Projets novateurs réalisés

Detection, characterization and analysis of unsafe video content on YouTube

A safe video platform can provide a healthy and suitable environment for users in general and children in particular. This project aims to utilize machine learning and deep learning techniques to identify and flag sensitive and questionable content (e.g., content related to violence, sexuality, etc.). The algorithm will leverage video frames extracted from the database for training and building a 3D Convolutional Neural Network (CNN) model. This model can detect and classify videos just like a human being. With this technology, BBTV can provide a more appropriate channel management solution to content creators that leverage its services.

Defect Functionalized Sustainable Energy Materials: From Design to Devices Application

Oxide materials have many interesting properties, in particular in the field of sustainability sciences. The Hosono group at the Tokyo Institute of Technology is world leading in synthesizing novel oxide materials with very interesting electronic properties with potential applications in the fields of superconductivity, catalysis and optically transparent electrodes for photovoltaics or displays. I will gain insights and experience in synthesizing these materials in Japan. To characterize the homogeneity of the materials I will use scanning probe techniques developed in the Grutter lab at McGill University. Defects play an important role in these materials – the outcome of this project is to measure the spatial homogeneity of the properties of oxide based materials and thus develop a deeper understanding of the role of defects in determining the material properties.

Mesure en temps réel de la cognition, de l’affect et de l’apprentissage dans les jeux vidéo

La première expérience de jeu (les premières 60 à 90 minutes) est déterminante, car c’est durant cette période que le joueur décide s’il aime le jeu, s’il désire continuer à jouer, et s’il peut le maîtriser. Relativement à l’apprentissage du jeu, le jeu propose un tutoriel visant à habileté le joueur au regard des principales fonctions à maîtriser. Ce tutoriel est intégré au scénario, avec le souhait de ne pas nuire à l’expérience de divertissement. Or, il est possible que ces tutoriels ou d’autres éléments du jeu nuisent au divertissement et/ou ne permettent pas de maitriser les fonctions pour jouer, ce qui peut conduire au rejet du jeu, alors que son achat représente souvent une somme substantielle. Le projet vise à répondre à la question suivante : quel est l’impact des tutoriels sur l’affectivité et la cognition du joueur dans sa première expérience de jeu? TO BE CONT’D

Développement et applications d’une source X dans la fenêtre spectrale de l’eau

Les chercheurs de diverses communautés scientifiques rêvent de pouvoir utiliser le rayonnement X dans un contexte d’imagerie où seraient combinées une résolution nanométrique et temporelle femtoseconde. Par exemple, une source ultra-brève dans la fenêtre spectrale de l’eau (2.3 à 4.4 nm) est parfaitement adaptée à l’imagerie dynamique en milieu aqueux. Cette application, parmi tant d’autres, a motivé dans la dernière décennie la mise en chantier de Lasers à Électrons Libres. Il s’agit d’infrastructures de recherche imposantes, qui nécessitent des investissements financiers de l’ordre du milliard de dollars, et qui ne sont pas disponibles au Canada. Une autre approche possible pour développer une source de rayonnement X femtoseconde consiste à utiliser la génération d’harmoniques élevés (GHE) dans un gaz. Nous avons développé une telle source au laboratoire ALLS de l’INRS-ÉMT et l’objectif de cette collaboration université-industrie est d’y coupler une plate-forme de microscopie X pour l’imagerie dans la fenêtre spectrale de l’eau.

Développement d’une méthode de mesure de l’exposition sonore effective intra-auriculaire pour une utilisation en milieu de travail

Malgré de nombreux efforts de réduction à la source, l’exposition à des bruits de forts niveaux en milieu de travail demeure un problème majeur au Québec et à travers le Monde. La mesure de l’exposition sonore est donc primordiale afin de déterminer et évaluer des solutions correctives à apporter pour bien protéger les travailleurs. Pour diverses raisons, les méthodes actuelles ne permettent cependant pas de rendre compte de façon précise des niveaux d’exposition reçus par un travailleur donné dans son milieu de travail. L’objectif principal du projet est de développer une méthode et des algorithmes de mesure des niveaux d’exposition sonore dans l’oreille ouverte ou occluse, pour ainsi permettre la mesure de la dose de bruit intra-auriculaire pour différentes conditions de bruit et de port de protecteurs auditifs. TO BE CONT’D

Human kinematic optimal control learning and wearable inertial measurement unit alignment for rehabilitation

Physiotherapy is a type of rehabilitation that aims to restore a patient’s quality of life after an injury, surgery, or stroke by improving their mobility. Through prescribed exercises and specialized equipment, physiotherapy helps the patient to regain their muscle strength, range of motion, and natural movement. Unfortunately, only rudimentary tools are available to the therapists for assessment and monitoring of patients. Our work focuses on developing wearable technologies that can help therapists with patient assessment and progress tracking. The proposed research will utilize the sensor data to learn a motion model for a patient. The learned motion model can be used to compare the patient’s performance to that of a healthy individual and to provide objective progress measures over long rehabilitation periods.

