University of British Columbia
University of British Columbia: cIRcle - UBC's Information RepositoryNot a member yet
88081 research outputs found
Sort by
Three papers examining mental health service provision with autistic adults in Canada
No full textAutistic adults experience higher rates of mental health challenges than their allistic (non-autistic) peers, and face numerous barriers associated with inaccessible and ineffective mental health systems of care. These barriers compound adverse mental health outcomes, increasing vulnerability to burnout, suicide, and premature mortality, and result in a need for more intensive and extensive mental health supports. Autistic adults have identified a need for translational research to highlight opportunities to improve accessibility and efficacy of supports and promote autistic-affirming mental health research and practices.
This dissertation comprises three papers that address these challenges and offer novel insights into the experiences of mental health and services among autistic adults and the mental health providers that support them. This project was guided by pragmatism, neurodiversity, and social work theories and the lived and professional experiences of the author and external project consultants. Interpretive description and participatory methods were used to generate applicable findings coherent with the lived experiences of autistic adults.
Sixty-seven participants (41 autistic adult clients and 26 mental health providers) shared their experiences in an online asynchronous qualitative questionnaire. Findings shared across three papers highlight opportunities for enhancing and strengthening autistic-affirming mental health services and research.
The first paper explores mental health experiences of autistic adults and offers an emerging theory of autistic mental health as “the place where autism, trauma, and mental health meet”. The second paper builds on this with an examination of experiences of support and offers an emerging framework that can be used to inform the development and enhancement of autistic-affirming policies and practices in mental healthcare. The third paper pivots to exploring experiences of online asynchronous participation in research and offers insight into the potential value of this method for increasing accessibility for autistic adults and mental health providers motivated to participate in research.
Together, these papers contribute to scholarly knowledge and offer considerations and implications that can be used by clinicians, researchers, organizations, policy makers, and autistic advocates to inform the development and strengthening of autistic-affirming mental health research and practice initiatives.Arts, Faculty ofSocial Work, School ofGraduat
Learning efficient representations for generalizable segmentation in data-limited scenarios
No full textAdapting visual recognition models to novel categories with minimal supervision is important, especially for dense prediction tasks like segmentation, where collecting pixel-level annotations is costly or impractical. These challenges are greater in data-limited domains and under distribution shifts, requiring models that can generalize with limited supervision. This thesis focuses on learning efficient representations to enable such generalization. We present three key contributions. First, motivated by how humans use global context in visual reasoning, we propose an object-centric global reasoning module that aggregates semantically similar features into latent tokens representing object instances or parts. These tokens interact through a global reasoning mechanism to enhance contextual understanding without requiring additional instance-level annotations. Second, we address generalized few-shot segmentation, aiming to adapt to novel classes from few examples while preserving performance on base categories learned with abundant data. We introduce learnable visual prompts and a causal cross-attention mechanism that contextualizes novel class prompts relative to base classes, while a transductive prompt-tuning strategy leverages unlabeled test images to improve test-time performance. Finally, we investigate training-free, open-vocabulary segmentation using vision-language models (VLMs) and propose an unsupervised entropy-based metric, InfoScore, to automatically select the most effective attention layers for segmentation. Our method generalizes across diverse VLM architectures and shows that minimal visual supervision—such as a single example per category—can significantly improve segmentation by reducing the semantic gap between class names and visual representations. Together, these contributions advance segmentation techniques toward scalable, adaptable solutions for data-limited and low-supervision settings.Science, Faculty ofComputer Science, Department ofGraduat
Flexible capacitive sensor arrays for human interaction and smart manufacturing
No full textThe full abstract for this thesis is available in the body of the thesis, and will be available when the embargo expires.Applied Science, Faculty ofElectrical and Computer Engineering, Department ofGraduat
Comparison of patient-reported outcomes between the Invisalign® palatal expander and rapid maxillary expansion : a prospective clinical trial
No full textObjectives:
To compare patient-reported outcomes between the Invisalign® Palatal Expander (IPE) and the Rapid Maxillary Expander (RME) in children undergoing maxillary expansion, focusing on pain, discomfort, functional impact, and quality of life during the first two weeks of treatment.
Methods:
This prospective clinical trial included 60 children aged 7–12 years (30 treated with IPE, 30 with RME) from five orthodontic clinics. Participants completed structured questionnaires at 1, 7, and 14 days after appliance insertion, assessing pain, discomfort, functional limitations, and quality-of-life factors using visual analogue scales (VAS) and fixed-choice yes-or-no questions. Statistical analyses were conducted using parametric and non-parametric tests, with significance set at P < .05.
Results:
Both groups experienced similar overall patterns of improvement across the 14-day period, with most patient-reported symptoms decreasing to mild or moderate levels by Day 14. The IPE group reported slightly higher discomfort at insertion, while the RME group experienced more functional limitations such as chewing and speech difficulties in the early stages. However, these differences were not consistently statistically significant and not necessarily clinically significant. Both appliances were generally well tolerated after the initial adjustment period, and no serious adverse effects were reported in either group.
