Doctoral Dissertations (FSCI)

Permanent URI for this collectionhttps://hdl.handle.net/10155/402

Browse

Now showing 1 - 20 of 76

Contextual topics: advancing text segmentation through pre-trained models and contextual keywords
(2024-09-01) Maraj, Amit; Vargas Martin, Miguel; Makrehchi, Masoud
Text Segmentation (TS) is a Natural Language Processing based task that is aimed to divide paragraphs and bodies of text into topical, semantically aligned blocks of text. This can play an important role in creating structured, searchable text-based representations after digitizing paper-based documents. Traditionally, TS has been approached with sub-optimal feature engineering efforts and heuristic modelling. In this work, we explore novel supervised training procedures with a labeled text corpus along with a neural Deep Learning model for improved predictions. Results are evaluated with the Pĸ and WindowDiff metrics and show performance improvements beyond any previous unsupervised TS systems evaluated on similar datasets. The proposed system utilizes Bidirectional Encoder Representations from Transformers (BERT) as an encoding mechanism, which feeds to several downstream layers with a final classification output layer, and even shows promise for improved results with future iterations of BERT. It is also found that infusing sentence embeddings with unsupervised features, such as the ones gathered from Latent Dirichlet Allocation (LDA), provides comparable results to current state-of-the-art (SOTA) TS systems. In addition to this, unsupervised features derived from LDA give the proposed system the ability to generalize better than previous supervised systems in the space. Furthermore, it is shown that with the use of novel language models such as Generative Pre-trained Transformers (GPT) for text augmentation, training data can be multiplied, while continuing to see performance improvements. Although the proposed systems are supervised in nature, they have the capability of fine-tuning a threshold variable that allows the system to predict segments more frequently or sparingly, further bolstering the practical usability of it. Due to the increasing competition in the supervised TS space, creating competitive systems often see contributions from larger research companies with more available resources (e.g., Google, Meta, etc.). However, unsupervised TS has been relatively unexplored in comparison with supervised efforts, since it is much more challenging to build a generalizable TS system. To this end, strong word and sentence embeddings are used to create an unsupervised TS system called “Coherence”, that blends the best of pre-trained models and unsupervised features to create a system that is capable of generalizing across various datasets, while achieving competitive results in the space. Since Coherence is unsupervised, inference is quick and requires no upfront investment (i.e., this technique can be picked up and applied to a domain without the need for fine-tuning).
Three essays on pricing, quality, and advertising decisions while anticipating a product recall
(2024-08-01) Jafarzadeh Ghazi, Amirhossein; Karray, Salma; Azad, Nader
Product recalls have been rising significantly in both numbers and severity across various industries. The consequences of recalls can be substantial, leading to direct and indirect financial losses for the entire supply chain. Using a game theoretic modelling approach, this thesis examines pricing, quality, and advertising decisions to manage product recalls and mitigate associated risks and losses. The core chapters of the thesis consider three relevant topics on product recall management. In the second chapter, we analyze quality and pricing strategies for two competing firms facing the risk of a severe quality-related recall, which makes the product hazardous and leads to its removal from the market. We develop a two-stage Nash game where the probability of recall depends on the firms’ chosen quality investments, and either firm can experience a recall. Our results indicate that the competitor of the affected firm by the recall should lower its price after the recall if consumers’ price sensitivity changes enough and may increase or keep its price the same otherwise. Surprisingly, considering the risk of a recall does not always lead firms to enhance their product quality. The third chapter considers a monopoly supply chain wherein a manufacturer sells a product through a retailer while anticipating a moderate or minor product recall and subsequent recovery process. Developing a manufacturer-Stackelberg game-theoretic model, we investigate how cost-sharing and revenue-sharing contracts serve as a retailer’s strategies to encourage the manufacturer to enhance product quality, thereby preventing the potential product recall and its associated costs, while also boosting demand. Our results reveal that revenue sharing through bargaining stands out as the most effective contract in driving product quality, diminishing recall probability, and generating higher profits for the manufacturer and the whole supply chain. Nevertheless, the retailer favours typical revenue sharing over the other contracts. In the fourth chapter, we investigate the optimal cooperative advertising and pricing strategies in a bilateral monopolistic marketing channel, including a manufacturer and a retailer, when anticipating a moderate or minor product recall and recovery afterwards. We develop a two-period cooperative advertising game model in which the manufacturer is the leader, and the retailer is the follower. We find that the manufacturer sets a higher wholesale price for its product when there is a recall risk compared to when uncertainties regarding the recall are resolved. Moreover, the retailer’s advertising initiatives and pricing strategies may increase or decrease following the recall, depending on the recall probability and the damage to baseline demand or advertising effectiveness. Finally, the manufacturer consistently favours cooperative advertising pre- and post-recall. This thesis offers several important managerial insights. First, pricing and quality, as well as pricing and advertising decisions, should be managed in an integrated manner to effectively mitigate product recall risks, as independent strategies are less effective. Furthermore, this thesis suggests that managers should adjust their pre- and post-recall strategies in response to changes in the market and consumer behaviour following recalls, such as lost sales and shifts in consumer sensitivity to price, quality, and advertising. Additionally, collaborative contracts between manufacturers and retailers focusing on quality efforts or advertising efforts can significantly enhance supply chain effectiveness when facing product recall risks.
