Projets novateurs réalisés

Effectiveness of Cognitive Behavioural Therapy in patients suffering fromdepression and in receipt of disability benefits

Depression is expected to become the second leading cause of disease burden

worldwide by the year 2020. Cognitive Behavioural Therapy (CBT) is one of the most

effective methods of treatment for depression. CBT may be less effective, or

ineffective, in the setting of patients in receipt of disability benefits who are likely to, on

average, suffer worse outcomes than patients not receiving benefits. Currently, there

is no review that has systematically assessed the effectiveness of CBT in patients

suffering from depression and in receipt of disability benefits. We will examine the

effectiveness of CBT in patients suffering from depression and in receipt of disability

benefits by performing a systematic review of studies that evaluate CBT and by

analyzing the administrative database of Sun Life Financial, a Canadian private

insurance company. This would have large implications in establishing if the current

treatment funds directed to CBT represent a good investment.

Application of Machine Learning to Vision-Based Pose Data for Exercise Classification

The research will be using visual information from the phone’s camera as well as demographic information from participants and implement various machine learning algorithms such as random forests, support vector machines, etc. to provide feedback regarding different exercises to the participant. Specifically, the algorithms will classify the exercise types. Furthermore, these algorithms will be optimized for use on smart phones. The partner organization intends to incorporate the algorithms in their mobile app for mass use. Such research methods allow for a more health-conscious use of smart phones and would give the partner organization a significant edge in technological development in the health-related sector.

Predicting Treatment Sensitivity in Hypotensive Patients

Anticoagulation with Warfarin is indicated and required for post-operative cardiovascular patients. However, it is a high-risk medication with a narrow therapeutic range where sub-optimal dosing can lead to complications and even death. While multiple risk factors have been associated to Warfarin sensitivity, the prediction of optimal Warfarin dosing strategies remains ineffective and requires trial and error and close patient monitoring. This work proposes the use of machine learning and reinforcement learning algorithms to more accurately predict Warfarin requirements in post-operative cardiovascular patients, leading to decreased hospital stay and re-admission rates and increasing cost savings at cardiovascular surgery centers globally.

Multi-institute domain adaptation by adversarial constrained medical time series representation learning

Hospitals strive to perform cutting edge medical treatment, treat all patients fairly, and reduce operating costs, while also enabling caregivers to spend more time interacting with patients. Artificial intelligence and machine learning promise these things. However, medical data provides unique challenges for machine learning. Currently, if a hospital wants to include an algorithm for automated decision making, they must either secure approval to collect additional patient data or change their care practices to replicate those at other institutions. This work proposes a novel application of artificial intelligence in medicine that creates a numeric representation of patients’ electronic medical records which is constrained to be similar across all hospitals despite each hospital having different underlying operating procedures. As a result, we can directly transfer algorithms which have proven to improve care at one hospital to another, without the need for additional data collection. This research has the potential to save lives of patients who otherwise might have been overlooked, improve patient quality of life, and set a precedent for quality healthcare globally within the next three years.

Machine learning in the operating room: focus, performance, and the medical record

This proposed study will significantly enhance our current understanding of how specific intra-operative factors can impact patient outcomes. Our proposed work will provide a proof of concept that machine learning can objectively predict a specific, high-impact post-operative complication, allowing us to move forward with scaling this work to a wide variety of surgical settings. Moreover, the use of machine learning to automatically assess high-risk points of a surgery has many implications, including the ability to direct risk mitigation efforts, work towards real-time assessment of operations, as well as standard-setting and credentialing for surgical procedures. The ability to automate the evaluation of a high-risk step of an operation in real time, and potentially change a patient’s outcome, undoubtably has the potential to significantly improve patient safety on a large scale.

The digital e-commerce logistics and supply chains management using block chain technologies

E-commerce has become one major marketing channel for many firms in Canada and world-wide and has increased dramatically in recent years. As firms migrate from traditional physical retail channels to combined physical and virtual channels, the shift brings new significant challenges to supply chain and logistics management. Blockchain is able to maintain authoritative records in a fully decentralized, secure, and trustless manner and far-reaching implications in supply chain management. Blockchain technology will revolutionize several aspects of supply chains; financial and non-financial. However, it will also face security, legal, regulatory and technological challenges. In this project we will study how to apply block chain technologies to e-commerce supply chain and logistics to improve traceability, security, and coordination.

