Projets novateurs réalisés

A simple method for subcutaneous vein detection in self-cannulation mapping of home hemodialysis

Although home hemodialysis (HHD) benefits patients including improving their lives’ quality, self-cannulation is currently a limiting and challenging step in their treatment. In this study, we aim to solve this problem by developing an epidermal film/patch with a vein visualization capability. A detachable mobile needle-guide tool can be mounted on the epidermal patch which can help patients to accurately perform self-cannulation through their skin. To conveniently connect and disconnect the needle guide segment to the epidermal pad, employing a simple and cheap magnet at the bottom base clip of the needle guide segment, as well as incorporating magnetic nanoparticles as nanofiller into the patch top-layer will fulfill our target. At last, the patients’ satisfaction toward their self-cannulation in comparison with previous methods will be monitored through surveys and questionnaires.

Few-shot learning en détection d’objets appliqué à la robotique

Chez Teledyne DALSA, nous développons une plateforme qui permet à nos clients d’utiliser l’apprentissage profond dans le monde de la vision industrielle en facilitant le développement de solutions intelligentes qui s’intègrent à notre gamme de produits. Bien le succès de l’apprentissage profond repose fréquemment sur l’abondance de données étiquetées, il est souvent difficile pour nos clients de récolter ces données en pratique. Le projet se concentre sur l’application du « few-shot learning » pour la détection d’objets dans le cadre d’une application robotique. L’objectif est d’améliorer nos solutions actuelles pour l’apprentissage avec un très petit nombre d’exemples étiquetées. L’étudiant devra assimiler l’état de l’art en « few-shot learning » afin de développer une solution pour la détection d’objets en vision robotique. L’étudiant pourra aussi mettre en pratique la solution proposée en développant une application démonstrative de « pick and place » intelligente à l’aide d’un bras robotique.

Unpacking the role of environmental stress and disease in salmon declines.

This program will investigate the role disease plays in salmon declines and the risks posed by open-net salmon farming to wild Pacific salmon. First, our ongoing partnership with the Broughton Archipelago Transition Initiative (BATI) will monitor salmon-farm impacts, serving as a testing ground for monitoring technologies to support Indigenous self-determination. Second, we will expand our work on the west coast of Vancouver Island, casting focus on the impacts of BC’s last major salmon-farm cluster on imperiled populations of Chinook salmon. Third, we will refine strategies for rebuilding of wild-salmon stocks through understanding of cumulative-stressor effects on survival coastwide. By investigating how individual fish respond to various diseases and environmental stressors, we seek to uncover why early marine survival of these populations is so poor, which many biologists consider to be the main driver of the widespread declines in BC’s wild salmon over the past three decades.

Duesytech

Duesytech is a product of Audio Concepts Inc which provides tech matching services to general customers and techs along with an online store. The intent is to provide tech matching services to general customers and techs while enhancing the traditional online shopping experience. Traditionally this is done through the application of a recommender system. However, given the sparse nature of the curent data set the application of self-supervised recommendation is anticipated to yield better recommendations. By applying Machine Learning to match what customers’ needs for tech services with techs, Audio Concepts will be able to more Duesytech forward to provide customers and techs with more robust matching services as well as enhanced online shopping expericence.

Fine-Grain MPI Year Two

The demand for computing power has been rapidly growing over the last decade. The ability to efficiently utilize computing resources and improve the productivity of applications is necessary for the competitiveness of any industry and it will become more critical as the demand for computational resources increases. The computer hardware sector has seen rapid advances with the introduction of multicore and many-core processors and has introduced many challenges for the software development community to efficiently utilize the new architectures. My research is focused on developing systems and techniques that allow improved utilization of parallel machines. The Message Passing Interface (MPI) is extensively used for constructing parallel applications in distributed environments. Our work is related to development of systems for high performance computing through extending and enhancing the capabilities of the MPI middleware. We have developed a system called Fine-Grain MPI that allows execution of hundreds and thousands of MPI processes on-chip or communicating between chips inside a cluster. The potential benefits of our approach are for multicore as the number of cores increase.

Westland Insurance Group – Communications, New Media, and Technology Internship Project

Westland Insurance has grown into one of Canada’s largest independently owned insurance distribution businesses. Our head office is located in Surrey, British Columbia, and we have over 190 locations across Canada. This internship project will involve identifying the best local and national practices around marketing, communications, media, employee learning management, digital content creation and technical systems management to optimize the organization’s brand awareness. The interns will search for best practices, identify gaps, recommend changes or enhancements, and complete a review of Westland’s existing practices and programs in a few different departments, focusing on communications, new media, and information technology. This project will impact various areas of our organization, supporting our teams in the above-mentioned departments to implement a number of most needed changes and improvements.