Development and evaluation summaries of a percutaneous nephrolithotomy (PCNL) surgical simulator
(2024-05-01) Sainsbury, Ben; Green, Mark; Ren, Jing
Traditional surgical training methods, such as the use of cadavers, and porcine models (pigs) fall short of providing the realistic, real-time feedback necessary for mastering complex procedures. Virtual reality (VR) simulators with haptics offer a significant improvement, delivering immersive and interactive experiences that closely mimic real-life surgeries. Cadavers are also costly, averaging $1,300 each, and are in limited supply. Simulation devices for teaching surgical skills have been shown to significantly reduce surgical errors, enhancing both the safety and effectiveness of surgical training (Anderson & Abrahamson, 2017; Badash et al., 2016; Bushey, 2013; Chen et al., 2021; Chiang et al., 2013; Fried et al., 2005; Pottle, 2019; Sommer et al., 2021). While many existing surgical simulators incorporate elements of VR and haptics, they often lack full integration of a 3D VR operating environment that immerses surgeons completely in the procedural context. The Marion K181 PCNL Simulator addresses this gap by providing a comprehensive VR surgical simulation platform that integrates advanced haptic feedback and accurate 3D models to create a realistic training environment. This simulator aims to reduce the significant number of deaths caused by medical errors in the US, which total approximately 251,000 annually (James, 2013). These errors lead to additional surgeries, lower patient quality of life, and substantial costs to the healthcare system, estimated at $5.95 billion per year (Badash et al., 2016). The evaluation of the Marion K181 simulator demonstrates its effectiveness in improving surgical skills, with users reporting high levels of satisfaction with the realism of the haptic feedback and the accuracy of the anatomical models. This platform not only enhances the technical abilities of surgeons but also provides a cost-effective, ethical, and sustainable solution for surgical training.
Advancing and expanding siRNA and saRNA therapeutics applications through chemical modifications
(2024-04-01) Giorgees, Ifrodet; Desaulniers, Jean-Paul
Oligonucleotides are short strands of DNA or RNA that are used to treat complex diseases like cancer and rare genetic diseases. They rely on biological pathways in our body to work. Two pathways that are important to this study are gene silencing and activation. Short interfering RNAs (siRNAs) silence genes, while short activating RNAs (saRNAs) activate them. Both types of strands can be used to create new cancer treatments. However, RNA-based therapies face challenges like instability, off-target effects, and low cell membrane permeability. To overcome these challenges, this study focuses on incorporating new chemical modifications into the RNA and assessing their impact on RNA activity. Our aim is to enhance RNA therapeutic efficacy for potential cancer therapy applications. The first goal focused on creating a combination therapy for cancer treatment by directly conjugating free base corrole molecules to siRNA. This novel construct created a combination therapy effect of gene silencing and simultaneous photodynamic therapy (PDT). This combination therapy is expected to be more targeted and non-invasive compared to traditional cancer treatments like surgery or chemotherapy. The second goal of this research involved exploring the potential of metal corrole molecules within siRNA for personalized cancer treatment. In this study, Ga-corrole was directly conjugated to siRNA, resulting in an advanced treatment consisting of live imaging and gene silencing. This novel construct created a new tool for siRNA real-time imaging applications that could potentially allow for real-time drug monitoring during cancer treatment. The third goal focused on discovering nuclease-resistant and active saRNAs targeting STING, which is a potential target for the treatment of solid tumors. In this study, a library of chemically modified saRNA was screened for their nuclease resistance ability and investigated for any potential correlations between chemical modifications, nuclease resistance and high gene upregulation activity. The results of nuclease stability assays revealed that the position of the chemical modifications within the RNA can significantly influence nuclease resistance. Furthermore, novel chemical modification designs were established for the synthesis of stable and highly active STING saRNA duplexes. In conclusion, this dissertation highlights novel approaches to enhance RNA therapeutics and employ RNA molecules for cancer drug monitoring or treatment applications.
Equivariant Kolmogorov-Arnold-Moser theory
(2024-03-01) Faulkner, Nicholas; van Veen, Lennaert; Buono, Pietro-Luciano
KAM theory is a collection of theorems and approaches to describe the stability of certain invariant tori within nearly integrable Hamiltonian systems. The invariant tori are associated with a set of torus frequencies, and a key result in the theory is that these frequencies must satisfy a Diophantine condition, with no rational resonances amongst the frequencies. In this paper we investigate the additional property of equivariance with a discrete symmetry group, Γ. The setting then becomes Γ-Equivariant, nearly integrable, Hamiltonian, ODEs. In this setting the Diophantine condition is no longer valid as the symmetry constraints imposed by Γ, generically force the frequencies to be repeated. Utilizing equivariant-singularity and linear representation theory we develop an alternative Γ-Diophantine condition and show that by modifying the Diophantine assumption in this way KAM theory can be applied to this new setting. Moreover, we demonstrate that this condition can be utilized in the Hyperbolic, Γ- Equivariant case.
