30921 research outputs found
Sort by
Converging paths: settler colonialism and the Canadian choral tradition
This thesis resists the grand narrative of Canadian music history by examining Canadian choral works through a settler colonial theoretical framework. I specifically focus on the politics of translation and the relationship between text and music. I exemplify my methodology in this thesis with a brief discussion of R. Murray Schafer's Miniwanka: Moments of Water (1971), explaining how his use of Indigenous words constitutes a form of extractive composition, or a use of Indigenous culture undertaken without the consent of those to whom it belongs in a settler-colonial context. I then apply this methodology to my primary case studies: Imant Raminsh’s Along the Flower Trail: Earth Chants (1982) and Andrew Balfour’s Nagamo (2022). Both works engage with the politics of translation: Raminsh sets Indigenous song in English translation, and Balfour sets English texts translated into Cree and Ojibway. I contrast Raminsh’s extractive methods with Balfour’s unsettling methods to point the way towards a reinvigorated musical practice that avoids reifying settler colonialism. These close readings provide methods to further understand Canadian music history and its close relationship with settler colonialism.October 202
Comparative analysis of polymerization efficiency of Icon resin infiltrant and Embrace Wetbond sealant under varying curing conditions
Purpose:
This study evaluates the degree of conversion (DC) of Icon resin infiltrant and Wetbond sealant
under varying clinical conditions, including differences in substrate type, curing angle, and
distance.
Methods:
Tabletop samples of Icon resin infiltrant and Wetbond sealant were prepared and cured at two
distances: 0 cm and 1 cm from the curing light unit. Fourier Transform Infrared Spectroscopy
(FTIR) was used to determine the degree of conversion of these samples. To simulate clinical
conditions, extracted teeth were mounted in wax to replicate the positioning and angles of curing
in the oral cavity. Icon resin infiltrant and Wetbond sealant were applied interproximally and
cured at distances of 0 cm and 1 cm. FTIR analysis was performed to measure the degree of
conversion, and statistical comparisons were conducted using ANOVA and Tukey's post-hoc
test.
Conclusions:
Results indicate that Icon resin infiltrant consistently achieves a significantly higher degree of
conversion than Wetbond sealant across all tested conditions. The degree of conversion was
influenced by substrate type and decreased with more distal curing angles, highlighting
challenges in achieving optimal polymerization in less accessible areas of the oral cavity. These
findings suggest that Icon may be a more reliable material than Wetbond sealant in terms of
degree of conversion. However, further research is required to evaluate other properties and the
clinical performance of these materials.Charles Lekic Graduate Pediatric Dentistry AwardOctober 202
Orientation-dependence of dislocation-induced incipient plasticity in Mg via nanoindentation technique
Magnesium (Mg) and its alloys have gained significant attention in recent years due to their lightweight nature and potential applications in transportation, aerospace, and biomedical industries. However, one of the major limitations to their broader application is the pronounced poor ductility. To better understand the root causes of this limitation, it is essential to investigate the plasticity across different crystallographic orientations.
This thesis investigates the onset of plastic deformation, with a particular focus on dislocation-mediated incipient plasticity in a coarse-grained Mg–2 wt.% Gd alloy using the nanoindentation technique. A nanoindentation protocol was designed to suppress the formation of deformation twins by applying low indentation loads (1000 µN), enabling the isolation and examination of dislocation nucleation events. A novel “grain signing–finding approach” was developed to accurately identify and repeatedly test specific grain orientations. This allowed for over 100 indents per orientation across 13 distinct grains, including basal, prismatic, and pyramidal orientations.
High-resolution scanning electron microscopy (HRSEM) and scanning probe microscopy (SPM) revealed no evidence of twinning, supporting that deformation at the nano-scale was mediated exclusively by dislocations. Load-displacement curves showed three distinct modes of the first pop-in event—(i) ideal pop-in, (ii) deviation before pop-in, and (iii) excursion before pop-in—all of which demonstrated strong orientation dependence. Furthermore, two types of pop-in sequences were observed: single and successive pop-ins, with successive events occurring only in basal orientations. The basal grains exhibited a higher median resolved shear stress (~1.3 GPa) compared to prismatic and pyramidal grains (~1.0 GPa), consistent with theoretical predictions for dislocation nucleation. This suggests the involvement of pyramidal ⟨c+a⟩ dislocations in basal orientations, facilitating plastic deformation along the c-axis. The activation volume associated with the first pop-in ranged from 27 to 40 ų—involving about ∼2 atoms of magnesium—supporting a surface dislocation nucleation mechanism and ruling out twinning, which typically requires a much larger activation volume.
