Innovative Projects Realized

Technical Business interns working within cross-functional teams to commercialize AI-powered solutions in the Public Services Sector (1)

“THIS IS A GENERIC TEXT PUT IN PLACE AS THERE WAS NO PROJECT OVERVIEW”

Blockchain Smart Contract Vulnerability Detection Using Quantum Convolution Neural Network

The proposed project aims to develop a Quantum Convolutional Neural Network (QCNN)-based approach to detect vulnerabilities in smart contracts, which are critical components of blockchain technology. By leveraging quantum machine learning techniques, the project seeks to enhance the accuracy and efficiency of identifying security threats in smart contracts, such as reentrancy attacks and integer overflows. This research will involve collecting and preprocessing a dataset of smart contracts, designing and implementing a QCNN model using frameworks like TensorFlow Quantum and Qiskit, and benchmarking its performance against traditional deep learning methods. The expected benefit to the participating institutions includes advancing their expertise in quantum computing and blockchain security, fostering collaboration between researchers, and contributing to the development of more secure and reliable blockchain ecosystems. This project will also provide valuable training opportunities for interns, equipping them with cutting-edge skills in quantum machine learning and cybersecurity.

Caractérisation des fibres de bois recyclées

Il est possible de recycler les panneaux de fibres à densité moyenne (MDF) par un procédé hydrothermique, décomposant la résine thermodurcissable urée-formaldéhyde liant les fibres de bois le constituant. Bien que ce procédé soit de plus en plus connu, nous disposons de peu d’informations sur les propriétés chimiques des fibres récupérées, ce qui a pourtant un impact crucial sur les différentes applications possibles de ces fibres récupérées. Ce projet a pour but d’étudier les propriétés chimiques, principalement la proportion des constituants (composants lignocellulosiques et résine résiduelle) des fibres de bois récupérés par procédé hydrothermique et d’évaluer comment l’utilisation d’acide faible et la température influenceront la proportion des différents constituants des fibres. À terme, ce projet propose de répondre à certaines problématiques que rencontre le Canada, comme le recyclage de panneaux de bois MDF et la valorisation d’acides faibles industriels.

Exploring Quantum Computing for Public Transit Origin-Destination Matrix Estimation

This research explores the integration of quantum computing with statistical methods to enhance public transit origin–destination (OD) matrix estimation. Traditional OD estimation relies on automated fare collection (AFC) and automatic passenger counting (APC) data, which often present challenges due to incomplete coverage. Scaling techniques like iterative proportional fitting (IPF) help address these gaps, but their accuracy declines at low AFC penetration rates.
This study proposes leveraging a hierarchical Bayesian framework alongside quantum algorithms, Quadratic Unconstrained Binary Optimization (QUBO) for combinatorial optimization and the Harrow-Hassidim-Lloyd (HHL) algorithm for solving linear systems, to improve OD matrix accuracy. These methods will incorporate APC-derived alighting probabilities and AFC trip-chaining data. The approach will be validated on the Sioux Falls and Calgary transit networks, assessing its robustness under varying data conditions.
This project, led by Dr. Saidi and Dr. Nassir, aligns with Canada’s strategic priorities in quantum technology and transit innovation. By enhancing OD estimation, it aims to improve transit planning, optimize operations, and contribute to more efficient urban mobility. As the intern, I will gain hands-on experience in quantum computing and transportation analytics, fostering collaboration between Canadian and Australian research institutions.

Predicting the bond dissociation enthalpies in lignin-derived molecules using quantum machine learning models

Bond dissociation enthalpy (BDE) is a fundamental chemical property for predicting molecular stability and reactivity. BDEs are crucial for understanding antioxidant efficiency, enzyme catalysis, surface functionalization chemistry, and drug discovery. This project will focus on predicting BDEs for C-O and C-C bond types in lignin-derived molecules, essential for efficient lignin decomposition processes in biofuel production and renewable chemicals. Computational approaches such as density functional theory (DFT) are computationally expensive which limit the size and diversity of BDE dataset for lignin-relevant systems. To address this challenge, we propose an active learning framework combined with Bayesian optimization which iteratively identifies the most informative BDE data points. We will use classical machine learning models to generate high-quality BDE data and minimize the requirements of costly DFT calculations. In addition, we will focus on the potential of quantum models, leveraging properties such as entanglement and superposition. We will develop a “quantum” active learning framework which will enable a more efficient exploration of chemical spaces in lignin-related systems. The use of quantum models will further enhance predictive capabilities by efficiently handling data-scarce scenarios. This work will contribute towards the efficient utilization of biomass for renewable energy applications.