Développement d’un logiciel d’analyse modale opérationnelle

L’Institut de recherche d’Hydro Québec (IREQ) est le leader en Amérique du nord sur des

recherches et développement en énergie. En collaboration avec L’École polytechnique de

Montréal et l’’École de technologie supérieure (ÉTS), un projet de recherche a déjà été

réalisé sur la vibration des turbines hydrauliques. À la suite du succès du projet, L’IREQ

souhaite développer une technique d’analyse modale opérationnelle sur des machines et

structures afin de faire le suivi modal de ses machines en assurant un fonctionnement

d’opération optimal

Apelin Analog Therapy as an Innovative Treatment for Cardiovascular Diseases

Apelin, an innate peptide, is a critical component of the apelin pathway, which is responsible for regulatory mechanisms of the cardiovascular system. Apelin is downregulated in patients with cardiovascular disease, therefore limiting the cardioprotective potential of the pathway. This project focuses on the optimization of a biological analog, able to withstand enzyme degradation with improved function that acts as a substitute for apelin. The intern involved in this project will have the opportunity to be a part of the development of a conceptual pharmaceutical into a biologic that can be used in clinical practice for patient treatment. Through the completion of this internship, PEARKO Therapeutics will have the assistance it needs to fully optimize its preclinical apelin analog formulations, bringing it a step closer to use in clinical practice.

Modélisation inverse de la charge de chauffage de l’eau d’un ménage

Les mesures d’équipements intelligents sont de plus en plus disponibles. Ces mesures permettent de caractériser le lien entre les habitudes des ménages et la consommation d’énergie du chauffage de l’eau. La meilleure connaissance de ce lien contribue au développement d’outils permettant d’anticiper la consommation d’énergie. Un réalisme accru de prévision permettrait à Hydro-Québec d’entrevoir d’autres méthodes pour gérer son réseau, évaluer ses programmes ou planifier ses activités. À terme, ceci permet d’augmenter la fiabilité de sa fourniture électrique tout en diminuant ses coûts de maintenances des infrastructures minimisant ainsi les impacts sur la tarification des clients.

Mechanical study and failure analysis on steel spirally tungsten inert gas (TIG) welded tubes

Spiral tubes and rings have been broadly used in the manufacturing sector, especially for the automobile industry. However, the failure of the rings will significantly shorten the service life of components, even leading to a disaster when the part is under operation. Generally, these tubes and rings are made from steel sheets spirally using the tungsten inert gas (TIG) welding method. Through the university-industry collaboration under the Mitacs program, the University of Toronto (U of T) team will give a fundamental understanding of failure mechanism of the (TIG) welded tube steels, and the partner company will improve their product quality and process stability by acquiring the study results from U of T on Process assessment, inspection method, and weld quality risk assessment.
This collaborative research provides theoretical knowledge in the failure mode from the TIG welding process and suggests a welding parameter window for the partner company to achieve a high-performance weld seam.

Experimental and numerical modeling of hydraulic fracturing in tight and light formations

Hydraulic fracturing technique has become an essential requirement in exploiting tight and light oil reserves in Saskatchewan and Alberta. The recovery in tight oil fields in these provinces is low with rapid production decline in hydraulically fractured wells. Also, it should be noted that fracturing horizontal wells can consume millions of gallons of water; a small portion of which is recovered after putting the wells on production. Considering the economic and environmental impacts of failed fractured wells, the need for research and development in the area of optimized fracturing design and fracture containment becomes evident. This research project aims at studying fracture initiation and propagation in tight formations as a function of various field and operational parameters. Results of this work will help in better design of fracturing jobs specific to Saskatchewan and neighboring provinces; with focus on increasing economic performance and reduction of environmental footprint of these stimulation operations.

How severe weather affects the auto insurance industry

In Canada, motorists are faced with a wide range of environmental conditions. Weather, in

the form of rain, snow, other frozen precipitation, fog or strong winds, occurs 10 to 20 percent

of the time, depending on the location and year. This project, which involves a three-way

partnership between State Farm Insurance, the Institute for Catastrophic Loss Reduction and

the University of Waterloo, explores weather-related driving risks in the Province of Ontario,

Canada, in order to better understand how risk is increased during inclement conditions. The

analysis is unique from previous studies in its focus on major storms, on urban versus rural

experiences of weather hazards, and in its exploration of accident characteristics and how

they may differ during inclement conditions versus clear, dry conditions.