DL-based defense against polymorphic network attacks
(2024-01-01) Sabeel, Ulya; Heydari, Shahram; El-Khatib, Khalil
Network security is of vital importance in our world dominated by internet systems. These systems are vulnerable to large-scale rapidly evolving attacks by sophisticated cyber attackers who can have an upper edge over the defensive systems. Artificial Intelligence (AI) based intrusion detection systems provide effective defense mechanisms against cyber attacks. However, these techniques often rely on the same dataset for training and validation as well as evaluation of AI models. Current research [1] also confirms that such trained models can accurately identify known/typical network attacks but perform poorly when faced with continuously evolving atypical/polymorphic cyberattacks. Therefore, it is crucial to develop and train an AI-based Intrusion Detection System (IDS) that proactively learns to resist infiltration by such dynamically changing attacks. For this purpose, in this research work, we propose an AI-based IDS system that can monitor and detect polymorphic network attacks with high confidence levels. We propose a novel hybrid adversarial model that leverages the best characteristics of a Conditional Variational Autoencoder (CVAE) and a Generative Adversarial Network (GAN). Our system generates adversarial polymorphic attacks against the IDS to examine its performance and incrementally retrains it to strengthen its detection of new attacks, specifically for minority attack samples in the input data. The employed attack quality analysis ensures that the adversarial atypical/polymorphic attacks generated through our system resemble realistic network attacks. Our experiments showcase the exceptional performance of the proposed IDS by training it using the CICIDS2017 and CICIoT2023 benchmark datasets and evaluating its performance against several atypical/polymorphic attack flows. The results indicate that the proposed technique, through adaptive training, learns the pattern of dynamically changing atypical/polymorphic attacks and identifies such attacks with high IDS proficiency. Additionally, our IDS surpasses various state-of-the-art anomaly detection and class balancing techniques.
Identification and characterization novel of cystine-loop ligand- gated chloride channels from Dirofilaria immitis: pharmacological analysis and novel compound screening
(2023-12-01) Varley, Sierra; Forrester, Sean
Dirofilaria immitis, otherwise known as heartworm, is a parasite that infects the hearts of dogs and causes serious health consequences. While preventative treatments are available drug resistance is developing at an alarming rate. Cystine-loop ligand-gated ion channels are important receptors in nematode neurobiology, and as such are promising drug targets. The UNC-49 (GABA-gated chloride channel) and ACC (acetylcholine-gated chloride channel) family of receptors have been characterized as potential drug targets in other nematodes. However, these receptors have yet to be identified or characterized in D. immitis. This thesis investigates the cloning and pharmacological characterization of 5 novel receptor subunits from D. immitis: UNC-49B and UNC-49C, ACC-1, LGC-46 and LGC-47. Additionally, novel derivatives of the antiparasitic drug levamisole were tested on ACC receptors to determine if any modifications enhanced levamisole action. D. immitis UNC-49B assembled as a homomeric channel and exhibited an EC50 of 5mM for GABA. UNC-49B also formed a functional channel with UNC-49C, which exhibited a decrease in GABA sensitivity. Additionally, the D. immitis UNC-49 receptors were significantly more sensitive to the open channel blocker picrotoxin compared to the same receptors from the sheep parasite Haemonchus contortus. Moreover, D. immitis UNC-49C, unlike other UNC-49C subunits, did not cause a decrease in picrotoxin sensitivity when assembled with UNC-49B. D. immitis ACC-1, LGC-46, and LGC-47 were unable to form functional channels through heterologous expression in Xenopus laevis oocytes. Levamisole derivatives were therefore tested on H. contortus ACC-2, which identified at least one showing a higher sensitivity compared to levamisole. Overall, this study characterized 2 novel UNC-49 receptors, identified 3 potential members of the ACC family in D. immitis and tested 8 novel derivatives of levamisole. This study lays foundation for the identification of more ligand-gated ion channels and will serve as a starting point for future researchers looking at new drug potential targets in D. immitis.