Conclusions:
Both IPE and RME are clinically acceptable options for maxillary expansion in children. Although there were some early differences in patient experiences between the two groups, both appliances showed favourable adaptation over time, with no major complications observed. These findings can help guide clinicians in setting realistic treatment expectations and discussing comfort considerations with patients and their families.Dentistry, Faculty ofGraduat
Virtual nodes for graph neural networks : techniques, impact, and theoretical analysis
No full textThis thesis investigates virtual nodes (VNs), a graph augmentation technique that improves the performance of Graph Neural Networks (GNNs). Although VNs have been widely used in practice, their theoretical foundations and systematic evaluation remain limited. To address this gap, we propose the concept of a projection graph, which provides a unified framework for analyzing Message Passing Neural Network+VN, and introduce a taxonomy of virtual nodes along two dimensions: effective density and locality. We review theoretical results on message-passing neural networks (MPNNs) with VNs, showing that even a single VN can enhance expressive power to a level comparable with graph transformers. We further analyze how multiple VNs reduce commute times and mitigate over-squashing, thereby lowering the depth required for long-range message passing. Extending these results, we demonstrate that multiple VNs provide stronger improvements against over-squashing. To validate these insights, we conduct experiments on tasks such as tree-neighbor matching and long-range graph benchmarks (Peptides-func and Peptides-struct). Our empirical study shows that (1) VNs consistently help alleviate over-squashing, (2) connecting VNs via the coloring strategy yields the best performance, and (3) the optimal number of VNs varies across tasks. Together, our theoretical and empirical findings establish VNs as both a practical augmentation and a principled mechanism for addressing structural bottlenecks in GNNs.Science, Irving K. Barber Faculty of (Okanagan)Computer Science, Mathematics, Physics and Statistics, Department of (Okanagan)Graduat
Investigating differences in return to work for workers with work-related chronic pain and concurrent psychological injuries in British Columbia
No full textWork-related chronic pain injuries in British Columbia, Canada, have been on the rise, with a 48% increase in accepted workers’ compensation claims from 2017 to 2022. Similarly, psychological health claims have also risen, with a 51% increase in the number of claims submitted from 2019 to 2023. This growth is expected to continue in the coming years due to changes in workplace dynamics and an aging population. Previous research has shown that these injury types have a bidirectional relationship whereby having one increases an individual’s susceptibility to having the other. This study aims to investigate differences in time to return-to-work (RTW) among psychological, chronic pain, and psychological with chronic pain injuries and to identify the most influential factors affecting the likelihood of RTW for each injury type. Workers’ compensation claims data were used in this study to identify workers (n=414,507) with an accepted time-loss claim due to work-related psychological, chronic pain, psychological with chronic pain, or other injury types occurring between 2012 and 2019, with up to 3 years of follow-up. Descriptive statistics were used to analyze differences in time to RTW. Multivariable logistic regression was used to assess the role of workplace/organizational, socio-demographic, and injury-specific characteristics in determining the likelihood of RTW for each injury type. Results indicate substantial differences in days to RTW events after injury, with psychological with chronic pain injuries having the longest RTW durations (median=1095 days [Interquartile Range (IQR):1095 – 1095 days]), followed by chronic pain (median=1095 days [IQR:327 – 1095 days]) and psychological (median=399 days [IQR:67 – 1095 days]). All other injuries had a
median RTW duration of 21 days [IQR:7 – 83 days]. Characteristics such as younger age, larger firm size, and having modified RTW increased the likelihood of RTW across injury types, while other variables had varying injury-specific effects. These findings suggest the need to modify current RTW programs to include better mental health and chronic pain management systems. Additionally, workers with concurrent psychological and chronic pain injuries warrant further investigation into how they can be further supported to promote earlier RTW outcomes, as over 75% of the cohort did not RTW within three years.Medicine, Faculty ofPopulation and Public Health (SPPH), School ofGraduat
Design of a six degree of freedom motion simulator for advancing sports bra comfort research
No full textEvaluating sports bra comfort in an objective manner is challenging because it is difficult
to reproduce experiments consistently with human subjects. Using a mannequin with ap
propriate breast prostheses and a motion simulator can overcome this limitation. However,
commercially available motion simulators are not optimized for reproducing the specific
combination of range of movement, speed and high acceleration observed during athletic
activity. Achieving this requires a purpose-built motion simulation platform with the right
combination of travel range, speed, and dynamic performance.
A Stewart’s platform motion simulator is particularly well suited for this task. To meet
the specific demands of torso motion simulation, a custom six-degree-of-freedom (6-DOF),
rotary-actuated Stewart’s platform was designed. A parametric torso motion trajectory was
synthesized from motion capture and inertial measurement unit (IMU) data collected in
collaboration with Lululemon and the Human Motion Biomechanics Lab (HuMBL). The
system was optimized from the ground up to replicate the most demanding portions of this
trajectory using impedance matching in the design phase, and the control system was tuned
for maximum dynamic performance.