Interestingly, the observation of successive pop-ins in basal orientations, along with higher stress levels, suggests the possibility of a cross-slip nucleation mechanism to accommodate ⟨c⟩-axis deformation.NSERC Discovery Grant (RGPIN-2019-05882) and Canada Research Chair program (CRC-2021-00512).October 202
Navigating the crossroads between early motherhood and mental health: exploring the role of eHealth in addressing mental health and at-risk substance use among a stigmatized and understudied population
Individuals with co-occurring mental health concerns and at-risk substance use (ARSU+) face barriers to accessing effective treatments. eHealth treatments may offer an alternative to address unmet needs, particularly for underserved populations such as mothers of young children. However, most treatments address mental health or substance use in isolation, which may limit accessibility and relevance for those with ARSU+. Alcohol and cannabis use were the focus of this dissertation due to their high prevalence, legal status, and frequent mental health comorbidity. This dissertation aimed to evaluate the feasibility, clinical utility, and effectiveness of eHealth treatments in general populations and/or mothers of young children. In Study 1, I conducted a meta-analysis and systematic review of 93 randomized controlled trials (RCTs; N = 51,833) to assess eHealth treatment effectiveness. Small effects were shown for alcohol use at post-treatment and cannabis use at follow-up. Moderator analyses revealed larger effects in studies with older samples, non-college/university samples, and those that targeted individuals with at-risk substance use. Treatment components associated with greater alcohol use reductions included relapse prevention, motivational interviewing, cognitive strategies, and emotion regulation and mental health support. In Study 2, I conducted a secondary data analysis of two RCTs evaluating the Building Emotional Awareness and Mental health program, a mental health and parenting treatment for mothers with young children. Feasibility and symptom trajectories were compared among those with ARSU+ (n = 33) and those with mental health concerns without at-risk substance use (ARSU-; n = 77). Significant reductions in depression, anxiety, anger, sleep disturbances, and parenting stress were observed over time in mothers with ARSU+. These improvements were comparable to those observed in the ARSU- group, even after adjusting for enrollment symptom severity. Together, findings suggest eHealth treatments are associated with reductions in alcohol and cannabis use in the general population and may be feasible and support symptom improvements in mothers with ARSU+. Results underscore that single-focus eHealth treatments targeting mental health or substance use alone may still offer preliminary benefit for individuals with ARSU+. These findings highlight the potential value of flexible and inclusive approaches to treatment development and implementation.October 202
Effect of freestream turbulence on flow structure around slanted-back Ahmed body
This study explores the influence of freestream turbulence (FST) on the flow around Ahmed bodies with rounded (RL) and squared (SL) leading edges, both featuring a 25° slanted rear surface. Time-resolved particle image velocimetry was conducted at Re = 0.17 × 10⁵ under FST intensities of 1.5% (baseline), 5.0%, 10.0%, and 15.0% generated by regular and fractal grids.
The SL geometry enlarged the roof recirculation bubble but did not alter wake length, while suppressing downwash and reducing vortex shedding frequency. For both geometries, increasing FST shortened the roof bubble through earlier shear-layer reattachment, whereas wake length remained unchanged. Over the RL slant, bubble size increased with FST up to 10% before stabilizing, linked to stronger interactions between the C-pillar vortex and the separated shear layer. Turbulence statistics showed geometry-dependent effects. In RL, FST suppressed Reynolds stresses over the roof and wake, while promoting transition via elongated streaks dominated by streamwise fluctuations. In SL, FST increased streamwise Reynolds stresses and turbulent kinetic energy (TKE) on the roof through energy redistribution to low-frequency motions (St < 0.23), though wall-normal and shear stresses were minimally affected. In both cases, wake TKE decreased due to weakened wall-normal stresses and disrupted vortex interactions. Frequency spectra and Spectral Proper Orthogonal Decomposition (SPOD) analysis revealed that in the RL geometry, FST suppressed dynamically important frequencies, including roof bubble pulsation, and disrupted the flapping motion and spectral coherence of wake recirculation bubbles. In the SL geometry, SPOD and quadrant-based analysis showed that FST amplified low-frequency flapping over the roof and promoted persistent C-pillar vortices that dominated the upper shear layer, disrupting classical vortex shedding in the wake. The results offer critical insights for refining aerodynamic models and developing flow control strategies tailored to real-world turbulent conditions.February 202
Transfer accuracy of 3D printed trays for indirect bonding of custom 3D printed orthodontic brackets
Objectives: To evaluate the transfer accuracy of custom orthodontic brackets using 3D printed transfer trays in a fully digital workflow.
