Innovative Projects Realized

AIDOX – Document Verification System

To validate structured trade contracts for language and economic term correctness, the existing document understanding systems use machine learning methods, natural language understanding, and text analysis, to extract data elements from financial documents. This can be expanded to a wide variety of financial documents, especially customer-provided reference material. The project focuses on document information extraction to extract the key data elements from financial documents and validate this data against the internal source of record.
The existing OCR-based approach has issues of error propagation (especially with noises), slow processing time due to the large model size, as well as the need for retraining for new document classes. To improve this, the first approach is to improve OCR engine performance by exploring OCR engines with pre-processing and post-processing techniques. The second objective is to implement OCR-less models which learn jointly with image and text contexts.

COMmunity-based nutrition RISK in Indigenous communities (COMRISK-Indigenous)

Poor nutrition is common in older adults and may be even higher in Indigenous communities than the general population. Yet, nutrition risk is rarely assessed as part of a health check-up. In this project, a First Nation and an interdisciplinary team will work in partnership to determine how to address poor nutrition in that community. The Mitacs intern will collate, interpret and communicate outcomes in order to generate a pathway for moving forward.

Readying For Change: Supporting Sustainability Progress Through Inner Transformation

There is increasing urgency to accelerate sustainability progress and address growing social and environmental challenges. Current approaches to solving these complex issues, such as new technology and governance, are failing to support the kinds of changes needed to collective wellbeing. Interest is growing around how individual values, worldviews, beliefs, and consciousness can be transformed so that they align with sustainability requirements. The project examines: i) what kinds of skills and capacities would be most helpful for enabling positive change, ii) how these approaches could challenge old ways of thinking and doing that are detrimental to people and planet, and iii) how new skills could be nurtured to support present needs. This is the first project to develop and test a novel readiness assessment tool and therefore offers immense value to the fields of sustainability and social innovation.

Decolonizing diagnostic assessments in mathematics

The overall purpose of this research is to explore the ways diagnostic assessments enable or disable growth in mathematics understanding for children most impacted by colonialism, in particular Black and Mi’kmaw children in Mi’kma’ki or what we now call Nova Scotia. This research stems from ongoing
collaborative relationships in local Mi’kmaw and Black communities that have focused on strategies for decolonizing pedagogy and content in mathematics classrooms. Yet, we recognize the disabling and dehumanizing power of current diagnostic assessment practices that contribute to the opportunity gap (otherwise referred to as an achievement gap) for students in mathematics achievement, resulting in
higher rates of individualized program plans (IPPs), streaming into nonacademic courses and lowered
expectations for Black and Mi’kmaw children. Much has been written about how these mathematics assessments are not culturally relevant or enabling, yet there is limited literature on how to create more appropriate diagnostic tools. Diagnostic assessments should be useful for educators, students, and their families to understand the underlying reasons for mathematics difficulties and create plans to support student learning….

Examining a Nature-Based Mindfulness-Compassion Intervention Delivered Via Virtual Reality: Applications for Older Adults

The global population is aging at an unprecedented rate, increasing the necessity for effective interventions targeting the mental health needs of older adults. The proposed research project is a partnership between Dr. Khoury’s mindfulness research lab at McGill (PhD Intern: Isabel Sadowski) and Super Splendide. We seek to investigate the feasibility, acceptability, and preliminary effectiveness of a 4-week nature meditation program offered via a multi-user 360°-HD video virtual reality (VR) application, called Toujours Dimanche. This program aims to support older adult mental health through incorporating elements of nature, mindfulness, and compassion. If successful, Toujours Dimanche has the potential to promote geriatric mental health through the use of a technology that is accessible, as well as safe. This project will also support Super Splendide in the effect development, market validation, and scaling of Toujours Dimanche.

Un nouveau regard sur les protéines et le microbiote intestinal afin de mieux comprendre les bienfaits des produits laitiers sur la santé.

La composition en nutriments a été traditionnellement considérée comme un point clé pour justifier et recommander la consommation de certains aliments. En ce qui concerne les produits laitiers, les guides alimentaires de nombreux pays ont principalement insisté sur le lien calcium-lait selon lequel un nombre minimal de portions quotidiennes de lait et produits connexes devrait être consommé pour maintenir une santé osseuse adéquate. Au-delà de ce constat, la recherche sur les produits laitiers au cours des dernières décennies a également documenté une relation bénéfique entre la consommation de produits laitiers et le risque de développer des problèmes de santé tels que le diabète de type 2 et l’obésité. Dans un tel contexte, le calcium s’est de nouveau avéré être un acteur important bien qu’il ait également été établi que les protéines sont importantes. Cette demande propose donc une continuation de cette preuve de concept qui devrait montrer qu’au-delà de la quantité, les protéines laitières peuvent faire la différence dans le contexte de la prise en charge clinique de l’obésité.

