Association automatique entre symboles et textes dans des documents d’ingénierie

La plupart des documents d’ingénierie comportent des symboles pour caractériser les systèmes qu’ils représentent. Ils comportent également des annotations (identifiants, notes, spécifications) sous forme textuelle pour préciser certaines propriétés importantes ou identifier les composants. Si l’association entre les symboles et les composants est intuitive pour un expert humain, il en va autrement pour un ordinateur. […]

Read More
Azurdev

Azurdev is a new venture oriented towards providing better patient care in the world’s hospitals. The team behind Azurdev has over 20 years of experience in providing digital entertainment platforms for the Hotel industry, and more recently to the Quebec hospitals. Owing to the market demand of digital health, Azurdev, in 2019, decided to develop […]

Read More
Élite Neurokinetix Inc

Élite Neurokinetix (ENK) développe une plateforme web facilitant le transfert de connaissances scientifiques aux athlètes et entraîneurs, afin d’accompagner l’entraînement, et d’optimiser la performance sportive, la sécurité et le bien-être des athlètes. La plateforme est ancrée dans une approche inclusive et centrée sur les besoins de l’athlète pour favoriser leur développement positif à long terme.

Read More
Machine learning for real-time parameters estimation and control of robotic laser cleaning

Surface cleaning is a technology used in a wide variety of industries, from heavy manufacturing and the energy sector through to conservation and restoration. Historically, technologies such as sandblasting and pressure washing have been used which have significant environmental and waste challenges. More recently, laser ablation has been used for surface cleaning. This last technology […]

Read More
Reduction of Software Rework Through theMitigation of Cognitive Biases

The work proposed will lead to mechanisms for reducing or eliminating selected Cognitive Biases – automatic, unconscious elements of the human reasoning system known to cause decision errors. While Cognitive Biases have been studied for decades, there has been almost no research into how to reduce or eliminate their effects. The end goal of the […]

Read More
Prédiction d’un indicateur de performance de ligne de production

Le passage des entreprises à l’industrie 4.0 à pour but de propulser la productivité, réduire considérablement les coûts de production et d’améliorer grandement la qualité des produits. Le projet a pour but de démontrer à l’entreprise partenaire que l’utilisation de modèle prédictif d’apprentissage machine afin de prédire un indicateur de performance (KPI) de ligne de […]

Read More
Disaster Recovery and Cloud Bursting as a Cloud Service

Cloud computing has revolutionized the way organizations consume computing power as a service. Infrastructure as a Service allows a company to move away from purchasing computing, networking, and storage resources to purchasing from a service provider in a public cloud. This project will research and develop off-site encrypted virtual machine storage and disaster recovery planning […]

Read More
Understanding and designing the female pelvic anatomy, a measuring device and an intravaginal device using 3-dimensional modeling techniques and Artificial Intelligence

Pelvic Organ Prolapse (POP) is a condition 1 in every 10 women is diagnosed with. The current non-surgical treatment for POP is an intravaginal device called pessary which has a 40% failure rate as its shape is not fitted to the female anatomy. Poor pessary design and performance arises from the limited data that is […]

Read More
Improving Health Information System Safety: Development of Novel Approaches for Identifying, Tracking and Preventing Technology-Induced Error

In Canada healthcare is being modernized and transformed through a range of new healthcare information technologies and systems. Applications of information systems such as electronic health records (EHRs), electronic medical decision support and an increasing range of mobile health applications promise to transform and improve healthcare and increase patient safety. However, although such technology has […]

Read More
Seq2Hypergraph: Link prediction on knowledge graphs using relation conditioned Transformer Networks

Learning from relational data is crucial for modeling the processes found in many application domains ranging from computational biology to social networks. In this project, we propose to work on developing new modeling techniques that combine the advantages of the approaches found in two fields of study: Machine Learning (through graph neural networks and transformer […]

Read More