Innovative Projects Realized

WCRI SimplySMART Project

Waterloo Co-operative Residence Inc (WCRI) is Canada’s largest student housing co-operative. It has been providing co-operative student housing and programming for post-secondary students in the Waterloo region for over fifty years. As a co-operative, WCRI relies on members’ involvement in providing affordable housing and creating a supportive and welcoming community.

Project SimplySMART focuses on documenting current WCRI processes, writing content for the website, conducting market research on e-commerce and ERP software vendors, identifying necessary winning conditions for digital transformation, devising execution strategy options, and preparing a strategy implementation plan.

Implementation of design controls through Quality Management System

Over 250,000 Canadians suffer from epilepsy, a debilitating neurological condition with no known cure. For many patients, the only solution is the surgical removal of the seizure-trigger zones in the brain. Unfortunately, current electrode systems that are designed to record brain activity and target the seizure foci are highly invasive, not very precise, and do not provide adequate coverage of the brain to sufficiently isolate the target sites. This limits the standard of care for patients struggling with epilepsy, as healthy tissue may get extracted as the surgeon compensates for the lack of high-resolution mapping. To help address these gaps, Neuraura Biotech Inc. has developed novel electrode technology that provides 3 times higher resolution and 20 times more coverage of the brain. To facilitate the first clinical trials of this recording and 3D brain activity visualization system, quality control systems must be expanded and regulatory approvals must be obtained. The successful translation of this Canadian technology to serve both researchers and patients will require strong adherence to regulatory requirements and market analysis for commercialization. This quality management and strategy internship will equip the student with the skills needed to commercialize products, secure partnerships, attain regulatory approvals, and work in a multidisciplinary start-up environment, which will be key in their future aspirations in serving patients. The intern’s contributions to risk analysis, project management, and regulatory approval applications will help accelerate the organization’s commercialization timeline and help us serve both researchers and patients with higher-quality devices for the treatment of neurological disorders.

Pre-training large models that generalize well across the chemical space

The intern is expected to develop a pipeline for training large machine learning models on drug-like molecules to learn a broad understanding of biochemistry. This pipeline involves developing a model that can learn on many biological and quantum mechanical tasks concurrently while evaluating and improving the structure of the model using low-cost approaches from recent literature. Then, he will implement transfer learning to evaluate the model’s performance in predicting the properties of unseen molecules for unseen tasks. This last part is very relevant for the partner organization since it is expected to improve their ability to predict molecular properties on real-world tasks and speed up the discovery of novel drugs.

Product Innovation and Process Improvement for SR.ai’s Investment Research Platform

The field of responsible investing is rapidly expanding, with even greater attention on the importance of responsible investment in 2021, as seen in the aftermath of the hugely impactful COP26 summit. Directing our financial resources in a sustainable direction has the potential to have a massive impact on helping us meet the Sustainable Development Goals set forth by the UN. Though the importance of better alignment between finance and sustainability is clear, and now has a very strong consensus around it, investors still lack the right tools to support their research process in terms of sustainability. They rely on high level data that is largely known not to be reliable, which leads to massive misallocations of capital. The mission of SR.ai is to bring more rigour to the world of responsible investment, using our technology backed by peer-reviewed research in sustainable finance, produced through collaboration with the University of Toronto RiskLab quantitative finance research center, led by our CEO Alik and his PhD Supervisor Luis Seco.

The mission of SR.ai is to help the responsible investment industry in directing global financial resources towards sustainable and socially responsible companies and initiatives. The objective of the collaboration is to improve SR.ai’s product, assist in our product management and business planning process, and identify “white space” opportunities for SR.ai’s AI technologies.

Toronto Seed Strategy – Pilot Orchard Documentation

In the spring and summer of 2022, the Toronto Seed Strategy Working Group (TSSWG – a group of landscaping business owners, commercial plant growers, and plant researchers, convened by Jonas Spring of Ecoman Landscaping and Gardening) is initiating a Pilot Seed Orchard where locally-sourced, native plants will be grown to produce seeds for native plant production. This Seed Orchard (hosted by NVK Connon Nurseries in Dundas, ON) addresses a surging demand for locally-sourced, native plants in the Greater Toronto Area, which currently lacks sufficient supply. Landscaping companies are seeing a demand for locally-sourced native plant gardens, but they can’t find appropriate or sufficient quantities of plants. Within Canada’s commitment to mitigate climate change, urban greening projects, ecological restoration, and gardening with native plants are important strategies to reduce water and fuel consumption, increase biodiversity, and provide ecosystem services – and they require significant quantities of locally-sourced native plants. As a pilot project, the TSSWG Seed Orchard will be an important model for similar organizations as cities and regions mobilize to rapidly address ecological restoration needs in the face of climate change. The Intern’s research position will focus on the 1) documentation and description of the Seed Orchard operations, through diagramming and writing, 2) documentation of seedling growth and harvest activities, using photography, and 3) compiling of all documentation into a brief report and presentation that can be easily used by all members of TSSWG, and shared with other regions working on similar native plant seed strategies.