Les effets de la pratique d’activités culturelles parascolaires à l’école secondaire sur le développement personnel et social des jeunes

En mars 2022, ACLAM, anciennement la corporation Secondaire en spectacle, a adopté une nouvelle raison d’être : le développement global et du plein potentiel des élèves des écoles secondaires par la pratique d’activités culturelles parascolaires. L’organisme doit toutefois être en mesure de faire valoir l’importance de ces activités pour l’élève. Le projet de recherche consiste à rassembler un maximum de connaissances à propos des effets de la participation des adolescents aux activités parascolaires du secteur des arts et de la culture. Les effets néfastes de la participation doivent être connus et compris si on veut mettre en place des mécanismes pour les réduire ou les éliminer. Les effets positifs, eux, doivent être valorisés, maximisés et mis de l’avant.

Influence of redox conditions on the fate of emerging contaminants in groundwater environments

Contaminants of emerging concern (CEC) have negative impacts on human and ecosystem health as well as the potential to develop antimicrobial resistance which is projected to result in 10 million deaths annually by 2050. Some examples of CECs include antibacterial agents, artificial sweeteners, and pharmaceuticals. Recently, the presence of these CECs has been found widespread throughout the environment including in our fresh waters. This research will focus on gathering preliminary field measurements on CEC fate and transport in groundwater environments known to have strong redox gradients. This will be completed at field sites near Sydney, Australia that are known to have CECs in the groundwater including recharge sites where treated wastewater is injected into the ground. Solid phase and groundwater samples will be collected and analyzed for CEC concentrations, redox conditions, as well as some other traditional geochemical parameters. Multiple sampling events will be conducted to better understand spatial and temporal relationships between CEC concentrations, redox gradients, and some other parameters. This research will help to better predict the transport of these CECs in groundwater allowing for future remediation.

Abnormal Detection for Language Assessment

A typical language test usually consists of four parts: speaking, listening, reading, and writing. Both audio and text data from the test takers will be collected and used by the automated scoring system. During the test, some test takers will intentionally/unintentionally provide abnormal answers, which may contain memorized content, repeated sentences, and meaningless or off-topic words, and may affect the accuracy of the scoring system. Pearson would like to develop a more robust system for automated scoring to identify and penalize abnormalities in the audio and text data, which could greatly improve Pearson’s competitiveness in the market and in the meantime contribute to the development of abnormal detection models.

Statistical Analysis and Machine Learning Approach to Retail Sales Forecasting Based on Localized Weather Features

In recent years, weather has been recognized as an important factor that can have a significant impact on
consumer behavior in certain industries. Predictive models that incorporate weather data can help industries to
adjust their inventory and marketing strategies to optimize sales.
This research project focuses on using machine learning and data analytics to determine any weather-related
signals on retail sales based on historical data and then to predict sales using weather information. The project
will involve literature review, developing complex SQL queries, engineering features, analyzing weather and
retail sales databases, and developing machine learning algorithms. The ultimate goal of this end-to-end project
is to create an accurate and scalable sales forecasting model that can be used by industries to optimize their
inventory and marketing strategies based on weather patterns.

Deep-learning cross-modality super-resolution imaging of the cellular network in bone

Many organisms, such as bone, depend on fluid-filled network structures for cellular transport, signaling and mechanosensing. Bone is made of a network called the lacunocanalicular network (LCN), which consists of lacunae that house osteocyte cells connected together by a network of 300-nm wide canals called canaliculi. An understanding of this vast network is incredibly important and could provide valuable insight into bone function and disease. While many techniques exist to visualize LCN, the large size of the network makes acquiring images at sufficient resolution while covering a large field of view difficult.

The proposed research aims to use and develop deep learning image-to-image translation to improve the resolution of microscopic images of bone and enable graph network extraction of the LCN directly from low-resolution (LR) images. It is hypothesized that deep learning can be used to perform same-modality super-resolution (SR) of LR confocal microscope images, as well as cross-modality SR of LR confocal microscope images to high-resolution Focused Ion Beam-Scanning Electron Microscopy (FIB-SEM), and to directly extract large area graph networks more efficiently.

Analyse en temps réel de signes vitaux grâce à l’application de techniques d’apprentissage automatique et à la synchronisation d’une pluralité de senseurs physiologiques

La photopléthysmographie (PPG) est une méthode non invasive mesurant les changements de volume sanguin
dans les tissus. Cette technique permet de dériver des métriques physiologiques importantes, telles que la
fréquence cardiaque, la saturation en oxygène et la pression artérielle. Cependant, pour des résultats fiables, il
est crucial de prendre en compte les différences physiologiques des individus, notamment relié à la couleur de
peau, au sexe ainsi qu’à l’âge et de surmonter les problèmes d’artefacts de mouvement.
Le projet vise à améliorer la fiabilité des métriques PPG en utilisant l’IA et un nouveau capteur portable de SiFi
Laboratoires pour améliorer la précision de la mesure de la pression artérielle en télémédecine. L’intégration de
ces technologies de pointe permettra d’obtenir des mesures de pression artérielle plus précises, ce qui peut avoir
des implications significatives pour la surveillance à distance de la santé des patients, en particulier pour les
personnes atteintes d’hypertension artérielle.