Projets novateurs réalisés

L2M – The Edge-Aware Diffusion for Contrast Enhancement

Artificial intelligence tools for contrast enhancement in medical imaging often improve visibility but may remove subtle structural details, reducing diagnostic reliability and clinical confidence. This project focuses on validating and preparing for the commercialization of the Edge-Aware Diffusion for Contrast Enhancement framework, which uses an edge-aware diffusion process to improve image contrast while preserving fine anatomical boundaries.

Through the Lab2Market Validate program, this internship will connect with radiologists, imaging engineers, and artificial intelligence specialists to evaluate clinical usability, interpretability, and integration needs. Insights from expert interviews will guide prototype refinement and the development of a business model and commercialization plan.

The project supports the growth of Canada’s medical technology sector by combining scientific validation with stakeholder engagement to build safer, more reliable imaging tools that improve diagnostic accuracy and reduce healthcare costs.

L2M – SensRIM

For manual wheelchair users, mobility means pushing thousands of times each day, often leading to pain, fatigue, and upper-limb injuries. SensRIM is an innovative, sensor-integrated ergonomic pushrim designed to transform this daily effort into data-driven mobility, enhancing independence, comfort, and health outcomes. By embedding soft tactile sensors that directly measure propulsion forces and detect user intent, SensRIM provides a novel, intuitive control input surface for power-assist applications while delivering real-time feedback on propulsion technique. This enables smarter power-assist control, fitness tracking, and clinical insights for rehabilitation. Bridging the gap between manual and power-assist mobility solutions, SensRIM offers a lightweight, ergonomic, and low-maintenance alternative that minimizes the moving parts found in traditional force-inferred control systems. With broad potential across rehabilitation, smart mobility, and health monitoring, SensRIM addresses critical unmet needs in a global wheelchair market projected to exceed USD 10 billion by 2032, offering a smarter, safer, and more independent approach to mobility.

L2M – GradientX

Nowadays, many young Canadians [2] and newcomers [1] begin their financial journey with limited access to guidance on saving and investing, which leaves them at risk of debt and missed opportunities to build wealth. Existing tools are often complex and designed for experienced investors, not beginners. This project aims to develop and validate GradientX, an AI-powered financial tutor that helps individuals improve their financial literacy through personalized guidance, real-time market insights, spending analysis, and a risk-free virtual investing “sandbox“. Through participation in the Lab2Market program, the project will conduct user discovery and market validation to identify key pain points and refine the business model. The collaboration will benefit Lab2Market by contributing to its mission of translating academic research into viable market solutions and fostering Canada’s next generation of entrepreneurial researchers.

L2M Launch – Commercialization of Natural Polymer-Based Wound Dressings

In collaboration with the Lab2Market Launch program, our goal is to transition our natural polymer-based wound dressing from the lab bench to commercial scale. Designed specifically for individuals with chronic wounds and fragile skin conditions, our hydrogel-based dressing aims to fill a critical gap in the market. Through the program, we seek to
deepen our understanding of the market landscape by interviewing potential clients, including healthcare professionals and patients. Their feedback will be invaluable in refining our product to better meet user needs.
Additionally, the L2M program provides an opportunity to connect with industry experts who have successfully brought similar technologies to market. We will also gain insights from regulatory specialists on securing government clearance and protecting our intellectual property.

L2M – Miniature EEG BCI with Wireless Power Charging

L2M – Miniature EEG BCI with Wireless Power Charging Proposal

L2M – Low-carbon modular concrete buildings

Housing shortage is one of the most serious societal issues that Canada is dealing with. On the other hand, the shrinking number of workforce is expected to increase the construction duration and cost of the new project. Furthermore, the influence of climate change in the form of wildfires and hurricanes threatens the conventional methods of construction. As such, the current project sets out to understand the market pull for a new mode and material of construction, i.e., low-carbon pre-cast and modular concrete structures. The factory-built structures are expected to increase the construction speed and enhance the quality of construction. The concrete product that are designed to use the minimal amount of cement and develop strength by capturing CO2 will benefit from climate resiliency and reduced environmental impact.

L2M ScaleUp

ScaleUp is a marketing tool that uses AI to help small businesses plan, create, schedule, and measure social posts in one place. This project will test ScaleUp with real firms, learn which features matter most, and shape an easy, affordable package. The intern will run interviews and small pilots, compare results before/after use, and report clear time savings and better engagement. The partner organization benefits from a validated go-to-market plan (target users, pricing, onboarding), a pipeline of early adopters, and practical evidence of productivity gains. The outcome is a ready-to-launch product with a simple dashboard, human-in-the-loop approvals, and “bring-your-own-API-keys” for data control.

