projects Realized

Sweat Club BSI Proposal

Sweat Club is a boutique-style gym located in downtown Fredericton that offers a variety of group fitness classes, including strength training and high-intensity interval training. The gym aims to provide a fun, inclusive, and motivating environment for clients of all fitness levels. However, like any business in the fitness industry, Sweat Club faces several challenges, such as competition from other fitness businesses, scheduling and logistics, marketing and promotion, keeping up with trends and changing preferences, managing costs, and safety and liability concerns. One significant opportunity for the business is to expand its clientele and target different age groups, particularly a younger demographic.

To combat these challenges, Sweat Club plans to hire a summer intern to improve and aid the business in all areas where they are currently facing challenges. The internship offers an opportunity for the intern to gain knowledge in the business field and the fitness industry. Summer interns can offer fresh perspectives and innovative ways to approach problems, which can benefit the business. Additionally, the intern can help increase productivity by taking on tasks that might otherwise be left undone due to time constraints.

Développement d’un système informationnel 360 pour l’optimisation des opérations

Videns Analytics est une compagnie qui approche cinq ans d’existence et qui regroupe une quarantaine de collaborateurs. Pour sa gestion interne, la compagnie utilise plusieurs outils qui touchent différents domaines comme par exemple:
– la gestion des heures des collaborateurs, utile notamment pour la facturation;
– la planification de l’affectation des ressources humaines sur les différents projets;
– les ressources humaines;
– la comptabilité;
– la gestion des clients et prospects (CRM).
L’objectif du projet est de développer un système d’informations analytique qui permettra de rassembler ces données et de les présenter sous un format qui facilitera la prise de décision et permettra d’optimiser la gestion de la compagnie. Le projet couvre:
– l’infrastructure cloud qui va héberger le système d’informations;
– le modèle de données;
– la base de données;
– la génération de code ETL en python qui va extraire et transformer les données des systèmes existants pour les intégrer au système informationnel;
– des tableaux de bord qui vont permettre de présenter les données de façon synthétique.

Transforming Community Economies

Communities across Canada are planning for significant population and job growth. They do this through Official Community Plans and Zoning Bylaws that regulate how land can be used, whether it’s for homes, jobs, shopping, parks, schools, and more.
Many of these communities were once established either as trading posts near a wealth of natural resources that could be shipped across the country and the world, or as bedroom communities within driving distance to the major service centres of bustling cities. This legacy remains an important component of local economies. Though today, a greater awareness of environmental challenges and the climate crisis is forcing communities to explore ways to green their economies.
How best can a community planner support this shift towards a more sustainable and less carbon-intensive economy?

Conception d’un algorithme LSTM de marche instable chez les personnes âgées

Azimut Médical est une entreprise montréalaise spécialisée dans le développement de vêtements gonflables de protection intelligents spécifiquement adaptés à la prévention des fractures de la hanche chez les personnes âgées. En effet, chaque année, plus de 330 000 personnes âgées souffrent d’une fracture à la hanche en Amérique du Nord seulement. Ces fractures coûtent plus de 20 milliards de dollars au système de santé et causent plus de 70 000 morts annuellement. De plus, au Canada seulement, les chutes représentent la cause la plus importante de blessures chez les personnes âgées. La moitié des personnes atteintes d’une fracture de la hanche perd l’habileté de marcher dans l’année qui suit de façon permanente, résultant par un impact majeur sur sa qualité de vie.
La problématique commune d’Azimut Médical et du stagiaire à travers ce mandat est de participer au développement d’outils intelligents permettant la prévention des chutes chez les aînés.
Pour répondre à cette problématique, Azimut Médical développe le produit Air Séqur: un vêtement gonflable novateur et intelligent qui permet une détection anticipée de la chute et séquentiellement, un déploiement de coussins gonflables absorbant l’impact de la chute.

Product Manager

NTWIST’s purpose has at its core the contribution to one of the most significant challenges globally and nationally: the Net-Zero Emissions goal. In this project, we will implement in-house developed digital technologies to help move forward into the industrial sector’s energy transition process. Particularly, we will implement a software solution powered by AI that will leverage energy efficiency in the mining sector significantly.

Contrast/modality-invariant anatomical recognition in clinical scans for multiple sclerosis study

Multiple sclerosis (MS) is a chronic and progressive autoimmune disease that affects the central nervous system (CNS), including the brain and the spinal cord. In MS, the immune system attacks the myelin that covers the neuronal fibers, resulting in inflammation and damage of CNS that leads to a wide range of symptoms concerning movement, sensation, and cognition. Big data analysis with multi-contrast/modality medical imaging can help improve our understanding of the disease. However, when collecting multi-centre clinical brain and spinal cord scans of MS patients, the file names are usually uninformative for the anatomy within the scan, making manual data organization for downstream analysis time-consuming. To tackle these challenges, the project will develop a new algorithm using few-shot learning to allow contrast-invariant anatomical identification. The resulting algorithm can greatly improve the efficiency of data collection and organization for the study of MS and biomarker discovery.

