Filter Results
21751 results
Objectives This dissertation had two overarching objectives: 1. To determine how stakeholders perceive women’s preferences for cervical screening modalities. 2. To understand methods to measure women’s cervical screening preferences, to inform the development and testing of a person-centered, evidence-informed approach to preference-elicitation. Methods The overarching conceptual framework was the Ottawa Decision Support Framework. The first objective was addressed by interview studies with (1) guideline developers and program managers and (2) health professionals and women considering screening. This was complemented by a systematic review of quantitative, qualitative and mixed-methods studies of women’s cervical screening preferences, using the Grading of Recommendations, Assessment, Development, and Evaluation approach to developing preference-based recommendations. This approach was also used in a systematic review of methods to elicit women’s preferences, addressing the second objective. These findings led to the development and field testing of a preference-elicitation tool using International Patient Decision Aid Standards criteria, and the development of a protocol for a population-based study of women’s preferences. iv Results Objective 1 Experts disagree about whether there is enough evidence to include alternative modalities in cervical screening programs. Women and health care professionals do not recognize that women face a choice to participate in cervical screening. A narrative synthesis of relevant literature presented challenges in aggregating preferences across diverse study objectives, designs, and contexts. Objective 2 Preference-elicitation approaches for cervical screening are heterogenous in design and rigour. I therefore developed and field tested a tool to elicit women’s preferences, which demonstrated that women found the tool helpful to identify their preferences. I then propose a study that uses multiple methods to apply the tool more broadly. Conclusions Synthesized preferences data might not be the optimal approach to incorporate preferences into cervical screening guidelines. A tool grounded in shared decision-making can help women identify their informed, values-based screening preferences.
Data Types:
  • Other
Plant homeodomain finger protein 6 (PHF6) is a chromatin adaptor protein structurally defined by its two zinc-knuckle-atypical PHD (ZaP) domains. This structural configuration mediates its interaction with dsDNA, miRNA, the nucleosome remodeling and deacetylase (NuRD) complex and regulators of rDNA transcription (Upstream binding factor (UBF) and RNA polymerase-associated factor 1 complex (Paf1C)), ultimately facilitating its role as a chromatin adaptor protein and regulator of gene expression. Mutations in the gene are implicated in Börjeson–Forssman–Lehmann syndrome (BFLS), a rare X-linked intellectual disability disorder characterized by large ears, truncal obesity, and long tapering fingers. BFLS is primarily caused by missense and nonsense mutations while deletions, frameshifts and mutations disrupting the structural integrity of the ZaP domains have been described in T-cell acute lymphoblastic leukemia (T-ALL) and acute myeloid leukemia (AML) patients. To provide more insight into PHF6 and BFLS, we have generated a Phf6 transgenic mouse line with a patient-related nonsense mutation (R342X). We show that the mutation drastically reduced Phf6 transcript levels and produced a truncated protein at very low levels in the developing brain. Mice were born at normal Mendelian ratios but mutant mice were significantly smaller than control littermates. Volumetric analysis of the brain via high resolution MRI revealed increased sizes of the amygdala, periaqueductal gray, and hypothalamus, and decreased volumes within the striatum, hippocampus and cerebellum. Studies of the pituitary gland revealed a postnatal defect in the growth of the anterior pituitary but not the posterior or intermediate regions. This change was reflected in altered expression levels of several hormones in the hypothalamic-pituitary-adrenal axis. Preliminary behavioral tests highlighted deficits in the anxiety and depression response of the mutant mice. Additional studies to fully characterize these mice are ongoing.
Data Types:
  • Document
The Upper Niger Basins (UNB) and the Inner Niger Delta (IND) are integral parts of the Niger River Basin, which flows through 10 countries and constitutes the third longest river in Africa. Natural climate variability and human interventions are two major factors affecting the hydrological regime in the UNB and IND. This study focuses on the later factor, by assessing the hydrological impacts of key existing and planned manmade structures and irrigation schemes in the UNB: the Sélingué (existing dam in Mali), four variants of the Fomi/Moussako dam (planned in Guinea), and Office du Niger (irrigation scheme located in Mali). The Fomi /Moussako dam will be located in the headwaters of the UNB and therefore, is expected to alter the hydrological regime in large parts of the watershed. Expected impacts include a reduction of the flood peak which will adversely affect critical ecosystems in the IND, and higher flows directly downstream of the dams in the dry season to sustain irrigation. These higher flows will, however, be consumed by Office du Niger irrigation scheme, leading to possible severe water shortages downstream of the irrigation scheme and in the IND. This is likely to affect the Malian economy and the poorest parts of its population, as the IND is crucial for the socio-economic and ecological preservation and development of the population surrounding it. The hydrological impacts of the dams and the irrigation scheme were evaluated in this study by developing a model of the IND and UNB using SWAT (Soil and Water Assessment Tool). After the model was calibrated, the effects of the dams and the irrigation scheme on selected flow statistics (mean and standard deviation) were determined at fourteen hydrological stations. In general, the results have shown that (1) the Fomi/Moussako dam will noticeably reduce the downstream high flows, and reduce the average flow; (2) if the Fomi/Moussako dam was to be built, the alternatives with the least storage volume (Moussako 388.5') will have the least impacts on the downstream flows. To assist in related decision making for various users, a Decision Support System (DSS) was also developed. The goal of the DSS is to help users analyze the effects of dams and irrigation on the flow regime by performing a comparative analysis (presence and absence of dams and irrigation in the river). A number of potential adaptation measures were also proposed.