Leveraging vehicular cloud computing through location and request prediction
(2023-04-01) Miri, Farimasadat; Pazzi, Richard
In recent years, the implementation of Vehicular Ad-hoc networks (VANET) has been acknowledged as a promising solution for monitoring road conditions. Despite the many benefits they can bring to society, the growing demand for communication, storage, and processing capabilities is giving rise to new challenges. For instance, the heightened need for communication in VANETs can cause network congestion. Additionally, the real-time nature of VANET applications, such as traffic management, accident prevention, and navigation, necessitates rapid and reliable communication, which can be difficult to achieve in a network that is constantly changing. Moreover, managing and scaling the resources of a large scale network like VANET with a large number of vehicles and road-side units (RSUs) is a significant challenge. There are essential contributions to deal with these problems, such as utilizing MEC (Mobile Edge Computing), and a hybrid architecture of cloud and fog computing which can create an efficient and adaptable resource management system. However, having unpredictable events such as accidents or road closures can cause rapid changes in the network topology, making it difficult to allocate resources effectively. Also, unpredictable events can lead to a lack of information, making it difficult to obtain accurate and up-to-date information about the network and hard to allocate resources effectively. Furthermore, allocating resources at the right time when unpredictable events happen without network congestion is another challenging problem that causes us to think about proposing a model that can satisfy delay sensitive applications requirements and decrease the monetary cost in hybrid cloud and fog architecture in the presence of unpredictable congestion. To reach our goal, we first need to estimate the traffic flow after an accident, predict the level of congestion based on the number of requests from different vehicles, and then predict the location of potential vehicles as mobile fog nodes in advance to form Vehicular Clouds. Finally, we propose a layer-based architecture with two prediction models and different modules to support safety and non-safety applications, and a task scheduling mechanism to decrease monetary costs and delay for delay-sensitive applications during congestion times to serve the vehicles requests in that region.
Integrated traffic analysis and visualization for future road events
(2023-04-01) Alghamdi, Taghreed; Elgazzar, Khalid
The existing traffic simulation methods are limited to specific synthetic scenarios. In addition, the natural structure of traffic and accident data requires modeling the dependent observations on multiple levels. Therefore, a system that utilizes hierarchical LMMs and GBM models are proposed which adaptively analyzes and predicts the traffic pattern based on hypothetical inputs. We developed a user-friendly interface to show the outcomes of the hybrid model. The proposed system encompasses three major components: (1) a road accident simulator and event profile to simulate an accident and predict its effects on traffic status; (2) a robust spatiotemporal traffic speed prediction model that integrates the impact of road accident with the prediction model to adaptively predict the future traffic status in response to this accident; (3) a traffic simulation tool to present the future traffic status. Our system provides satisfactory prediction results in terms of predicting with small errors, obtaining optimal hyperparameters, and less computational complexity. The hierarchical structure of the spatial component in our approach effectively captures the correlation in traffic status across different spatial points on the same road. Furthermore, computing the traffic speed at different spatial levels and how it interacts with lagged prior traffic speed over the past four periods and a day prior up-scaled the system efficiency. Evaluation is conducted to test the functionality, usability, and viability. Performance evaluation shows that the event profile model achieves small error rates with an MSE of 0.24 and an RMSE of 0.53 on the testing data, demonstrating satisfactory performance. For traffic status, the integrated model achieves high accuracy with low computational complexity. The boosted LMMs achieved high performance on the test data with an R2 of 0.9190 and an R2 of 0.9291 on the full-fitted dataset. The MAE and RMSE are 0.27 and 0.80, respectively, indicating that the fitness of our data was excellent.
Agent-based modeling framework for adaptive cyber defence of the Internet of Things
(2022-12-01) Rafferty, Laura; Hung, Patrick
The adoption of the Internet of Things (IoT) continues to increase significantly, introducing unique challenges and threats to cybersecurity. In parallel, adaptive and autonomous cyber defence has become an emerging research topic leveraging Artificial Intelligence for cybersecurity solutions that can learn to recognize, mitigate, and respond to cyber attacks, and evolve over time as the threat surface continues to increase in complexity. This paradigm presents an environment strongly conducive to agent-based systems, which offer a model for autonomous, cooperative, goal-oriented behaviours which can be applied to perform adaptive cyber defence activities. This thesis aims to bridge the gap between theoretical multi-agent systems research and cybersecurity domain knowledge by presenting a novel applied framework for adaptive cyber defence that can address a wide range of challenges and provide a foundation for significant future research in systems modeling for cybersecurity. Belief-Desire-Intention (BDI) agent architecture is extended within this work through a novel application of knowledge graphs to provide a scalable data model for agents to understand their environment, infer the context of threats, create goals associated with security requirements, and select plans based on possible actions and expected results. The framework has been implemented to demonstrate the feasibility of the architecture and evaluate the design properties through applied security use cases. While the experimental results have demonstrated the value of the framework applied to IoT systems, the concept can be easily expanded to other domains. This thesis provides the foundation to inspire further research works in this area for continued development, application, and optimization to support the advancement of the industry and bring autonomous, adaptive cyber defence to realization.
Molecular dynamics simulations and neural network solutions for applications in biophysics
(2022-12-01) Nagel, Andrew; de Haan, Hendrick W.
