14277 research outputs found
Sort by
Combining minimally invasive technologies to develop and focus a marine monitoring "macroscope"
No full textPublishe
Impact of ad blockers on computer power consumption: a comparative analysis of browser ad on and built-in browsers feature
No full textThe increasing prevalence of online advertisements has driven the widespread use of ad blockers, which aim to enhance user experience by reducing unwanted content. However, the impact of ad blockers on system power consumption, particularly across different hardware configurations and ad-blocking implementations, remains underexplored. This thesis provides a comprehensive investigation into the energy use of ad blockers in various computing environments, including systems with AI accelerators, ARM-based processors, and browsers equipped with built-in ad-blocking features. The research comprises a multi-faceted analysis of power consumption influenced by ad blockers under different hardware and software conditions. First, the study explores the use of ad blockers in systems integrated with AI accelerators, comparing traditional CPU/GPU-based methods to AI-enhanced approaches. It was observed that ad blockers such as uBlock Origin and uBlock Origin Lite, when paired with AI acceleration, achieved significant reductions in power usage and memory consumption, particularly for multimedia-heavy websites.
Further investigation targets the energy implications of ad blockers on ARM-based CPUs, which are widely utilized in mobile and embedded systems. A detailed comparative analysis across multiple browsers—including Chrome, Brave, Vivaldi, and Firefox—revealed that the combination of efficient browsers (like Brave and Kiwi) with lightweight ad blockers (such as uBlock) resulted in an approximate 15% reduction in power consumption compared to traditional setups without ad-blocking features. This emphasizes the importance of the optimal pairing between browsers and ad blockers to achieve enhanced energy efficiency in low-power environments. Additionally, this thesis examines the efficiency of browsers with integrated ad-blockers, such as Brave, Opera, and Librewolf, compared to standard browsers without built-in ad-blocking. Results indicate that these integrated ad-blockers significantly reduce CPU and GPU
workloads, resulting in up to 44% lower power consumption, especially when browsing video-heavy websites. These findings demonstrate the potential of integrating ad-blocking features directly into browser architecture to achieve meaningful energy savings, thereby extending battery life for mobile devices and reducing overall energy demands. Lastly, a comparative analysis of popular ad blockers—AdBlock, AdBlock Plus, Ghostery, uBlock, and uBlock Origin—across different content types and website categories highlighted substantial variations in energy efficiency. The results underscore that while ad blockers generally contribute to reduced power usage, their effectiveness is influenced by factors such as the type of online content and specific ad-blocking technology used. Media-rich websites particularly benefit from the use of lightweight and well-optimized ad blockers, which significantly decrease system resource consumption. Overall, the findings presented in this thesis provide new insights into optimizing power consumption in web browsing through the strategic use of ad blockers. By demonstrating the benefits of different ad-blocking strategies across various computing environments, this research contributes to sustainable computing practices, offering practical guidelines for users, developers, and policymakers aiming to reduce energy consumption and promote efficient digital technologies.Includes bibliographical reference
Propagation dynamics for nonlinear evolution systems in a shifting environment
No full textSpatial and temporal heterogeneities play a crucial role in population biology. In
particular, climate change and intensi ed environmental pressures have highlighted
the signi cance of shifting habitats as a framework for understanding their impact
on population dynamics. As a result, this thesis is devoted to the investigation of
spreading properties and traveling waves for three classes of evolution systems in
shifting habitats.
An overview of the mathematical analysis of propagation phenomena and shifting
habitat problems is presented in the introduction. The main theoretical tools used
in this thesis are presented in the preliminaries. In Chapter 3, we rst consider a
species living in an environment surrounded by two favorable habitats. To describe
its dynamics, we study a nonmonotone integro-di erence equation with a nonsymmetric
dispersal kernel. In Chapter 4, to understand the e ects of climate change
and industrialization on a population with juvenile and adult stages, we propose a
stage-structured model and explore its propagation dynamics in a habitat that shifts
from unfavorable to favorable habitats. In Chapter 5, for species whose juveniles
have more limited dispersal ability than adults, we further investigate a time-delayed
reaction-di usion equation in a time-periodic shifting environment surrounded by two
unfavorable habitats. At the end of this thesis, we present some open problems and
directions for future research
Biomechanical and neuromuscular responses to low back pain
No full textLow back pain (LBP) is a major healthcare burden globally, but causation is still not fully understood. A fundamental aspect to be explored is the relationship between LBP and changes in movement and function. While adaptations have been observed, the direction and mechanisms of this interaction are unclear.
