Innovative Projects Realized

Machine Learning Aided Detection of Brain Aneurysms

Intracranial aneurysms are relatively common, occurring in 2-5% of the general population. Rupture of the aneurysm can result in a stroke with a devastating 30-day mortality of 45%. Further, severe medical conditions are possible in which up to one third of patients may die before reaching the hospital and one third will become severely debilitated. Aneurysms are, however, difficult to find on magnetic resonance and computed tomography angiography scans (MRA and CTA scans) – especially when small or located close to the bone. Through the use of the existing knowledge and expertise of the neurovascular team at Toronto Western Hospital, as well as their database of CTA and MRA scans, the aim of this project is to improve on efforts in this field through the development of an artificial intelligence and machine learning method for the detection of intracranial aneurysms with a diagnostic accuracy approaching that of trained neuroradiologists. As unspecialized general radiologists read most neuroimaging exams due to a lack of specialized physicians in remote and rural areas, a long-term goal is to use results from this work to build a tool for the support of imaging review by non-specialists in emergency situations.

Automated Retail Area Cluster Detection

The main objective of this project is to develop a retail cluster detection algorithm and improve the accuracy of the identification. The final deliverable will be an algorithm that runs through Google dataflow that will be able to ingest a month of users’ location breadcrumbs and output user-location clusters. The output of the algorithm will be a unique cluster identifier that will be used in the visit algorithm for visit identification. Besides latitude and longitude data, the project will have access to altitude information. This can be useful for determining clusters in multi-level locations (such as malls).This algorithm will help recognizing the customer visiting pattern observed based on user ID and allow accurate profiling for weather related content or advertisement targeting for enhanced user experience

Communication network modeling for simulation of wide-area control and protection applications in power systems

Computer simulation is a widely used tool for design and analysis of various aspects of modern power systems.

One area where simulation has been invaluable is the study of protection systems used to avoid catastrophic

failures and blackouts, by rapidly detecting and shutting down faulty equipment in the system. High-speed data

networks are becoming an integral part of the modern power grid, which uses computer-based monitoring, and

protection systems. However, relying on data networks presents a unique set of challenges in terms of ensuring

reliable and timely exchange of information between measurement units, computers, and protection equipment.

Therefore any comprehensive study of a power system protection scheme must also include an analysis of the

impact of the of the data network. At present, software tools are not available for realistic simulation of datanetworks

within widely used power-system simulation software such as PSCAD. The goal of the proposed

research is to develop a set of data-network simulation models, which can…………………..

Hybrid RNA sequencing of Douglas-firs exposed to simulations of future extreme climate events

Douglas-fir is one of the most economically important tree species in Canada and it is at risk of maladaptation to local environment due to climate change. Thus, the aim of my project is to understand the adaptation of Douglas-fir to future extreme climatic events. Our group at Ensminger Lab at University of Toronto monitored the performance and survival of four different Douglas-fir genotypes under simulations of future drought and heat events. During the experiment, we also collected samples for long- and short-read gene sequencing. In collaboration with Jill Wegrzyn, one of the world’s leading conifer genome researcher, we will create novel hybrid gene sequencing assemblies. We will also develop new algorithms to identify genes that confer improved performance in Douglas-fir under extreme climatic stress. This work will improve our understanding of plant responses to climate stressors and provide genomic tools for the identification of Douglas-fir genotypes that are adapted to perform better in a future climate. Ultimately, these tools can be used to inform tree improvement programs, and forest management practices and conservations.

Rheological and microstructural properties of micellar solutions

Surfactant solutions are of immense scientific and technological interest. They can self-assemble into wide variety of interesting and highly dynamic nanostructures including spherical, cylindrical, vesicle, worm-like or lamellar type. Micellar solutions typically contain surfactants and salts that are dissolved in water. Beyond a critical micelle concentration, the addition of salt or surfactant transforms spherical micelles to isolated rod-like micelles (in dilute regime) that at sufficiently high concentrations of salt and surfactant (in the semi-dilute regime), form entangled worm-like micelles. The projects aim at understanding the relationship between the microstructure-process-property for micellar solutions that have a broad range of applications such as detergents, paints, and lubricants. The project will be performed through employing rheological characterizations. The expected outcomes are the detailed models describing the change in the rheological properties and microstructures in the micellar solutions as a function of processing parameters such as temperatures and shear rates.

Evaluating the effectiveness of a multimodal intensive episodic treatment centre for youth Child Sexual Abuse (CSA) victims: The Be Brave Ranch

Child sexual abuse (CSA) is an all too prevalent form of childhood trauma worldwide. Nearly 1 in 5 women and 1 in 10 men experience some form of abuse in their childhood. CSA has been linked to numerous long-term effects: PTSD, depression, anxiety, substance abuse, eating disorders, sexual dysfunction, suicidal ideation, and cognitive impairment. Growing awareness of the problem has led more individuals to seek professional treatment. At present, there is no gold-standard for CSA treatment, particularly in youth. The Little Warriors Be Brave Ranch in Alberta, Canada is an innovative multimodal residential treatment program for children and adolescents with histories of CSA. We will gather mental health data from children and adolescents to measure symptom progression over time. We will also collect information from their caregivers, to gain better insight into the effects of treatment on them. Our goal is to develop a comprehensive framework for CSA treatment programs.