As computing resources evolved and became more accessible over time, much of scientific research shifted towards utilizing computational techniques. In particular, biophysics is a field of science that has continually benefited from the advancement of computers. In biophysics, numerical models of biological phenomena generated from techniques traditionally used in mathematics and physics are solved using computational methods. One of the oldest and most popular of these computational methods is molecular dynamics (MD) simulations. On the contrary, deep learning is a newly emerging family of methods that have only recently found success in the biophysics community. In this dissertation, I present several contributions to biophysics by employing both MD simulations and a deep learning-based approach referred to as the neural network method (NNM). Specifically, the collective motion of ensembles of bacterial twitchers, the structure and dynamics of a phytoglycogen nanoparticle, and nanoparticle mobility through the slit-well microfluidic device are studied using MD. In each of these applications, varying modelling resolutions are chosen, reflecting the trade-off between modelling accuracy and computational efficiency. In addition to utilizing MD, the NNM is applied in this dissertation to solve partial differential equations modelling phenomena in the slit-well microfluidic device. That is, both the driving electric field and the solution to a parameterized equation modelling the mean first passage time of nanoparticles through the device are generated using the NNM. In all applications of the NNM, the accuracy and effectiveness of the technique are analyzed and benchmarked against results obtained using MD simulations.
Synthesis, characterization and biological investigation of self-delivering and modified short interfering RNAs (siRNAs)
(2021-07-01) Salim, Lidya; Desaulniers, Jean-Paul
Aberrant gene expression is a hallmark of disease, so it is of great interest to develop targeted therapies that provide a means to regulate gene expression. The RNA interference pathway serves as a natural defense system against invasive genetic information and results in gene silencing by targeting and degrading mRNA. Synthetic short interfering RNAs (siRNAs) can use this endogenous machinery and have emerged as a novel class of gene-silencing therapeutics. Unfortunately, the development of RNAi therapeutics has been hindered by several challenges associated with the nature and structure of RNA. To harness their full potential, siRNAs must be chemically modified to improve their pharmacokinetic profiles. This dissertation reports the use of two bioconjugates, cholesterol and folic acid, to improve the cellular uptake and delivery of siRNAs and explores the incorporation of a novel sugar moiety within siRNAs to assess its effect on gene-silencing activity. Cholesterol has been extensively used as a delivery vector for nucleic acids. In this work, we show a novel way to functionalize siRNAs with cholesterol, via a triazole linkage, and demonstrate the efficacy of these self-delivering siRNA. Despite their promise, lipid-conjugated siRNAs tend to accumulate in areas like the liver and kidneys, so there is great interest in developing siRNA-conjugates to target other cells and tissues. Based on this, we explored the use of a folate ligand to selectively deliver siRNAs to cancer cells via the folate receptor. This receptor is highly overexpressed in numerous cancers and has become an important molecular marker in cancer research. Here, we show that centrally modified folate-siRNA conjugates display enhanced gene-silencing activity and can be selectively delivered to folate receptor-expressing cancer cells. Lastly, we explore the incorporation of a novel glucose moiety, triazole-linked to uracil at position one, in the sense or antisense strand of siRNAs. The resulting siRNA duplexes contained a single 3′-6′/2′-5′ phosphodiester linkage and achieved good gene-silencing activity. Together, this dissertation demonstrates the efficacy of several chemical modifications at improving some of the limitations associated with siRNAs, providing new avenues for the development of safe and effective RNAi therapeutics.
Calculating thermochemical equilibrium for multiphysics simulations of nuclear materials : development of yellowjacket gibbs energy minimiser
(2022-11-01) Bajpai, Parikshit; Piro, Markus
Nuclear fuels and structural materials are highly complex systems that are remarkably challenging to understand and model. Material behaviours are influenced by multiple physical phenomena such as mechanics, chemistry, heat and mass transport, etc. Moreover, lower scale phenomena inform and drive the phenomena at larger scales. The strong interactions between multiple physics at different length and time scales creates a need for multi-scale, multi-physics modelling tools. In nuclear fuels and structural materials, the problem gets compounded by the fact that, in addition to an extreme environment, the composition of the system changes with time. For such complex systems, computational thermodynamics plays a valuable role in predicting many phenomena and is often necessary for understanding and informing others. For this reason, there has been an increasing interest in incorporating equilibrium thermodynamics calculations in multi-physics frameworks such as the Multiphysics Object Oriented Simulation Environment (MOOSE). To simulate corrosion in molten salt reactors, a new MOOSE-based tool named Yellowjacket has been developed and this work contributes to it. The objective of this work is to develop a new equilibrium thermodynamic solver to provide thermodynamic material properties and boundary conditions for Yellowjacket and other MOOSE-based codes. While several thermodynamics codes already exist, the new software, called Yellowjacket–GEM, adds native equilibrium thermodynamic capability to MOOSE and aims to address several concerns such as computational performance, limitations on system size and models, and Software Quality Assurance (SQA). Yellowjacket–GEM exploits the fundamental laws of thermodynamics to solve a non-linear, non-convex optimisation problem. Several thermodynamic models, including the Modified Quasichemical Model in Quadruplet Approximation (MQMQA) were implemented, and state-of-the art numerical solvers in Portable, Extensible Toolkit for Scientific Computation (PETSc) were used to efficiently solve the optimisation problem. In doing so, the work contributes to the understanding of MQMQA which until recently wasn’t well comprehended. Ensuring that the solver gives a true equilibrium solution also requires solving a global optimisation problem without severely compromising performance and reliability. Several global optimisation methods were compared through numerical experiments to objectively select the best approach for implementation. The C++ code follows MOOSE coding standards and SQA procedures and enables direct coupling of thermodynamic equilibrium calculations in multiphysics simulations performed using MOOSE.