Data Types:
  • Document
Violence against Indigenous women is a major public health concern worldwide and Canada is no exception. Multiple forms of violence inform the broader context of violence against Indigenous women. Nurses are likely to encounter Indigenous women in a variety of settings, but evidence suggests that nurses may lack understandings of violence. This thesis explored the following question: How does extant qualitative research conducted in Canada, contribute to understanding the health and wellbeing of First Nations, Métis and Inuit (Indigenous) women who have experienced violence? During the development of this thesis, significant gaps were highlighted including underrepresentation of Inuit women in the literature, limited focus on health promotion, and lack of methodological approaches to systematic reviews that were participatory and inclusive of the community. Therefore, a secondary aim of this thesis was to privilege perspectives of Inuit women and their communities, by developing a study protocol for a collaborative and community centered approach to reviewing and assessing the extant literature. A configurative and inductive approach based on thematic synthesis was used to systematically search, retrieve, analyze and synthesize extant literature. Post-colonial feminist theory and intersectionality were used as theoretical lenses to emphasize intersections between multiple forms of violence and locate the problem within the broader context of colonization and oppression. Sixteen studies were included in this review, fifteen qualitative and one mixed methods study. Four themes with subthemes emerged based on analysis and synthesis of findings in the included studies: 1) ruptured connections between family and home, 2) that emptiness… my spirit being removed, 3) seeking help and feeling unheard, and 4) a core no one can touch. These themes represent interconnected pathways that influenced health among Indigenous women, and have implications for healthy public policy, clinical practice, and nursing education.
Data Types:
  • Document
Although the concept of fractional calculus was known for centuries, it was not considered in engineering due to the lack of implementation tools and acceptable performance of integer order models and control. However, recently, engineers and researchers started to investigate the potentially high performance of fractional calculus in various fields among which are acoustics, conservation of mass, diffusion equation and specifically in this thesis control theory. The intention of this thesis is to analyze the relative performance of fractional versus integer order PID controller for a quadcopter. Initially, the dynamics of the quadcopter is presented with additional consideration of the ground effect and torque saturation. Then, are introduced the concept of fractional calculus and the mathematical tools to be used for modeling fractional order controller. Finally, the performance of the fractional order controller is evaluated by comparing it to an integer order controller.
Data Types:
  • Document
The non-sustainable nature of fossil fuels as feedstocks for valuable chemicals, combined with the environmental damage caused by their extraction and combustion, increases the need for the development of a bio-based economy. While industry and public opinion are slowly shifting towards acceptance of this change, efficient technologies for the depolymerization and subsequent separation of lignocellulosic biomass fall short of the ever-increasing demand. In particular, there are currently no efficient, sustainable mass scale methods to convert lignin, the most abundant source of aromatic molecules on Earth. The use of oxovanadium(V) catalyst complexes to aerobically cleave C‒C bonds has been demonstrated previously and remains an attractive option for incorporation into a sustainable bio-based economy. Two new triphenoxyamine oxovanadium(V) catalysts with reduced steric bulk and electron density at the metal center (vs. previously reported complexes) have been synthesized for aerobic oxidative diol C‒C bond cleavage. These complexes were found to cleave less activated and more complex substrates than previous generations, including cyclic diols and polyalcohols. Several insights into the reaction pathways of this class of complex were elucidated through a series of kinetic studies. Experimentally, the rate of C‒C bond cleavage of both pinacol and hydrobenzoin was determined to be unaffected by substitution of the O‒H bonds with deuterium, suggesting that currently proposed mechanisms need to be revised. Multiple catalytic regimes were observed during anaerobic reaction, which were not altered significantly by the brief addition of O2. A series of density functional theory calculations revealed a plausible mechanism for the trialkoxy complex that did not involve a proton transfer in the rate determining step, instead suggesting that ligand-arm dissociation-reassociation play a significant role in the reaction. In a second project, new bisphenoxyamine-N-appended base ligand with less steric hindrance and electron density at the metal center, has been synthesized utilizing similar design principles gained from work with triphenoxyamine catalysts. When reacting with lignin model compound 1,2-diphenyl-2-methoxyethanol, this new complex displays a higher selectivity towards aldehydes and esters (relative to previous bisphenoxyamine-N-appended ligands), leading to a higher rate of C‒C bond cleavage. Investigations into the mechanism of bisphenoxy complexes, as well as the role of the N-appended base in reactivity, were performed using substrate pre-complexed bisphenoxy compounds. Thermolysis at 60 and 100 °C produced almost exclusively oxidative C‒H bond cleavage product benzyl methyl ether, with evidence for overoxidation product benzoic acid observed. Thermolysis of labelled substrate pre-complexed revealed that N-appended base may impede C‒C cleavage of 1,2-diphenyl-2-methoxyethanol by forcing the methyl ether away from the oxovanadium(V) center. Through the use of these multidentate phenoxyamine ligands, advances have been made towards sustainable oxovanadium catalysis in the pursuit of efficient and selective lignocellulosic disassembly for a sustainable bio-based economy.