Methods and Materials: Twenty-one participants who were treated with LightForce (LightForce, Burlington, MA, USA) brackets were bonded using 3D printed transfer trays. Intraoral scans (iTero, Align Technology, San Jose, CA, USA) were taken after brackets were bonded. The virtual bracket setup was compared to post-bonding scans using digital model superimpositions. Discrepancies in bracket position were assessed, and one-sided t-tests were used to determine if the mean transfer discrepancies were within clinically acceptable limits of 0.5 mm in linear dimensions and 2° for angular dimensions.
Results: The mean bracket position differences were 0.21 mm, 0.39 mm and 0.38 mm for mesiodistal, buccolingual, and occlusogingival directions, respectively. The frequency of bracket positions within 0.5 mm ranged from 73% to 95%. Mean bracket position differences were 1.69°, 1.09° and 1.70° for torque, tip, and rotation, respectively. The frequency of bracket positions within 2° ranged from 69% to 84%. P-values ranged from 0.00 to 0.694. The mean differences were statistically significant (p<0.05) and within clinically acceptable limits for all teeth in mesiodistal and tip dimensions.
Conclusion: 3D printed transfer trays have high transfer accuracy in many dimensions. Bracket position discrepancies in angular dimensions are more frequent than in linear dimensions. A digital workflow using virtually planned bracket positions and 3D printed transfer trays, as judged from this study, may be implemented with clinical confidence in indirect bonding of orthodontic bracket.October 202
Advanced kernel-based approaches for robust inference in intractable models
Modern statistical and machine learning models often involve high-dimensional parameters, latent variables, or complex hierarchical structures, leading to likelihood functions that are both difficult to express in closed form and costly to differentiate. With data volumes growing and architectures such as deep neural networks becoming more complex, conventional likelihood-based methods struggle to provide reliable inference and model diagnostics, giving rise to intractable likelihoods and gradients.
In response to these challenges, this thesis develops and evaluates new Kernel Stein Discrepancy (KSD) methods for efficient and robust inference in intractable settings. First, it introduces Gradient-Free Kernel Conditional Stein Discrepancy (GF-KCSD), a nonparametric goodness-of-fit test that requires no gradient information. GF-KCSD offers theoretical guarantees and strong empirical results, outperforming standard approaches in importance sampling scenarios.
Next, the thesis investigates GF-KCSD’s interpretability and error control under measurement errors, demonstrating its robustness and highlighting its ability to maintain Type-I error control and stable power even when data are noisy.
Finally, while KSD-based methods (or score-based methods in general) continue to gain traction, they often encounter mode blindness. To resolve this, the thesis proposes Mode-Sensitive KSD-Bayes (MS-KSD-Bayes), a weighted extension of KSD that balances multimodal posterior distributions within a generalized Bayesian framework. Across the Galaxy velocities and lung-cancer gene expression datasets, MS–KSD–Bayes shows improved mode sensitivity and comparable or tighter posteriors relative to KSD–Bayes and gradient-based methods, within the light-tailed, moderate-dimensional regimes we study.