A privacy preservation framework for smart connected toys
(2019-09-01) Yankson, Benjamin; Hung, Patrick
Advances within the toy industry and interconnectedness have resulted in the rapid and pervasive development of Smart Connected Toys (SCTs), built to aid children in learning, socialization, and development. A SCT is a physical embodiment artifact that acts like a child user interface for toy computing services on Cloud. SCTs extend the capability of the traditional toy into a new area of computer research by incorporating the physical component of a traditional toy combined with networking and sensory capabilities of mobile devices using ubiquitous technologies. These SCTs are built as part of the Internet of Things (IoT) with the potential to collect terabytes of personal and play information; introducing ever-increasing privacy, and serious safety concerns for children. SCTs can gather data on the context of the child user’s physical activity state (e.g., walking, standing, running) and store personalized information (e.g., location and activity pattern) through a camera, microphone, Global Positioning System (GPS), and various sensors such as facial recognition or sound detection. Privacy concerns itself with the protection against the intrusion of someone’s defined space without explicit consent, in such a manner that the defined space is protected from intrusion, interference, and information access by a non-authorized entity. In addition to privacy and safety concerns, criminals using a child’s Personal Identifiable Information (PII) can create false identities to engage in a variety of financial frauds and other crimes. The challenge is, with so many SCTs in the market, how to develop a framework and techniques to protect the privacy of children’s data; and in the case where a privacy breach occurs (cybercrime), how to develop a digital forensic framework to analyze data on SCTs. This thesis surveys the cybersecurity and the privacy landscape of state-of-art emerging technology and development in SCTs; investigates technical and legislative related privacy issues in SCTs with functionalities that collect, process, and transmit PII; and presents a privacy preservation framework that will address privacy challenges within SCTs. In addition, this thesis develops a SCT digital forensic processing framework for privacy breach or crime-related investigation. The privacy preservation framework includes a context data model and a privacy-preserving data-modelling framework. The data context model is an abstract model that organizes elements of data and standardizes how they relate to one another and to properties of the related entities in SCTs based on eXtensible Markup Language (XML). The privacy preservation framework is depicted by Petri-Nets and will identify offensive (non-privacy compliant) content intended for storage or transmission, tag, classify, alerts, and secure delete content. The framework consists of four major components, including (1) Systematic Privacy Impact Assessment Table (SPIAT); (2) Privacy-Preserving Context Ontology (PPCO) Model; (3) SCT privacy preservation data model using Petri-Nets; (4) Empirical Study to support Petri-Nets model as one of the most promising tools to support data flow privacy; and (5) Case studies of SCT digital forensic investigation.
The analysis of the entomological and chemical decomposition of human remains to ultimately assess the viability of the domestic pig as a substitute in forensic applications
(2022-09-01) Skopyk, Angela D.; LeBlanc, Helene
Death investigations often rely on the minimum post mortem interval (minPMI) estimations provided by forensic entomologists. The models accepted by courts in Canada are not based on research involving humans but rather on a human substitute, the domestic pig. However, now that facilities for human decomposition research are opening, we are faced with the prospect that pigs may not be so similar to humans as originally thought. The purpose of this research is to analyze the entomological and chemical decomposition of human and pig remains to determine if the domestic pig is an appropriate substitute for humans in research applied to minPMI estimations. Two human (n=4, n=2) and one pig (n=2) study were performed in Sydney, Australia, while three pig (n=3, n=2, n=3) studies were performed in Oshawa, Ontario. The environmental conditions were monitored as well as the accumulated degree days (ADD), rates of decomposition, primary dipteran colonizers, and volatile organic compound (VOC) production. Domestic pigs in Oshawa, Ontario, had rates of decomposition that were highly alike. Rates of insect colonization were rapid, with little to no delay. The production of 5 known apneumones showed no significant difference (p < 0.05) between pigs. Human decomposition in Sydney yielded results with varied rates of decomposition and colonization – some with long pre-colonization intervals (Pre-CIs), which affected the rate of decomposition. It was noted that the donors with long Pre-CIs were likely to have been taking strong peri-mortem antibiotics due to their antemortem health conditions. These antibiotics could have affected the donor’s microbiome, killing the beneficial bacteria that produce apneumones. The domestic pigs observed in Sydney showed decompositions, colonizations and VOC productions more similar to the pigs in Ontario than to the humans in Sydney. Humans live differently than domestic pigs with varying diets, habits, body types, and medications that can influence their decomposition and colonization after death. Since this cannot be said for the domestic pig, it is recommended that the researching community aim to shift future research to human donors so that the data collected can be applied to human death investigations while considering comorbidities and how it affects insect colonization.
