Projets novateurs réalisés

Alouette – ML demand prediction platform for SMEs

Small and medium-sized enterprises (SMEs) face challenges such as inflation, supply chain disruption, and extreme weather, resulting in missed orders, overstocking, or understocking. Traditional forecasting methods based on seasonality and trend cannot provide accurate market demand, leading to inefficiencies and sales losses. The proposed research project aims to develop a machine learning-based demand prediction platform. The platform will analyze large amounts of data from internal and external sources, identifying difficult patterns and relationships to provide accurate market demand predictions in real-time. Apana will integrate with this platform as the “Software-as-a-Service” model. This will enable SMEs to perform real-time analysis of incoming data for their own demand forecasting. The overall objective of this project is to establish Apana as the go-to data hub for SMEs to manage their retail operations. We aim to grow our business based on the scale of users and the extension of different types of data services.

Évaluation de l’effet antinociceptif des opioïdes courant chez les poissons en utilisant le modèle de la shuttle box

L’objectif de ce projet est de trouver des médicaments antidouleurs pour les poissons en observant leur
comportement dans une eau à température variable. Le but est de différencier un effet analgésique d’un effet
sédatif des médicaments testés en analysant la vitesse de nage des poissons. Les mesures seront effectuées à
distance à l’aide d’une caméra et d’une analyse par ordinateur sans interaction humaine afin de limiter le stress
des poissons. Ultimement, ce projet de recherche permettra d’améliorer la prise en charge des conditions
douloureuses chez les poissons, qui peuvent survenir lors de maladie, de traumatisme avec des congénères ou
suite à des interventions humaines. Dans un Aquarium, il est important de préserver le bien-être des poissons de
la collections et de prolonger leur durée de vie par des chirurgies au besoin. Les résultats de l’étude serviront
donc aussi les besoins des pensionnaires de l’Aquarium de Montréal.

Assessing Support Needs of Tenants in Two Ottawa Community Housing Buildings

Some people living in social housing are at risk of losing their housing. This is due to serious mental illness, substance use, neurocognitive decline, physical frailty, disabilities, or other issues, along with an absence of adequate support. Ottawa Community Housing has approached researchers at the University of Ottawa to conduct a study to learn more about the needs of these tenants and to help identify the resources and services that can be drawn upon to assess, place, and support these clients. The current study will provide OCH with data to inform its next steps in advocating for system change around housing selection, placement, and conditions, and securing supports for tenants at risk of losing their housing. The research will also inform the social housing field where concern over this issue is mounting but there has been a very limited amount of research.

Development of a Performance Measurement and Quality Improvement Strategy at Ottawa Salus

Ottawa Salus is a non-profit corporation who have served the Ottawa community for over 40 years. They have over 100 staff members delivering supportive housing and mental health services to the community. The organization creates opportunities for adults living in the community with mental health and substance use issues to live independently. To achieve this, they have a strong commitment to improving the quality of their services. The goals of this project are to first make a committee of several people from the organization and consult with them during the project. The intern will conduct interviews with different staff members about how they use quality improvement information (e.g., client surveys). Finally, in consultation with the committee, the intern will produce recommendations for what the organization can do to develop a quality improvement strategy.

Seamless Day Programs: Opportunities, challenges, and Promising Practices

As part of its 10-year ChildCare plan, the Government of BC recently funded “Seamless Day” pilot programs in various communities across the province. These programs involve service sharing between schools and child care providers, with the intention of providing continuous care, support, and learning for young children throughout the school day, as well as before and after-school. The proposed project is designed to support the interest of the Early Childhood Educators of BC (ECEBC) in exploring emerging opportunities, challenges, and promising practices related to Seamless Day programs. The work will assist ECEBC in supporting educators who work in Seamless Day programs, as well as inform the organization’s broader efforts to advance early childhood care and learning in the province. Using a participatory research approach, informed by pedagogies outlined in the BC’s Early Learning Framework, the project will engage a range of collaborators in dialogue and action toward transformative change.

Model validation of high-speed imploding liquid cavities at industry scale

Fusion power has the potential to deliver clean, safe, and abundant electrical power on an
industrial scale. Due to the extreme conditions required, involving temperatures of hundreds
of millions degrees Celsius, it remains a technical challenge, particularly on a commercial
scale. In a concept called Magnetized Target Fusion (MTF), plasma is compressed to fusion
conditions using an imploding cavity of liquid metal; the liquid is spun by a rotor to form a
cylindrical cavity which is imploded by injecting fluid through channels in the rotor walls. This
project will validate our current model of an imploding liquid cavity at industry scale, and
investigate technological designs by which the implosion efficiency can be further improved.