The aim of this thesis was to improve our understanding of the methods used to investigate LBP and movement (Study 1 & 2), and to explore the relationship between LBP and movement/function (Study 3).
Study 1 investigated the concurrent validity of skin-based sensors to measure spine kinematics and found that they should be considered to estimate intervertebral angles only. However, if their limitations are addressed appropriately, they can be a useful tool to monitor spine kinematics. Study 2 explored whether blood or brain markers are associated with heat/capsaicin pain. There were no differences between heat/capsaicin pain and placebo, suggesting an absence of the systemic neurophysiological response seen in clinical LBP. Study 3 explored the relationship between pain-states (pain vs no pain) of two types of LBP (recurrent and heat/capsaicin) and changes in movement/function. We found recurrent LBP to be associated with some changes that dissipate during recovery. However, these changes were not reproduced with experimental pain, indicating that it is not only the experience of pain that induced these changes. Additionally, when comparing outcomes between the recurrent LBP group and controls, differences between populations were found regardless of pain state. This
suggests that the relationship between LBP and changes in movement/function is bi-directional and more complex than a causative one-directional relationship.
Collectively, this thesis contributes to our understanding of current methods to understand LBP, including experimental pain and skin-based spine kinematics. We establish that pain-states in recurrent, but not experimental, LBP are associated with changes in movement and function. Additionally, there are differences between people with recurrent LBP and controls beyond the immediate experience of pain, highlighting the complexity of the interaction between LBP and movement/function. Advancing our methods to assess this relationship further will allow us to better tease out the reasons for the changes found
Predicting survival among colorectal cancer patients using automated body composition
No full textColorectal cancer is among the most prevalent cancers worldwide, yet its current
staging system is limited by variability and paradoxical survival outcomes. Body
composition analysis from routine computed tomography (CT) scans offers predictive
markers of patient survival, such as sarcopenia and adiposity. However, manual segmentation
of CT images is labor-intensive and prone to inter-observer variability. In
this study, we validated the accuracy and reliability of automated CT-based segmentation
using the Data Analysis Facilitation Suite (DAFS) Express, achieving high concordance
with manual segmentation, with mean DICE scores above 96% for all tissue
areas and only 0.10% poor cases across a cohort of 5,973 cancer patients. The DAFS
Express method also showed strong concordance with manual analysis in mortality
association, as indicated by similar hazard ratios, confidence intervals, and p-values.
Furthermore, we developed a deep learning model that combines clinical and body
composition biomarkers to predict survival outcomes in colorectal cancer patients.
Our integrated model achieved a time-dependent concordance index (Concordancetd-
index) score of 0.7298 (p < 0:001), significantly outperforming models based solely on
clinical biomarkers. Models relying exclusively on body composition features had the
lowest Concordancetd-index scores, suggesting that body composition alone may not
be a reliable predictor of survival.Includes bibliographical references (pages 67-76
Career paths and planning: career development for undergraduate students in Manitoban universities
No full textThis qualitative, multi-institutional research study was conducted to better understand the mandate for career development in Manitoba’s public universities, the interpretation of the mandate by public universities, and the effectiveness of the services, supports, initiatives, and programs provided in assisting students with their career paths and planning. The research includes an environmental scan of publicly available documents from the Government of Manitoba and Manitoba’s public universities and semi-structured focus groups with undergraduate students. This study explores the alignment between Manitoban career services centres’ program offerings and students’ engagement with those services and supports throughout their post-secondary education experiences
Exploring postsecondary students' motivations for cannabis use
No full textThis qualitative project explored Canadian postsecondary students' motivations for cannabis use, with a particular focus on the use of cannabis to manage psychological well-being. Participants (N = 14) were emerging adult postsecondary students in Newfoundland and Labrador (NL) who used cannabis to manage mental health, recruited through postings by NL postsecondary institutions. Data were collected using semi-structured interviews and analyzed using inductive thematic analysis.