Collectively, these contributions form a gradient-free, mode-sensitive suite of KSD methodologies that significantly expand feasible statistical modeling for intractable likelihoods, ultimately enriching the toolkit for modern data analysis and inference.February 202
A-I-D for cascades: an application of the Behaviour Change Wheel to design a theory-based intervention for addressing prescribing cascades in primary care
Abstract Background Prescribing cascades, which occur when a medication is used to treat the side effect of another medication, are important contributors to polypharmacy. There is an absence of studies that evaluate interventions to address them. We describe an application of the Behaviour Change Wheel (BCW) to design theory-informed interventions for addressing prescribing cascades within interprofessional primary care teams. Methods The BCW framework was applied to guide intervention development. This report describes the first seven steps. Three behaviours were developed based on data collected from two qualitative studies exploring why and how cascades occur across practice settings. A target behaviour was selected and the COM-B model was applied to identify relevant factors for interprofessional primary care teams. Relevant intervention types, policy options, and corresponding behaviour change techniques (BCTs) were identified, and intervention examples drafted. Prioritization of behaviours and intervention examples were guided by the APEASE criteria. Results The three behaviours involved supporting: (1) healthcare providers (HCPs) to ask about, investigate and manage cascades, (2) the public to ask about prescribing cascades, and (3) the public to share medication histories and experiences with HCPs. The team selected the HCP behaviour, A-I-D (ask, investigate, deprescribe), for intervention development. Psychological capability and physical opportunity were the most relevant COM-B components. Ten intervention options comprised of BCTs were developed, which are ready for further prioritization by stakeholders. These can be grouped into: provision of educational materials for use by HCPs; provision of consultation or training to support HCPs; and knowledge mobilization strategies. Through the process, the team identified that development of a practice guidance tool, which assists HCPs to investigate and manage prescribing cascades, is needed to support further intervention development. Conclusions The BCW framework guided the design of intervention examples to support primary HCPs practicing in interprofessional teams to address prescribing cascades. When identifying interventions for future consultation, creation of a practice guidance tool was prioritized as it underpins all proposed interventions for addressing prescribing cascades in practice. Further research is needed to determine what primary HCPs would need in this practice guidance tool and how it will be used in practice, to support its development
Improving shell-based phantom materials and quality assurance in breast microwave sensing
Breast Microwave Sensing (BMS) systems require adequate testing and performance before reaching clinical trials. This thesis presents improved materials for shell-based breast phantoms and quality assurance devices to provide comparable testing in the field. Previous studies utilized 3D-printable plastics to create shells for shell-based phantoms; however, the low permittivity of these plastics impacts the phantoms’ dielectric accuracy. The liquids filling shell-based phantoms also often contain air bubbles, leading to undesirable microwave scattering. This research explores new tissue-mimicking materials to overcome these issues. The low-permittivity 3D-printed plastic filament have been substituted with graphite, carbon black, and resin to replicate the skin’s properties within the 0.4-9.0 GHz range. Diethylene Glycol Butyl Ether (DGBE) solutions have replaced glycerin and Triton X-100 to simulate the characteristics of adipose and fibroglandular tissue. The resin-based material more accurately mimicked the properties of ex vivo tissue samples than 3D-printed plastics. DGBE solutions demonstrated superior dielectric properties compared to glycerin and Triton X-100 solutions. These DGBE solutions are preferable over glycerin and Triton X-100 due to their lower viscosity, reduced tendency for air bubble formation, better short-term stabil-ity, temperature stability, and enhanced long-term stability, which supports the reusability of these materials. Quality assurance devices were designed and 3D-printed to analyze and compare BMS systems, an area the field has been lacking in the literature. The materials studied in this work improve on existing breast phantoms by ensuring dielectric accuracy and stability, offering better experimental utility for microwave breast imaging. At the same time, the devices presented can be used for quality metrics to appropriately compare BMS systems on a fundamental level.February 202
Primary prevention of food allergy: beyond early introduction
Abstract Food allergy typically begins early in life and persists as a lifelong condition. Delayed introduction of allergenic foods followed by years of hesitancy to introduce these foods early may have contributed to the increase in food allergy prevalence in recent decades. Most infant feeding guidelines focus on the importance of early introduction of allergenic foods in infants at around age 4–6 months. However, regular, ongoing ingestion of allergenic foods is also critical for the primary prevention of food allergy. Similarly, intermittent exposure to cow’s milk formula (CMF) in early infancy increases the risk of cow’s milk allergy (CMA), while regular exposure (if it is introduced) prevents it. Families hesitant to introduce allergenic foods to their infant at home (despite education) should be offered introduction in a primary care clinic. Infants who have failed primary prevention should be referred to an allergist for consideration of early infant oral immunotherapy (OIT)