Forward and reverse genetic approaches to investigate cellulose biosynthesis in Physcomitrium patens
(2021-04-01) Behnami, Sara; Bonetta, Dario
Cellulose biosynthesis is a common feature of land plants and involves multimeric complexes composed of cellulose synthase A (CESA) proteins and other structural proteins. The exact stoichiometry of CESA proteins and interactions among proteins within cellulose synthase complex (CSC) is not well understood. Therefore, cellulose biosynthesis inhibitors (CBIs) are useful tools in decoding fundamental aspects of cellulose biosynthesis. Here, I characterize the CBI indaziflam, with a unique mode of action for resistance management, which prevents plant growth by inhibiting cellulose biosynthesis. Arabidopsis thaliana indaziflam resistant mutants were identified through forward genetic screening. Since indaziflam is also active in moss Physcomitrium patens, a forward genetic approach to screen indaziflam resistance was applied on P. patens, and positional cloning combined with next-generation sequencing revealed two point mutations in CULLIN1 (CUL1) and AUXIN/INDOLE-3-ACETIC ACID INDUCED (Aux/IAA) which both are involved in auxin signaling pathways. The mutants were also cross-resistant to synthetic auxin 2,4-D. It is predicted that indaziflam affects plant growth and development and impacted the production and remodeling of plant cell wall directly or indirectly. Moreover, to gain insight into the nature of the protein composition of CSCs, I employed a strategy called Biotin identification (BioID), aimed at identifying proximate and vicinal proteins in vivo associated with CESAs in P. patens. I generated multiple BioID-CESA translational fusions by homologous recombination to identify biotinylated proximate proteins. Due to limitations, including but not limited to different behaviors of fused proteins tagged at the C- or N-terminus, decreased expression level, longer incubation time with biotin, higher incubation temperature, and large size BirA* tag, I was not able to identify any interacting proteins. Another finding of this thesis is that P. patens can be used to produce known natural products that are difficult to obtain by chemical synthesis. An in vivo combinatorial biosynthesis approach was pursued in P. patens to obtain rare cannabinoids with beneficial biological activity. The outcome was to produce rare cannabinoids and some pathway intermediates. This idea's other significant result is designing different drug-candidate-producing moss strains, especially within the chemical class of cannabinoids.
Assessing nearshore water quality and biological condition in the Kawartha Lakes using a community science approach
(2022-08-01) Smith, Erin D.; Kirkwood, Andrea
The Kawartha Lakes, located in south-central Ontario, are a popular tourist destination, with a growing permanent resident population. Consequently, land development in its watersheds continues, and the specific land-use – water quality relationships in this region are unknown. The nearshore zone is where land use has its first impacts on the lake, and provides vital habitat for most lake inhabitants at some point in their life cycle. Despite its importance, the nearshore zone is rarely monitored and further investigation is required to understand human impacts on the nearshore zone. My thesis aimed to elucidate the relationships between land use and the abiotic and biotic condition of the nearshore zone in the Kawartha Lakes. To examine Lake Scugog’s nearshore water quality patterns and relationship with land use, 12 volunteers collected water samples from spring to fall for three years. Land use had significant impacts on chloride at buffer scales and phosphorus at the sub-watershed scale. I also monitored the nearshore biotic community (phytoplankton, zooplankton, and macroinvertebrates) at eight sites in Lake Scugog. Two years of sampling found that macrophyte abundance significantly influenced the phytoplankton, zooplankton, and macroinvertebrate communities. During the pandemic restrictions in 2020, community scientists on 16 lakes collected monthly water samples from June-September. A subset of four lakes had nutrient data across 3-years (2019 - 2021), which allowed comparison of nutrient conditions before and during the pandemic. There were significant differences in water quality between watersheds and a notable impact of Lake Scugog on downstream lakes. There was not a significant impact of pandemic restrictions on nearshore water quality in these lakes. A focal study on Balsam, Cameron, Sturgeon, and Pigeon lakes involved nearshore water quality monitoring (2019 and 2021) and biological sample collection (2021). There was a separation of distinct water quality profiles that grouped Balsam and Cameron, and Sturgeon and Pigeon. Exploring relationships between land use, water quality, and the biotic communities I found that phosphorus was important for driving phytoplankton, zooplankton, and macroinvertebrate community abundance. Overall, these findings provide important information for lake managers in understanding the role of land-use and nearshore ecological condition in lake health.
