L2M – Flow

The goal of this project is to validate and improve the scientific, clinical, and user-centered value of Flow’s AI-powered features in preparation for commercialization. The focus will be on developing a structured evaluation framework, conducting hands-on testing, and iterating based on clinician feedback. This complements the customer discovery and business model testing activities of the […]

Read More
Management and mining of big energy sector data

With advances in techniques, high volumes of valuable data are generated in many domains (e.g., energy sector) at a rapid rate. Consequently, a scalable and flexible system for efficient storage and fast management of these distributed data is needed. In this proposed research project, we plan to design and implement a cloud-based data storage & […]

Read More
Applied AI for EdTech: Automated Reporting and Analytics in VidaNovaVLE

Fenix Alma Solutions Inc. is a Canadian EdTech company that develops VidaNovaVLE™, a purpose-built, partner-driven Virtual Learning Environment designed specifically for medical and health sciences education. The platform supports curriculum management, assessment, clinical scheduling, competency-based medical education, and accreditation reporting for leading institutions across North America. As the number of institutional partners grows, the volume […]

Read More
Implémentation d’un pipeline TALN pour la structuration de textes cliniques par concepts médicaux normalisés (UMLS, SNOMED-CT-CA)

Le projet vise à développer une infrastructure capable d’exploiter les informations cliniques contenues dans les dossiers médicaux électroniques. Une grande partie des données médicales du Québec est consignée dans ces dossiers, mais les textes n’ont pas de structure précise et restent difficiles à exploiter automatiquement. Cette complexité peut ralentir la prise de décision médicale et […]

Read More
Adversarially Resilient Federated Learning for Intrusion Detection in 5G Networks

This project will design and test a new privacy-preserving, attack-resilient cybersecurity system for 5G networks using Federated Learning (FL), an approach where multiple devices can train models together without sharing sensitive data. The goal is to improve protection against cyberattacks such as denial-of-service and model poisoning, which threaten the reliability of modern mobile services like […]

Read More
L2M – AI Powered Seed Analysis

As part of the Lab2Market program, this project explores the commercialization potential of the group’s previous research pertaining to the use of computer vision, artificial intelligence and machine learning to precisely detect and identify seeds. Over the 4-month duration of the project, the team will participate in Lab2Market workshops, panels, and lectures from successful founders […]

Read More
AI based Animal Behavior Recognition: Dog Pose and Labeling

This project will develop a real-time Artificial Intelligence based system that accurately estimates 3D skeletal pose from monocular video of dogs and reliably classifies fundamental behaviors to support veterinary health monitoring and behavioral assessment applications. Recent research has demonstrated significant progress in automated animal behavior recognition. The intersection of computer vision, animal behavior analysis, and […]

Read More
Linguistic Cues for the Extraction and Synthesis of Speech in Extreme Noise

Malaspina Labs Inc. develops algorithms for enhancing speech in extremely noisy audio signals for real-time applications such as hearing aids and mobile devices. Many classical approaches to speech enhancement are not applicable to these domains because of severe hardware and psychoacoustics constraints such as far-field, single microphone use cases (no beamforming),

Read More
L2M: AI-assisted marine ecosystem monitoring

This project aims to turn raw underwater video into decision-ready science. Our software automatically detects, classifies, and counts marine species, then delivers results in a simple web dashboard with linked video clips, maps, and exportable reports. Instead of annotating from scratch, ecologists just review and correct AI suggestions—cutting review time by more than half. Built […]

Read More
GOWL-Edit, un assistant logiciel pour l’édition graphique d’ontologies OWL 2, de requêtes SPARQL et de règles SWRL (Semantic Web Rules Language) dans la syntaxe G-OWL étendue

L’Institut de Recherche en Électricité du Québec (IREQ), qui est le centre de recherche d’Hydro-Québec, a pris le virage des technologies du web sémantique depuis quelques années. L’IREQ doit gérer une quantité énorme d’informations provenant de ses équipements réparties dans tous le Québec. Les chercheurs de l’IREQ ont notamment choisis les technologies du web sémantique […]

Read More
Inventing the Future of AI Applications: Applied Research in Machine Learning at AXL

AXL Labs is the technical arm of AXL, a Toronto-based venture studio that creates and launches companies focused on human-centric artificial intelligence (AI). Their main goal is to leverage human-computer interaction (HCI) and AI in designing and deploying end-to-end solutions for industry and academic applications. Organizations that partner with AXL typically have business problems where […]

Read More
Debugging ML via Feature-guided Analysis: Analyzing Neural Network Robustness

Neural Networks (NN) use a set of individual units (neurons) connected together to learn a specific behavior from a dataset. For example, NN excel in classification tasks where given a dataset labelled with presence or absence of a feature in each entry, they are able to detect the feature presence on new inputs. This technology […]

Read More