Innovative Projects Realized

Econochef – Implementing MLOps for Small Businesses

This project will look into the tools available to implement machine learning in production and find the best ones to use in a small business environment. The goal is to find a set of tools that can be used by developers with little or no experience with machine learning that is still performant enough to add value to a software product. We will implement our solution in Econochef’s mobile app that will recommend recipes to users by training a model according to a user’s culinary preferences. This feature will allow Econochef to set itself apart from the competition.

Social Enterprise and the Pursuit of Net Zero in Newfoundland and Labrador

Memorial University’s Centre for Social Enterprise presents this type of organization in the following light:

“Social enterprise challenges the predominant understanding of business as pursuing wealth at the expense of people and the planet. A traditional view of business measures a healthy economy by wealth gained through the trade of goods and services. Social enterprises, however, produce goods and services for trade but direct their surpluses towards social goals. They offer inspired, tangible solutions that elevate vulnerable populations, mitigate environmental impact and connect individuals with communities. Social enterprises view community prosperity that balances economic and social development as the most important indicator of a society’s success. They understand that communities are strongest when their individual members are too, when social and financial barriers are lifted and opportunities are generated for all. This economy of care values performance indicators like social well-being, environmental health and cultural diversity to measure quality of life.”

Livestock Monitoring and Analysis: A Deep Learning Approach on Edge Devices

Livestock monitoring is a topic of great interest for livestock farming since it would help identify the most appropriate feed cycle, improve and predict animal health, thus reducing investment cost, and providing other important knowledge to improve the breeding process. This research project proposes the development of a non-invasive monitoring system for livestock using video sequences of groups of pigs as experimental subjects. The system will be able to identify the different pigs, classify the different activities that they perform, and analyze different wounds on the skin to locate the most probable cause of them. The proposed system will be based on artificial vision through the use of a camera with deep learning architectures and machine learning techniques, which will be located in strategic places to monitor the behavior of the animals and their environment.

Developing approaches for the production and processing of single-cell protein

The development of a cheap, environmentally sustainable, protein-rich animal feed is of key interest to the growing livestock and aquaculture sectors. Cvictus is in the process of commercializing an integrated low-carbon footprint production platform that can convert coal to methanol, which can then be used to grow a single-cell protein (SCP) animal feed that would address the major concerns of meat and dairy producers. This project will focus on examining the impacts of changing various process conditions to establish approaches for SCP production and processing that result in the highest yields and/or nutritional quality. These studies will generate invaluable data that will help direct the build of Cvictus’ integrated production platform and consequently facilitate sustainable utilization of Alberta’s vast coal resources.

SEE Mitacs Internship

Through research and iterative testing, an intern will develop resources that will guide Social Entrepreneurs through a process of building branding and communications strategies for their ventures. The student will research industry best practices and then apply their gatherings by creating a curriculum that entrepreneurs can work through at their own pace. The student will then refine their materials by guiding one or two of SEE’s entrepreneurs through their curriculum and adjusting the materials based on entrepreneur feedback. The project deliverables will be a set of resources that represent a repeatable process for development of communication strategy for social entrepreneurs that SEE can provide to all their entrepreneurs in the future.

Image Deblurring for Mobile Devices Year Two

The wide proliferation of digital cameras has resulted in a massive flow of images captured then shared online in various forms. At the same time, capturing high-quality images with embedded cameras remains a significant challenge. One of the most common image degrading effects is motion blur, produced by camera movement while taking a photograph. This problem is widespread and has attracted considerable research and development efforts, but there is still no satisfying solution in place suitable for a wide variety of conditions. In this research project, we will consider the problem of image deblurring for mobile devices. We will consider existing algorithms that estimate the camera movement and remove motion blur, and we will develop new ones. For the industrial partner, this project will result in an enhancement of its product portfolio for the mobile device industry, video surveillance and visual data processing.

Software Quality assurance internship supporting R&D in the development of a commercial stroke decision support tool

Acute stroke is a leading cause of global death and disability. Quick access to robust treatments significantly improves patient outcomes, however, accurate interpretation of diagnostic imaging and effective communication of suspected findings remains a serious challenge for all stroke networks. Circle Neurovascular Imaging (Circle NVI), a wholly owned subsidiary of Circle Cardiovascular Imaging (Circle CVI) and a joint venture with the Calgary Stroke Program, is building a software as a service business aimed at providing physicians with an acute stroke decision aid tool, StrokeSENS™ that will leverage artificial intelligence (AI), advanced image-processing, and network connectivity to support physicians and other healthcare team members to simplify stroke care, with a focus on enabling front line stroke centers. Circle NVI’s mission is to simplify stroke care by augmenting front-line physician decision making.
Developing software improvements and new software features for our StrokeSENS software, in addition to building solutions for business-to-business partners with one small team is challenging. Funding through this Mitacs BSI internship will enable us to have an additional Quality Assurance (QA) team member working with our Research & Development and platform development teams.