Water Penetration of Brick Veneer Walls Made of Clay Bricks with Large Voids

Rain water penetration can cause severe damage to building materials and facilitates the growth of mould, endangering the health of the occupants of the building. This Mitacs project will determine the water penetration resistance of various clay brick veneers used in various buildings. This research will also provide very important test data that may allow engineers and builders to use much lighter clay bricks (clay bricks with large voids) as the brick veneer. Use of lighter bricks translates to less carbon footprint and hence, is a sustainable choice. The research work will be completed using wall tests in special environmental chambers under the joint supervision of Dr. Sreekanta Das at the University of Windsor and Mr. Patrick Kelly at the Clay Brick Association of Canada.

Analyse d’impacts économiques et fiscaux d’une politique d’achat local par des organismes du secteur public à Sherbrooke

Le projet chercherait donc à établir si l’achat de biens et services locaux par les organismes de la ville tels que le Centre Hospitalier, l’Université de Sherbrooke, la mairie, entre autres, serait juridiquement possible et économiquement intéressant. L’intérêt économique pourrait concerner les organismes eux-mêmes : est-il plus intéressant pour l’acheteur de s’approvisionner localement plutôt qu’avoir recours à l’achat groupé ? Les intérêts économiques et fiscaux pourraient aussi concerner la ville entière, la province ou même le pays : est-il plus intéressant pour la ville de Sherbrooke que les acheteurs s’approvisionnent localement plutôt qu’avoir recours à l’achat groupé ? Etc. Une question centrale serait : est-ce que le désavantage/avantage pour l’acheteur peut-il être compensé par l’avantage/désavantage pour la/le ville/province/pays ?

Visual attention in deep learning for detection and classification

Visual attention refers to the mechanism of dynamically and selectively focusing on a subset of the visual input stimuli for detailed analysis, which is part of the visual perception process of the early primate vision. It has been successfully integrated into the design and implementation of many artificial visual recognition systems with applications to image classification, object detection, object sequence recognition, as well as image captioning and visual question answering. This research will explore various visual attention mechanisms with the goal to improve the generalization, robustness and efficiency of various DCNN models and algorithms developed for Epson’s computer vision and machine learning core technologies and products.

Adaptive Artificial Agent for Smart Home Thermostat

How can we make the connected home smarter and more energy efficient? Ecobee’s research aims to address this question by making cutting-edge improvements to its smart home thermostat. There is huge potential to save energy in the residential heating and cooling space – homes with an ecobee smart thermostat can save up to 15% of their energy consumption. Ecobee’s aim is to make use of customer-volunteered usage data in the creation of an adaptive agent which can control the comfort level in the home autonomously. This agent will eliminate the need for close monitoring and adjusting of the thermostat by the user and use less energy than a house with a traditional thermostat, while still retaining maximal comfort for the residents. TO BE CONT’D

Collaboration between Polytechnique Montréal and Lund University to realise a petrochronological study in the vicinity of the Allochthon Boundary Thrust (ABT) in the Grenville Province and its Sveconorwegian extent, the Mylonite Zone.

Study of the core of ancient mountain belts is key to the understanding of current tectonic processes. In Canada and Sweden, the Grenvillian and Sveconorwegian orogenies are the relict of a major tectonic event, similar to the continent-continent collision between India and Eurasia. In both locations, a major tectonic boundary i.e. the Allochthonous Boundary Thrust (Canada) and the Mylonite Zone (Sweden) is exposed and is actively studied to understand tectonic processes, specifically exhumation or the path followed by rocks from deep within the Earth’s crust back to the surface. This project focuses on the exhumation of high-grade metamorphic rocks in both locations. Samples are studied under optical microscope and scanning electron microscopy, key minerals are dated, and metamorphic conditions are calculated from trace element concentrations and thermodynamic modeling. Thus, this study will rebuild the Pressure-Temperature-time history of key samples to assess their role in their respective orogeny. Then, this new information will be used to test existing tectonic models.

Dynamic Pricing for Optimizing Demand and Profitability

In this research project, a surcharge optimization algorithm will be developed to help the partner company to dynamically determine the premium charged for order pickup. The objective of the project is to smooth the demand curve for the order pickup timeslots: popular timeslots are congested, making it harder to deliver a positive experience, and other windows are idle, and overstaffed. In the project, large volumes of purchase history data including pickup and delivery “hot-spots”, historical pricing information, historical inventory, and shorts on pickup and delivery orders are to be utilized to determine the premium to be charged to improve both overall labor planning efficiency and customer experience.