The first study explored students' motivations for using cannabis and the associated effects. Students described diverse cannabis motives, including relaxation, socializing, experimentation, productivity and therapeutic motives (i.e., managing mental or physical health). Participants also detailed secondary consequences of cannabis, including adverse interpersonal, motivational, psychological and physical health impacts.
The second study explored therapeutic cannabis motives in further detail. Students used cannabis to manage cognitive and emotional states like stress, anxiety, difficult emotions and racing thoughts. Cannabis was also used to manage sleep, pain, nausea and appetite. Students also described factors contributing to increased cannabis use, including changes in routine, worsening health symptoms and academic pressure.
The findings further knowledge of postsecondary students' cannabis motives, which may inform the prevention and treatment of cannabis use problems. Postsecondary students require further education on the health impacts of cannabis use and alternative coping strategies for managing stress and psychological well-being
Structural study of asphalt binder by X-ray diffraction ( XRD) and Peak fitting
No full textA study of asphalt binder samples was conducted using x-ray diffraction (XRD). XRD
patterns were obtained using a Rigaku` D/Max 2200 V-PC, with Jade software
(version 6.1). The profile fits and full width at half-maximum (FWHM) were obtained
using Pearson VII, pseudo-Voigt functions and the Generalized Fermi Function
(GFF). Aromaticity and crystallite parameters for the asphalt binders were calculated
using the data from the three functions (Pearson VII, pseudo-Voigt, Generalized
Fermi Function GFF). In comparison to aromaticity, results for the crystallite
parameter were strongly correlated with variations in Pearson VII exponent and
pseudo-Voigt Lorentzian.
The results reveal clear correlations among Pearson VII, Pseudo-Voigt, and
Generalized Fermi Function (GFF), with regard to the aromaticity, and crystallite
parameters. Also, the results indicated that XRD diffraction was useful for
determining the compositional and structural characteristics of asphalt binders which
is important for understanding the aging of asphalt binders and pavement
Wavefronts and speed selection of spatiotemporal biochemical models
No full textThis thesis is devoted to the study of wavefronts and speed selection of spatiotemporal biochemical models.
We first consider a model of precursor and differentiated cells, with the appearance of non-isolated equilibria on a line in the phase space. This system exhibits transition properties, shifting from monostable to bistable and from monotone to non-monotone, through changes in the impacting parameter.
We establish the existence of traveling waves in the monostable monotone case, the monostable non-monotone case, and the bistable monotone case, and derive some propagation properties of the traveling waves for each case.
The critical degenerate case in this precursor and differentiated cells model exhibits fundamentally different and significantly more intricate dynamics. We here develop new ideas to prove the existence of traveling waves by a continuation argument via perturbation in a weighted functional space.
To extend the above ideas to more complex models, we investigate the Belousov-Zhabotinsky system with non-local delayed interaction, which has non-isolated equilibria and exhibits a dynamical transition structure from bistable to monostable.
The existence and uniqueness of bistable traveling waves, as well as the existence of monostable waves in the degenerate case, are first proved. In addition, we derive some necessary and sufficient conditions for speed selection and show how parameters impact speed selection results under the monostable case.
Climate change has great impacts on the survival of biological species. Lastly, we consider a reaction-diffusion equation with a shifting environment, where the reaction function shifts from a favorable to an unfavorable environment. The exponential stability of three types of forced waves in this model is proved. Furthermore, numerical simulations are carried out for each model, which match with our theoretical results.Includes bibliographical references (pages 159-163