Intelligent 5G/B5G Service Function Chain Provisioning

This project aims to solve a primary problem of 5G/B5G Networks, namely the Service Function Chain (SFC)
provisioning problem. This project will improve the solutions for this problem in two aspects. First, the SFC architectures considered in this project are futuristic/generalized ones that do not appear in existing systems. These architectures reduce the processing time of services by concurrently performing tasks at different locations. These sophisticated systems represent a complicated SFC provisioning problem that has received very little attention in the literature. Second, this project will improve existing provisioning solutions by robust traffic prediction frameworks. Most of the existing traffic prediction frameworks consider only one location at a time and rely on historical data strongly. Data traffic patterns at different locations correlate mutually. And historical data is not usually available because of stringent privacy agreements. This project will apply the latest Artificial Intelligence and Machine Learning techniques, especially online learning, to design robust solutions and evaluations. For Ciena – a leading organization in the networking service industry, this research project adheres to Ciena with innovative technologies of 5G/B5G Networks. Ciena may get prototypes from this project that lead to products in the future. The research outcomes can also confirm the effectiveness

Development of fabrication process for AFM probes

Atomic Force Microscopy (AFM) is one of the most widely used imaging tools by the researchers and industries
to view and characterize tiny objects. This tool uses a very tiny and sharp probe to scan or feel the surface of the
specimens and then it produces a topography image of the scanned surfaces. This research project aims to
develop the cleanroom fabrication process for “soft” probes with low force constant as needed for biological
imaging, durable probes with long lifetime using diamond as the tip material, and high aspect ratio silicon tips for
accurately imaging deep trenches or tall structures.

Powered by ARC Brand Opportunity Platform

We are looking for an intern to help add a new and exciting component to our platform which will connect our athletes/streamers with brand opportunities and be a core growth driver for the ARC Profile product. The purpose of this component is to act as an interface for a brand to post multiple opportunities such as sponsorships and scholarships and give them the ability to interact with applicants seamlessly. The design of the platform will facilitate these connections by pushing appropriate candidates for the available opportunities. For example, if Golf Town was looking for a match, the program would suggest qualified golfers to them, making the connection process more efficient The platform will also offer performance-based rewards in the form of trophies or product giveaways that brands can link to their products to. Another key component of the platform will be direct transfer of funds from the brand to the user and contact information for shipping of product or equipment. While all these features are certainly beneficial to brands, this will also help athletes and streamers to build their careers and portfolios. The platform will act as an interface for players to easily view and apply to opportunities and even have options pushed directly to them if they are a match for the brand.

GenGuard Software Platform Reliability and Maintainability Enhancement

Modernizing plant breeding programs through integrated genomic and phenomic data management and analysis is an important organizational innovation for many public and private plant breeding organizations. This project involves enhancing the reliability and maintainability of the GenGuard software platform to support marker-assisted breeding with the partner organization. The objectives of the project include improving the website reliability, the data integrity, and the software maintainability of this system. The outcomes of the project will enhance operational efficiency and data security for the partner organization.

Image-based phenotyping to accelerate analysis of breeding field trials

Plant breeding companies develop new seed varieties by growing thousands of candidate varieties in large field trials and manually selecting the best-performing varieties. Variety selection is informed by genomic data, but plant populations still need to be grown in field conditions and screened for specific physical characteristics, called phenotypes, such as yield, disease resistance, harvestability and other important traits. In-field measurements of these phenotypes are expensive, tedious and time consuming, and much of the collected data is not directly used by breeders. Digital phenotyping has the potential to efficiently provide critical information to plant breeders by analyzing aerial images of field experiments collected by unmanned aerial vehicles (UAVs). However, the process of transforming UAV images into actionable information for plant breeding companies remains a significant bottleneck in this important process. To capitalize on the potential efficiency and genetic gain enabled by UAV based imaging of field trials, we propose to work with Nuseed Canada to integrate and refine our PlotVision software into Nuseed Canada processes for their crop breeding field trials.