Modèles de cohabitation entre personnes aînées et personnes étudiantes : un examen de portée

Ce projet, imaginé par la FADOQ – Région Estrie, vise à former un tandem bigénérationnel en matière de partage de logement en regroupant des personnes aînées qui acceptent de loger des personnes étudiantes universitaires recommandées par l’Université de Sherbrooke, moyennant différentes modalités. L’intérêt d’une personne aînée à participer au projet se traduit, entre autres, par une présence humaine rassurante et par certains services que la personne étudiante pourrait lui rendre pour favoriser son maintien à domicile. De son côté, la personne étudiante, faisant actuellement face à une rareté de logements, pourra bénéficier de l’accès à un nouveau bassin d’habitations locatives à des coûts et à des conditions fort intéressantes ainsi que de la présence de la personne ainée favorisant son intégration sociale dans la région.

Avion sans pilote dans l’atmosphère martienne

Objectif: Conception préliminaire par analyse seulement d’un avion capable d’effectuer un vol propulsé et autonome dans l’atmosphère de Mars. Travail individuel sous supervision.
– Préparation: conception d’un avion, atmosphère martienne, revue de la littérature scientifique
– Liste des requis, mission typique, préparation du cahier des charges
– Estimation du poids initial
– Performance aérodynamique: charge alaire
– Aérodynamique des profils d’aile
– Dimensionnement de l’empennage vertical et horizontal, surfaces de contrôle
– Génération de maillage (Pointwise), solveur fluide (Ansys Fluent ou STAR-CCM+, ou autre)
– Simulation numérique (CFD): aérodynamique des ailes et de l’ensemble de l’avion, courbe polaire
– Systèmes de propulsion et d’énergie, dimensionnement des moteurs électriques, des cellules photovoltaïques et des batteries, description de leurs caractéristiques respectives et contraintes en opération, etc.
– Performance: décollage et atterrissage
Si le temps le permet:
– Environnement du lanceur (Terre)
– Charges et structure
– Stabilité et contrôle

Data-driven Automated Hypovigilance Detection using Physiological Parameters

In this research project, the intern will explore ways to detect hypovigilance with vital signs collected from the Intensive Care Unit (ICU). The intern will first study existing research and models to understand which features are important for detecting these conditions. Next, they will analyze vital sign data from ICU patients to identify the most informative features for detecting hypovigilance and delirium. Using this information, the intern will then apply statistical analysis and machine learning techniques to develop models that can accurately detect these conditions. Different combinations of input variables will be tested to find the most effective model. The performance of these models will be evaluated using various metrics, and they will be optimized for use with new patients. Finally, the intern will write a scientific paper to share the project’s findings. The partner organization will benefit from improved methods for detecting hypovigilance which is very important in the transportation sector.

Investigating the endothelial cell population in different tissues using Raman microspectroscopy

Healing Trauma Using Digital PTSD Therapies for Anyone, Anywhere

Savyn offers digital therapies based on Eye Movement Desensitization and Reprocessing (EMDR) and written exposure therapy, which are digitized and adapted versions of existing therapy protocols provided in person to heal trauma and PTSD. Savyn offers access to patients and clients who have PTSD but cannot get the help they need due to a shortage of therapists, higher costs, and geographic barriers. The goal is to heal trauma. We are removing barriers to accessing PTSD therapies by digitizing and bringing virtual therapies with proven efficacy in patient populations. Savyn’s EMDR offering is currently undergoing clinical validation with PTSD clients at Sick Kids Research Institute, in collaboration with the Principal Investigator, Professor Ben Dunkley at the University of Toronto. The studies so far show promising results to the patients using the digitized EMDR by improving distress levels and improving PTSD symptoms by the fifth and sixth sessions.

Secure and Practical Federated Learning with Multi-Key Homomorphic Encryption

Nowadays, our data is stored and processed in the cloud, and there are increasing privacy concerns. Federated Learning (FL) has recently emerged to allow users to keep their data private and collaboratively train machine learning models. To ensure that no client data can be revealed, there are multiple challenges in designing a system to be used in real applications: (1) different types of heterogeneity (data, resource, infrastructure), (2) efficiency, and (3) scalability. In this project, we aim to design a secure and flexible FL system, SmartFL, with multi-key homomorphic encryption (MKHE) algorithms. From this project, we expect peer-reviewed top conference or journal papers and patents. We expect that by the end of the project, the working prototypes will be developed, which can be merged into HCL Canada Inc’s product in the future. The prototype will be demonstrated to the partner organization to show its scalability, efficiency, and practicality.