Numerical Study of the Tillage Forces on Soil Plows

The industrial drainage plow is a mechanized tool used for the installation of agricultural drainage and land reclamation. A large machine controls the depth and angle of the mole to determine the grade and depth of the tile for piping or cable to be installed. The plow is used for agricultural drainage, trenching, and installation of underground cable ducts. To make the industrial partner’s plow more competitive in the industry, this research aims to optimize the plow’s shape for various soil conditions to increase the tile installation speed and to reduce the drag force, which will reduce the fuel consumption. Numerical methods will be used in this study to predict the tillage forces on the plow. The numerical models will be validated by comparing the numerical results with available experimental data. Numerical simulations will be carried out for the plowing processes under different working conditions with different plow shapes. Also, the load distribution on the mole will be obtained based on the numerical results, which will allow the optimization of the structural components of the machine to make the machine more efficient and more durable.

Dynamic In-Game Tracking and Mapping

This research proposes a way to make speed up video game testing by creating a tool that can understand how a player moves through the game without any special information. Usually, video games are tested manually to find bugs and problems, which takes a lot of time and effort. The proposed solution would use gameplay footage to automatically figure out how the player moves and remember the surrounding environment. This information would make testing easier by reducing the manual effort involved. With faster testing in place, video games can be created faster, which will be a great benefit for video game companies and the industry in general.

Picking pithy plants: pith selection by wild, white-faced capuchin monkeys (Cebus imitator)

In this project, I examine the process of food selection by wild, white-faced capuchin monkeys using a novel multi-methodological approach. Capuchins are highly omnivorous, eating over 100+ fruit species as well as dozens of invertebrate prey species. However, when it comes to plant pith, the spongy interior of tree branches, capuchins are uncharacteristically picky, consuming only a handful of species of 300+ tree species available. To understand this selection process, I examine how dietary pith species differ from non-dietary pith species in multiple criteria, including mechanical hardness, relative amount of pith, macronutrient composition, and chemical profile. In my Mitacs Globalink internship, I will develop skills in chemical analysis by analyzing plant pith chemistry through gas chromatography-mass spectrometry at the German Center for Integrative Biodiversity Research (iDiv) in Leipzig, Germany. I will identify the chemical compounds that differentiate dietary piths from non-dietary ones, seeking ultimately to understand the chemical basis of food choice by a wild omnivorous primate.

Social structures in Brown Bears

The research project focuses on getting new insights in the social behavior of Brown Bears. The social system of a species describes the pattern of relationships between and among individuals and social groups and how these are distributed across the landscape. Animal sociality can affect individual fitness, and vice-versa, and is often a crucial factor in population management and sustainable long-term conservation. As human presence, e.g., hunting, is affecting population dynamics, instable social structures within a population might strengthen this effect over time. Thus, advancing such knowledge further is paramount for the conservation and sustainable management of wildlife populations. The visit to Canada would allow the student to get in contact with renown social behavior researchers and would greatly extend his skills/understanding. By participating in symposiums, the student would be able to show his research and share his ideas with a large crowd. The stay would also make it possible to work closely with the PhD students Canadian supervisors, which will have benefits for the ongoing PhD project.

3D Printed Bioreactor System for Analysis of Trabecular Bone Remodelling

Osteoporosis is a prevailing skeletal disease that was diagnosed in over 2.3 million Canadians as of 2016-2017, with 80% of this population being women. Osteoporosis imposes many health complications such as increased likelihood of bone fractures and implant failures. Since bone is a dynamic material that is constantly remodelling, it’s important to understand how osteoporotic bone remodels and how different factors such as mechanical or biochemical stimuli affect this remodelling. Previous studies have successfully implemented a polycarbonate bioreactor system which can keep bone alive for up to seven weeks for long-term observation of trabecular bone remodelling. Studies in the Center for Health Innovation (CHI) have implemented a 3D printed bioreactor which is significantly less expensive than the polycarbonate bioreactor, and increases the possible bone core size to meet mechanical compression testing standards. The present study aims to validate this 3D printed bioreactor system in the Centro de Bioingeniería lab in Viña Del Mar, Chile. Bovine bone cores 10 mm x 10 mm will be maintained and observed over a 21-day period for bone mineralization while being subjected to mechanical and biochemical stimuli.

Cyberentrepreuneur profiles based on the vulnerabilities they exploit and sell

The project aims to distinguish profiles of software vendors that allow the exploitation of computer security vulnerabilities. Data about the vendors and the vulnerabilities they exploit in their software will be collected from illicit software sales forums. The vulnerabilities will be classified according to what they can do when exploited. Then profiles will be developed based on the different types of vulnerabilities sold by each vendor. The profiles will be developed using artificial intelligence and interpreted with a criminological perspective. The creation of profiles allows the production of cyber security intelligence and contributes to the prevention of cyber attacks. The advantage for the partner organization lies in the integration of a criminological perspective to the technical analyses. Through this research, high-value cyber threat intelligence will be integrated in the product, thus better protecting the partner’s clients, including Canadian enterprises.