L2M Validate / Qc Winter 2026 / Smooth Detector: Combating Fake News in the Society with AI

The Smooth Detector project addresses the urgent problem of multimodal misinformation, fake news that combines misleading text and manipulated images, which spreads rapidly across social media and overwhelms small news organizations, fact-checkers, and public health agencies that lack the resources to combat it effectively. Smooth Detector analyzes posts across multiple formats simultaneously, identifying subtle inconsistencies that reveal forgeries, such as persuasive text paired with mismatched images. Unlike existing tools that provide simple yes/no answers, Smooth Detector returns confidence scores that help users prioritize which content needs urgent human review and which can be safely published or automatically flagged. The system is designed to be affordable and easy to use, delivered as a simple plugin for WordPress and Drupal websites or as a cloud-based API, making advanced detection accessible to community newsrooms, independent bloggers, and regional health communicators who currently have no practical defense against sophisticated misinformation. This MITACS internship will accelerate commercialization by executing customer pilots with 3-5 partner organizations, building the technical integrations (plugins and dashboards) that non-technical users need, collecting real-world performance data to validate detection accuracy and time savings, and developing the business materials required for market entry and investor fundraising. The expected benefit to Smooth Detector is a validated, market-ready product with proven customer demand, deployed infrastructure, documented impact through case studies, and strategic partnerships that position us to scale across Canada’s 7,000+ community news outlets and 600+ million blogs worldwide. This directly supports our mission to democratize misinformation defense and protect public trust in critical information sources.

Noninvasive tumor genotyping for cancer detection

Non-small cell lung cancer is one of the leading causes of cancer-related deaths, mainly because relapses are often detected too late. This research project aims to validate an innovative method to monitor patients after treatment by detecting tiny traces of tumor DNA circulating in the blood. Using a highly sensitive genome sequencing approach, it will be possible to identify a recurrence of the disease much earlier than with current imaging techniques. For the Institut du cancer de Montréal and its partners, the expected benefits are significant: providing doctors with a reliable tool to adjust treatments more quickly, reducing the need for costly follow-up exams, and improving both quality of life and survival for patients. For the public, this means access to more effective care, less stress and uncertainty during medical follow-ups, and a more efficient use of healthcare resources. Finally, this project will strengthen Canada’s leadership in precision oncology and deliver concrete benefits to society as a whole.

L2M – IUAide by ReFem Technologies

ReFem Technologies is developing IUAide, a new medical device that makes intrauterine device (IUD) insertion safer, faster, and less painful. The current procedure uses outdated tools that cause discomfort and inefficiency for both patients and clinicians. IUAide combines gentle stabilization, local anesthesia, and real-time placement feedback into one easy-to-use device. Through this project, ReFem will refine its product design, test its prototype, and strengthen its business and financial strategy. The project will help the company prepare for investment and market launch, improving women’s healthcare while supporting innovation and job creation in Canada.

L2M – Electrochemical cellulose nanocrystal production

Our process takes bulk cellulose and upgrades it to cellulose nanocrystals — a valuable nanomaterial. This novel method now holds a provisional patent, but we need Lab2Market (L2M) to help us validate our process and implement it in industry. L2M will aid in identifying our potential customers, making sure our tech links well with existing infrastructure, and make a business proposition.

Système d’optimisation de la surveillance de la culture via des robots en utilisant les LLVM et l’optimisation VSLAM3D

L’agriculture dans les régions nordiques fait face à des défis majeurs : saisons courtes, conditions climatiques difficiles et manque de main-d’œuvre. Ce projet vise à développer des robots agricoles capables de travailler de manière autonome dans ces environnements exigeants, sans nécessiter une supervision humaine constante.
Grâce à l’intelligence artificielle, ces robots pourront non seulement se déplacer dans les champs, mais aussi comprendre ce qu’ils voient et prendre des décisions adaptées. Imaginez un robot capable de reconnaître différentes cultures, détecter des maladies des plantes et planifier ses actions simplement à partir d’instructions en langage naturel, tout comme un travailleur agricole expérimenté. Le projet combinera deux technologies innovantes : des systèmes de navigation visuelle qui permettent au robot de se localiser précisément dans l’environnement, et des modèles d’intelligence artificielle capables d’interpréter des images et du texte pour comprendre les tâches à accomplir. Les résultats contribueront à améliorer la productivité agricole dans les régions nordiques, renforcer la sécurité alimentaire locale et réduire la dépendance à une main-d’œuvre qui se fait de plus en plus rare.