Prediction models for pain volatility and engagement patterns of mobile pain app users

Pain is among the top 3 most common reasons for seeking medical help. ManagingLife has developed a mobile-based app, called Manage My Pain, to help chronic-pain patients by providing a simplistic, customizable and comprehensive interface to track pain symptoms and pain experience at a frequency chosen by the users. ManagingLife is interested in understanding the benefits of the app use on its 27,000 and constantly growing user base by identifying user cohorts that ultimately experience improvement in pain experience given self-disclosure tracking behaviours. We plan to develop scientific methods that will help predict the change in pain ratings, i.e., pain volatility of the app users over time. We also aim to predict the level of user engagement in future based on the current usage behavior and users’ profile information such as gender and age. Through this research, Manage My Pain platform will have analytic capabilities that will enable targeted product development, drive retention and engagement for chronic pain patients.

Potentiel pesticide des liquides pyrolytiques

Le monde agricole québécois est à la recherche d’alternatives écologiques aux produits phytosanitaires conventionnels dont l’utilisation massive provoque des effets négatifs sur l’environnement et la santé humaine. Dans cette optique-là, le projet Mitacs propose d’étudier le potentiel antifongique et herbicide de liquides produits par pyrolyse de biomasse lignocellulosique. La pyrolyse est un procédé qui décompose thermiquement et chimiquement la biomasse en absence d’oxygène. Le liquide pyrolytique produit est constitué de molécules résultant de cette dégradation. Le liquide pyrolytique possèderait des propriétés antifongiques, et des capacités herbicides grâce à une ou plusieurs des molécules qu’il contient. Ce projet va permettre aux organismes partenaires (UQTR, CRIQ et IRDA) de se positionner dans un créneau innovant pour l’agriculture québécoise.

Multivariate analysis of neuroimaging and non–invasive brain stimulation data to assess functional and structural reorganization in the brain following stroke

The brain is a dynamically evolving structure. Rehabilitation interventions take advantage of the malleable properties of the brain that persist even after an injury from stroke. Various interventions are used in clinical practice at an attempt to restore normal neurological function. Non-invasive brain stimulation is a novel technique that may act as a facilitator of recovery. Due to the complexity of processes that occur within the brain, multiple measures of neurological function are needed to capture the benefits that non-invasive brain stimulation may have on individuals with stroke. To gain a complete picture of this recovery, these

measurements should be assessed in a unified framework. This project will use advanced statistical analysis to understand the story of the brain’s road to recovery during stroke rehabilitation.

Process Improvement for Plasma Transfer Arc Additive Manufacturing

Metal additive manufacturing (MAM) is newly developed manufacturing technology that is faster and more cost-effective compared to traditional methods. Although there are various types for AM, the product quality improvement in plasma transfer arc (PTA) and wire-feed AM processes will be the subject of this project. To achieve this goal, the intern will determine the key parameters of the existing PTA2M and wire-feed machines in InnoTech and then will select appropriate measuring technologies to obtain data from these AM processes. Then, the intern will develop a model for both processes based on either first-principles or system identification strategies. In order to improve the products’ quality, he will develop a process model to estimate the unmeasurable variables and then designs a closed-loop controller. The controller will be designed based on the model and then will be implemented for the actual systems in InnoTech. TO BE CONT’D

Evaluation Capacity Network: Aligning Evaluative Thinking and Practice among Early Childhood Stakeholders

This project aims to provide much needed evidence to non-profit organizations working with vulnerable children and their families in five inner city schools. The All in for Youth initiative is a collaboration of eight organizations offering integrated, multi-dimensional supports to improve academic outcomes and resiliency of vulnerable children, support family health and stability, get communities involved, and inform policy and systems change. Individual non-profit organizations do not always have the resources, expertise, time, or capacity to intentionally gather evidence to support critical reflection of their services. Balancing the needs of all organizational representatives into one evaluation will be challenging. This collaborative of organizations will provide a case study where the ECN research team can more closely examine how partners develop intersectoral relationships, create common governance models, navigate systems change, integrate programs and services, develop common client outcomes, and collect and analyze evidence to inform and improve organizational practices, programs, and policies.

Long-term response of boreal forest bird community to changes in habitat structure and configuration due to forestry and other land uses

There are both immediate and long-term effects of forestry on boreal birds. To minimize incidental take of birds and nests during harvest, foresters need accurate tools for predicting abundance of nesting birds in the peak breeding season, to time harvests year-round. Longer-term effects of harvest and other land uses on boreal birds are unknown, because most studies of forestry practices on boreal birds occur over short periods. I will analyze >20 years of forestry treatment and bird abundance data from Alberta to 1) predict nesting bird abundance in different forest stands within a forest harvest plan; and 2) under future land use scenarios created by timber supply and land use simulation software. These analyses will predict how boreal birds respond to changes in forest structure and enable the Mitacs partners (West Fraser, Al-Pac) to minimize incidental take and create long-term plans balancing forest harvest with maintenance of boreal bird habitat.