Development of an Augmented Reality Laparoscopic Training System for Gynecological Surgeries

This project aims to create a training system for laparoscopic procedures focused on gynecological surgeries using a 3D-printed torso phantom. The mannequin features apertures for inserting laparoscopic tools (trocars) and a camera, with visible 3D-printed organs inside. The goal is to develop a training module that integrates virtual anatomical models with real-world views using HoloLens 2 augmented reality glasses. These models will assist trainees by overlaying relevant anatomical structures or other virtual models useful for training.

Implementing image-based SSL methods for laparoscopic surgical video data

The proposed MITACS GRA project is part of the Human Surgeome Project at the German Cancer Center, which uses advanced machine learning and deep learning technologies to improve surgical practices. By analyzing large amounts of surgical video data, the project aims to enhance surgical training, skill assessment, and workflow. It will develop systems that provide real-time feedback to surgeons, helping to improve efficiency, safety, and decision-making during surgeries. Ultimately, this project seeks to make surgery safer and more standardized, reduce complications, and improve patient outcomes, making healthcare more efficient and evidence-based.

Gamifying Shakespeare: Theorizing and Designing Game-based Digital Media for Stratford Festival Audience Engagement

The aims of the partner organization are to: bolster audience engagement, i.e., to enhance the experience of those already attending the Stratford Festival; bolster audience attendance, i.e., to draw new crowds to the Stratford Festival; and establish the Festival as a leading digital media creator in the field of professional theatre. With the intern’s focus on enhancing player/audience engagement, the creation of a digital game based upon the partner organizations staged productions (beginning with one in the 2015 season) will allow audiences new ways to connect emotionally and empathically with the partner organization and their work.

ESROP-NUS-Measurement of magnetic field vector by a compact diamond quantum sensor

Measurement of magnetic field vector by a compact diamond quantum sensor

Measuring magnetic field is important for material and device characterization. Although various methods have been developed, challenges persist in accurately measuring local magnetic field vectors. Our lab has been developing a compact diamond quantum sensor to overcome the challenge. The diamond sensor uses nitrogen vacancy centres which have unique spin properties enabling room-temperature optical read-out of spin states. By measuring the magnetic resonances of the spin states, we can extract precise magnetic vector information. Students in this project will learn about the optically detected magnetic resonance (ODMR) system and gain hands-on experience in characterizing and benchmarking the quantum sensor.

Investigating neurescence in dopaminergic neurons of PrknR275W mice

Parkinson’s disease (PD) is a devastating brain disorder accompained by the death of dopamine-producing neurons. Aging is the main risk factor for PD, but early-onset forms also exist and are often linked to mutations in PRKN gene, which encodes for a cellular protein called Parkin. Evidence suggests that these mutations may perturb dopamine neuron function and increase brain inflammation, although the exact mechanisms behind the neuronal death remain unclear. Recent discoveries suggest that “neuronal senescence,” a form of cellular aging, may contribute to age-related brain diseases including PD, but its role in triggering neuronal death is not fully understood. This work aims to study senescence markers in dopamine-containing neurons of a new preclinical mouse model of juvenile parkinsonism that harbors a mutation in PRKN gene. Neuronal senescence could serve as a potential therapeutic target since it can be modulated by senolytic drugs, that have already been developed for other diseases. The findings of this research may lead to new neuroprotective treatments for PD.

Innovative Use of Sidoarjo Mud as Aggregate Replacement in Self-Compacting Geopolymer

The proposed project aims to develop a new, eco-friendly construction material by combining fly ash and Sidoarjo mud to create a self-compacting geopolymer mix. This innovative material will replace traditional aggregates, making it more sustainable and reducing environmental impact. The project leverages the international collaboration to address waste management issues and promote sustainable construction practices. Participating institutions will benefit from this project by advancing research in sustainable construction and gaining access to new, cost-effective materials that can be used in building infrastructure.

Understanding the chemistry of cancer with a smart paper chip

The purpose of the project is to design a paper-based device for the co-detection of nitric oxide (NO) and glucose in a tumor angiogenesis model. By combining paper technologies to electrochemical sensing, we envision that an easy-to-use, smart tumor model can be built. The devices will be tested on co-cultured endothelial and tumor cells, to track the levels of glucose and NO in this model of tumor communication. Pharmacological alterations will be tested as a proof of concept.