Opportunities for the deep neural network method of solving partial differential equations in the computational study of biomolecules driven through periodic geometries
(2022-08-01) Magill, Martin; de Haan, Hendrick; Waller, Ed
As deep learning emerged in the 2010s to become a groundbreaking technology in machine vision and natural language processing, it also ushered in many new algorithms for use in scientific research. Among these is the neural network method, in which the solution to a differential equation is approximated by varying the parameters of a deep neural network trial function. Although this idea has been explored with shallow neural networks since the 1990s, it has experienced a resurgence of interest in recent years now that it can be implemented with deep neural networks. A series of empirical and theoretical studies have acclaimed the deep variants of the neural network method for being able to solve many classes of traditionally challenging partial differential equations. These early works emphasized its potential to solve high-dimensional, highly parameterized, and nonlinear equations in arbitrary geometries, all without requiring the discretization of the geometry into a mesh. Problems exhibiting these challenging features abound in computational biophysics, and this thesis presents recent efforts to adapt the neural network method for use in this _eld. The investigations in this thesis center on models of biomolecular motion in periodic geometries. Such models arise, for example, in the study of microfluidic and nanofluidic devices used for the separation of free-draining molecules. These problems exhibit many of the characteristics for which the neural network method is appealing, and serve here as non-trivial test problems on which to characterize its performance. Perspectives from biophysics, numerical analysis, and deep learning are combined to elucidate the true potential of the neural network method as a technique for studying such differential equations. Altogether, these works have moved the neural network method closer to being another reliable numerical method in the computational biophysicist's toolkit.
Impact of neck muscle fatigue on upper limb sensorimotor integration
(2019-08-01) Zabihhosseinian, Mahboobeh; Murphy, Bernadette
Neck muscles have a high density of sensory receptors which project to the central nervous system, and have important role in sensory motor integration (SMI). The cerebellum is important for SMI and it is known to undergo neuroplastic changes in response to performing a novel motor acquisition task. Alterations in neck sensory input impacts motor output to upper limb muscles, cerebellar disinhibition, and performance accuracy in response to novel motor acquisition. Despite this, the impact of cervical extensor muscle (CEM) fatigue on cerebellar-motor cortex plasticity and SMI pathways in response to a novel motor acquisition task has not been yet investigated. Study one used short latency somatosensory evoked potentials (SEPs) to investigate the differential effects of CEM fatigue on motor learning and retention; and on sensorimotor processing from distal hand muscles. CEM fatigue impaired upper limb motor learning performance in conjunction with differential changes in SEP peak amplitudes related to SMI. Study two used a paired pulse cerebellar-motor cortex transcranial magnetic stimulation (TMS) technique to determine whether CEM fatigue alters cerebellar disinhibition in response to novel motor skill acquisition. Neck fatigue led to a lessened capacity for cerebellar disinhibition coupled with diminished motor learning relative to a control group. Study three used an eye-hand tracking protocol to investigate the effect of CEM fatigue on accuracy of pointing to both visual and hidden targets. CEM fatigue reduced the accuracy of upper limb tracking to a hidden target. Study four used short- and medium-latency SEPs to determine the impact of CEM fatigue on motor performance accuracy and retention of proximal upper limb muscles, as well as neural processing changes in response to novel motor acquisition using proximal upper limbs muscles. CEM fatigue had minimal impact on proximal upper-limb motor performance accuracy, but lead to differential changes in both short- and medium- latency SEP peak amplitudes related to SMI. Overall, this thesis suggesting that SMI areas including the cerebellum are impacted by CEM fatigue, likely because the altered afferent input from the neck due to fatigue alters body schema, impacting awareness of upper limb position sense, resulting in decreased upper limb performance accuracy.
Functionalized carbon surfaces for clean electrochemical energy systems
(2022-03-01) Fruehwald, Holly M.; Easton, Brad; Zenkina, Olena
The development and implementation of clean energy technologies is the way to overcome the global energy crisis and reduce pollution. Therefore, new energy solutions are rapidly needed. Fuel cells (FCs) may become one such solution. FCs are devices that utilize chemical reactions to directly produce electric energy. While these devices are ideal clean energy sources for the transportation industry, conventional FCs are based on expensive materials that implement platinum catalysts on a carbon support (Pt/C). The high cost and limited availability of platinum hinders the applicability of FCs. Nitrogen and metal-doped carbon supports have been investigated as a non-precious metal replacement for costly precious metal-based materials in various electrochemical energy applications. However, the design of non-precious metal materials (NPMMs) involves high temperature pyrolysis treatments, which leads to an almost random distribution of nitrogen atoms on the surface and therefore can limit efficiency. In this work, a method to graft only the most active nitrogenous groups and/or transition metals on the surface of carbon supports was developed. Specifically, diazonium coupling chemistry to covalently attach molecularly defined moieties bearing a terpyridine (tpy) group onto the surface of carbon supports, followed by the introduction of Fe (or other transition metal) centers into anchored tpy groups. Pyridinic nitrogenous groups, which are the basis of tpy, are believed to be required for high activity in the oxygen reduction reaction (ORR) in FC applications and are thought to increase capacitance in SC applications. Upon metal coordination into the tpy sites, the metal-N3/C catalyst shows promising activity for the ORR and SC applications and opens the door for molecularly controlled, inexpensive, and efficient materials. Importantly, upon an energy intensive heat-treatment, the material’s activity does not improve. Confirming that this system’s properties are dictated by the molecularly defined tpy-Fe units, new efficient NPMMs can be fabricated using energy-saving conditions. This design can be applied and optimized to create a family of NPMMs for various clean energy applications. Modifying carbon-based materials opens the way for new low-cost materials for clean energy systems such as FCs, SC, and water oxidation.

Browse

Recent Submissions