The resulting platform enables repeatable reproduction of high intensity torso motion and
provides a cost-effective, high performance solution for objective comfort testing of sports
bras.Applied Science, Faculty ofEngineering, School of (Okanagan)Graduat
A lectin-based magnetic CRISPR screening protocol permits rapid discovery of genes regulating cell-surface glycosylation
No full textThe full abstract for this thesis is available in the body of the thesis, and will be available when the embargo expires.Pharmaceutical Sciences, Faculty ofGraduat
The mechanochemical hydrostatic organs of Chaoborus larvae : function under pressure
No full textMany aquatic animals regulate their position in the water column by controlling their buoyancy. This requires body compartments of regulated volume filled with matter that is less dense than water. However, only one group of insects has achieved this: the aquatic larvae of the midge genus Chaoborus (Diptera), which can alter the volume of four internal air-filled sacs. This thesis reveals the unique mechanochemical system underlying air-sac function, including how it is regulated, its mechanical properties, how it converts changes in pH into mechanical work, and how it has evolved to function at depth.
The air-sac wall was found to contain resilin, an insect protein that swells reversibly in response to increases in pH, causing the fusiform air-sacs to expand lengthwise while maintaining a constant radius. Pharmacological investigation demonstrated that the air-sac wall is acidified by proton pumps (VHA) in a surrounding epithelium, shrinking it, while cAMP-signalling mediates alkalinisation and expansion. Finally, disrupting the air-sac epithelium in vivo revealed that the hemolymph has a pH of 7.4 while the air-sacs, without stimulation, are maintained near pH 6, providing the pH range for air-sac operation.
Exposing air-sacs to hydrostatic pressure ramps revealed that those of deeper diving species and older instars could withstand more pressure before failure. Transmission electron microscopy showed that deep diving C. edulis had thicker air-sac walls relative to their radius, compared with two shallow diving species. There was no relationship between failure pressure and air-sac radius within species. In air-sacs, resilin imparts compressibility. The deepest diving species, C. edulis, possessed less resilin per length of wall, while deeper diving species, more generally, possessed stiffer air-sacs.
Production of mechanical work by air-sac resilin was quantified by manipulating both pH and pressure. Dive depth correlated with work production: air-sacs of the deepest diving species, C. edulis, produced over 10-fold more work than the shallowest. With less resilin lengthwise, the greater work output by air-sacs of C. edulis must be due to differences in resilin composition.
This thesis shows that Chaoborus species have adapted the pressure-tolerance and work capacity of their air-sacs to the depth of their aquatic environment.Science, Faculty ofZoology, Department ofGraduat
The risk of cancer and its associated risk factors in patients with glomerular disease : a population-based study.
No full textPatients with Glomerulonephritis (GN) may be at increased risk of developing de novo malignancies due to a combination of chronic immune system activation, tissue inflammation, immunosuppression use and as a consequence of the progression to end stage kidney disease (ESKD). However, there are limited epidemiologic data to accurately quantify cancer risk in GN and to determine the contributions of traditional and GN-specific risk factors including estimated glomerular filtration rate (eGFR), proteinuria and immunosuppressive therapies. This thesis intends to bridge these gaps.
Chapter 1 introduces GN, its background, epidemiology, and the hypothesized relationship with cancer. Chapter 2 presents a narrative review summarizing existing observational studies and highlighting their limitations and methodological challenges. In Chapter 3, we assembled one of the largest cohorts of biopsy-proven GN patients by linking provincial GN registry with the Cancer Registry and other administrative databases spanning twenty years to conduct robust investigation of cancer risk and its associated factors in patients with GN. We showed that patients with GN have a substantially increased risk of cancer compared to the general population. This excessive risk was mostly observed for lymphoma, colorectal, kidney, and lung cancers, and was present both before and after the onset of ESKD, particularly among patients younger than 40 years. In Chapters 4 and 5, we found that cancer risk accumulates gradually with ongoing kidney dysfunction rather than being triggered solely by the onset of ESKD. To study the effect of immunosuppression on cancer risk in patients with GN, we used a data-driven approach to identify the most etiologically relevant exposure metrics for each immunosuppressive agent in relation to cancer risk. High cumulative exposure over years to immunosuppressive therapies, including antimetabolites, corticosteroids, and cyclophosphamide, was associated with increased cancer risk. The dosing thresholds linked to increased risk could be readily reached with conventional treatment protocols.
Findings of this thesis highlight that while immunosuppression remains central to GN management, treatment decisions should involve shared decision-making, with a preference for low-dose agents that are not associated with increased cancer risk when possible, and careful reassessment as cumulative doses approach identified thresholds.Medicine, Faculty ofMedicine, Department ofGraduat