Low latency, robust and reliable multi-party interactive live video streaming

A multi-party live video communication, such as live tutorials and fitness classes, are an emerging application which involves a large number of users from different places with heterogeneous network conditions like 3G/4G/5G or Wi-Fi networks. Video Content/Service providers usually deploy their Content Delivery Networks (CDNs) over the public Internet to avoid expenses of dedicated connectivity. Thus, they often seek solutions to provide seamless services over the changing conditions of Internet that can introduce packet error, packet loss, or out-of-order packets. However, existing solutions neither scale well with the number of users nor adapt to diverse network conditions of users that require the same video to be streamed in different bitrates depending on user network condition. In this project, we will develop error detection and correction with retransmissions techniques for low-latency multi-party live video streaming. Moreover, we will devise a prediction-driven proactive CDN server scaling and placement algorithm in which CDN servers are spawned/teared down dynamically according to users’ demand change.

Refining of Kisolite Clay for Cosmetic and Phamaceutical Applications

There are four main objectives for the proposed project:

(1) To investigate three techniques of aiding the wet screening process including: (a) inclined screening combined with water jets, (b) sonication, and (c) a combination of those two techniques, to find optimum operating conditions for each technique and allow them to be evaluated and compared.

(2) To implement the most suitable technique on the industrial scale and establish its optimum operating conditions.

(3) To design and integrate all unit operations for producing the refined clay and optimize the overall production process.

(4) To refine the Kisolite clay to a suitable size range for inclusion into nanofibre structures (if time allows).

AI & Sustainable Finance: using NLP to analyze ESG disclosures of Canadian companies

Sustainable finance is a rapidly evolving discipline, where investors are in the process of acquiring and developing new tools allowing them to evaluate the sustainable impact of the companies they choose to invest in. Public companies have adopted the practice of publicly disclosing ESG-related information under the form of public reports following ESG disclosure frameworks and standards. Analyzing these reports represents a challenge for investors because of the vast quantity of reports and information they contain, the structure of the information, and the difficulty of quantifying and scoring the information. Investors often rely therefore on the ratings offered by rating agencies, but many investors would like to better equip themselves to do that work for more transparency on their evaluations. Part of the AI and Sustainable Finance currently explores the opportunity for a Natural Language Processing tool capable of collecting online reports published by Canadian companies, extracting, and analyzing said information.

Caractérisation d’espèces fongiques du Canada et du Rwanda

Les champignons produisent une panoplie de molécules qui ont des utilisations multiples dans plusieurs domaines tel que dans l’industrie agroalimentaire, les applications environnementales via la biodégradation de déchets et rejets industriels, et la médecine. Des composés issus de plusieurs champignons ont démontré une capacité à inhiber des virus et des bactéries, ralentir des maladies neurologiques et neutraliser la prolifération de certaines cellules cancéreuses. Récemment, un intérêt marqué s’est manifesté envers les molécules produites par les champignons psychotropes: Les résultats d’essais cliniques et recherche neurologiques ont démontré que les extraits de ces champignons sont capables de significativement améliorer les cas de dépression récalcitrante, d’anxiété débilitante, de syndrome de stress post-traumatique, spécialement chez les vétérans de guerre, et à assister à éliminer la dépendance aux substances, notamment les opiacés. Ce projet vise donc dans un premier temps de caractériser avec précision les champignons issus des inventaires de Mycocultures inc., et du Cercle des Mycologues de Montréal à travers une analyse moléculaire de leur ADN.

Security and safety intelligent systems on smart spaces: privacy perseveration approach

We are witnessing the rapid development of the Internet of Things (IoT) which provides ubiquitous sensing and computing capabilities to connect a broad range of things to the Internet while connecting people and spaces. That opens a wide range of opportunities to foster user-centered service, improve safety, and well-being. Intelligent IoT-based solutions have been shown to have significant potential in people safety and risk environment monitoring, due to their ability to operate at a fine granular level and provide rich low-level information.
To obtain mean-full insights into data generated from ubiquitous IoT devices in smart spaces, artificial intelligence (AI) techniques have been widely exploited to train data models for enabling intelligent IoT applications. Traditionally, data is sensed and communicated to data analytics and AI functions which are placed in a cloud server or a data center for data learning and modeling. This incurs critical limitations, such as the offloading of massive IoT data to the remote servers may be infeasible due to the required network resources and the incurred latency. The use of third-party servers for AI training also raises privacy concerns such as data breaches as the training data may contain sensitive information and also open extended security challenges.

Testing whether southern resident killer whales are endangered because of the collapse of winter and spring Chinook populations in California that occurred over a century ago

We will test the hypothesis that southern resident killer whales have been endangered since the late 1800s when mining, logging, dams, pollution, and water diversion destroyed the Chinook populations that returned in the millions to Oregon and California. We will estimate historic run sizes of Chinook salmon available to northern and southern resident killer whales based on catch records and the numbers of cans of Chinook processed at canneries (from California to BC during the 1800s). We will also compare historical seasonal abundances with relative numbers of Chinook returning to spawning rivers today—and will compare seasonal numbers of Chinook needed to sustain resident killer whales by locations to identify when and where the nutritional bottleneck southern resident killer whales are experiencing is occurring—and whether fishing restrictions imposed in BC are ultimately benefitting southern resident killer whales or not.

Multimodal Representation Learning for Healthcare Data

The abundance of electronic health record (EHR) data has accelerated the adoption of data-driven methods to automate various tasks ranging from patient care to resource management in hospitals around the world. The use of specific types of data such as X-ray images, doctor’s note and others have been used individually to implement machine learning models suited for a specific task. However, such single mode of data is not able to provide an overall picture of the patient health. In this project, we aim to develop an end-to-end machine learning model that can learn a combined representation of all these different types of data which can be fine-tuned for multiple downstream tasks such as treatment outcome prediction, hospitalization and length of stay prediction and many more.