Exploring the agronomic and soil health benefits of liming in the Prairies

Soil acidity in Alberta and northeastern British Columbia has been recognized as a problem since the early 1960’s. With the current application of N fertilizers, this problem has persisted making it unsustainable to maintain high crop yields in the long term. This project hypothesizes that the one-time application of lime to increase soil pH will lead to increased crop yields and promote soil health. On-farm research will be undertaken to demonstrate directly to producers how to employ lime as a management tool and its benefits to farm income. This project will benefit producers, crop commissions as high yield implies increased farm income and dues and the Ag industry at large.

BSI Internship – Dylan Barry

The project will consist of the intern utilizing the knowledge acquired from their academic institute and applying it to the partner organization’s development team to assist in getting the company’s product ready for commercialization. The overall objective of this project is to leverage machine learning and Natural Language
Processing (NLP) to create an engaging learning environment for financial intelligence.
Primary project objectives include:
1. Bibliographical Review
2. Develop user stories framework and design conversation flow
3. Build and configure backend infrastructure for Intelligent Bot
4. Train natural language processor to understand user intent
5. Incorporate innovations into the user interface

Dynamic Appointment Scheduling for Radiation Therapy

Radiation treatment (RT) is one of the primary methods for treating cancer. The Radiation Medicine Program (RMP) in Princess Margaret Cancer Centre (PMCC) is Canada’s largest of its kind. It delivers around 11,000 courses of radiation treatment for over 8000 patients annually. RT uses Linear Accelerators (LINAC) to deliver high-energy rays to destroy or shrink cancer cells. With 16 LINAC units and uncertainty in future demand, scheduling appointments at RMP is a complex task. Given the increasing demand for RT, reducing waiting times through advanced approaches to scheduling is of high interest. In this project, we will develop new mathematical models and methods for dynamic scheduling of appointments at PMCC and evaluate their performance using both simulation experiments and in a pilot implementation at PMCC. The methods and tools developed through this project are expected to reduce patient waiting times, improve patient outcomes, and reduce costs.

EcoBloc Next Stage Operations & Growth Plan

EcoBloc sustainably improves regional food security and food safety while also reducing the impact of harmful chemicals on water, soil, and animals. Eco Bloc’s non-lethal technology protects assets in the global food supply chain, including animals, animal feed, crop inventory, seeds, equipment and other assets against biosecurity risk and asset destruction originating from mice and rats. This project will bring forward the product development stages by utilizing a software engineer and the organizational capacity in terms of partnerships and marketing through support interns.

Triware Technologies Edge-to-Cloud Program

Triware Technologies Inc. (Triware) is a leading Newfoundland-based information technology consulting and support firm, with public and private sector clients throughout the province. The company has extensive experience in business technology consulting and management with private sector clients, government organizations and public agencies. Our previous and current engagements include roles and responsibilities such as current, support provision, project management, web / application development and technology consulting.

As our clients move out of the pandemic and resume full business operations we recgonize an opportunity to help drive cloud adoption, allowing small-to-medium business and small-to-medium enterprises to become more agile and better address business intruptions moving forward. To achieve this Triware requires a focus on shifting our current and future clients from local networking and data management to a cloud-centered approach.

Grain Quality and Traceability Mobile App Development

Farms currently suffer from undervalued commodities, spoilage, and traceability issues. While Canadian growers can market their own grain, this places extra responsibilities on the grower such as monitoring the time, price, and place they wish to sell their grain. To determine the price of grain one must be able to determine the quality of the grain and the quantity. To grade the quality of the grain, samples must be taken to identify characteristics such as damage caused by insects, moisture, or temperature. Grain sampling can be a time-consuming process because traditionally a representative sample requires a person to take a sample from grain flowing from the auger at regular time intervals.

VeriGrain Sampling Inc. has produced a system to automatically (using the extractor) and semi-automatically (using only the mobile app) help perform grain sampling and data management. They want to build onto their mobile app to provide further functionality for growers, buyers, and consumers of the agriculture industry. This BSI project focuses on two items: 1) Improving the mobile application’s user interface and 2) integrating a grain analysis source into the mobile application.