Data Types:
  • Other
Agriculturalists in Lebanon are exposed to a wide range of vulnerability factors that have direct impacts on farmers’ livelihoods. To evaluate the effects of those factors and the livelihood challenges they breed, this research analyses how two agricultural activities shape farmer livelihood vulnerability, namely urban agriculture and fruit and vegetable value chains. First, I analyze how vulnerability factors influence urban agriculturalist livelihoods and assess if urban agriculture is an adequate solution to lower their livelihood vulnerability. Second, I analyse how various actors partaking in fruit and vegetable value chains are exposed to different vulnerability factors and how this impacts their individual livelihoods. Conceptually, the urban agriculture component of this research builds from the vulnerability framework and sustainable livelihoods approaches scholarships. The agricultural value chain section engages with literatures centered on the vulnerability framework, sustainable livelihoods approaches and value chain analysis. This thesis concludes that the main vulnerability factors associated with urban agriculture relate to physical, financial and human capitals, and that urban agriculture is not an adequate tool to reduce the livelihood vulnerability of urban agriculturalists. I also conclude that social, financial and human capital barriers significantly affect agricultural value chains actors’ livelihoods. Overall, the vulnerability of the different actors is linked to their socioeconomic status, which dictates the amount of human capital they possess, thus their ability to adapt to changing conditions and external stressors. I posit that human capital is key to both urban agriculture and agricultural value chains, as this asset dictates the vulnerability of individual livelihoods and Lebanese agriculturalists’ ability to sustain their livelihoods.
Data Types:
  • Document
At the neuromuscular junction, nicotinic acetylcholine receptors (AChRs) convert chemical stimuli into electrical signals. They are heteropentameric membrane protein complexes assembled from four evolutionary related subunits (two α subunits, and one each of the β-, δ-, and ε-subunits), arranged around a central ion-conducting pore, which is regulated by the neurotransmitter acetylcholine. Understanding how the binding of acetylcholine leads to channel opening is of fundamental importance. While it is known that channel opening results from a global conformational change involving the cooperative action of all five subunits, how the subunits achieve this cooperativity is unclear. Our hypothesis is that this subunit cooperation is maintained through coevolution of the subunits, and thus studies of subunit coevolution can provide insight into subunit cooperativity. Using an ancestral reconstruction approach, combined with single-molecule patch clamp electrophysiology, we have begun dissecting the mechanistic consequences of preventing coevolution of the acetylcholine receptor β-subunit. This approach has allowed us to identify new amino acid determinants of acetylcholine receptor function.
Data Types:
  • Other
In this thesis, learning refers to the intelligent computational extraction of knowledge from data. Supervised learning tasks require data to be annotated with labels, whereas for unsupervised learning, data is not labelled. Semi-supervised learning deals with data sets that are partially labelled. A major issue with supervised and semi-supervised learning of data streams is late-arriving or missing class labels. Assuming that correctly labelled data will always be available and timely is often unfeasible, and, as such, supervised methods are not directly applicable in the real world. Therefore, real-world problems usually require the use of semi-supervised or unsupervised learning techniques. For instance, when considering a spam detection task, it is not reasonable to assume that all spam will be identified (correctly labelled) prior to learning. Additionally, in semi-supervised learning, "the instances having the highest [predictive] confidence are not necessarily the most useful ones" [41]. We investigate how self-training performs without its selective heuristic in a streaming setting. This leads us to our contributions. We extend an existing concept drift detector to operate without any labelled data, by using a sliding window of our ensemble's prediction confidence, instead of a boolean indicating whether the ensemble's predictions are correct. We also extend selective self-training, a semi-supervised learning method, by using all predictions, and not only those with high predictive confidence. Finally, we introduce a novel windowing type for ensembles, as sliding windows are very time consuming and regular tumbling windows are not a suitable replacement. Our windowing technique can be considered a hybrid of the two: we train each sub-classifier in the ensemble with tumbling windows, but delay training in such a way that only one sub-classifier can update its model per iteration. We found, through statistical significance tests, that our framework is (roughly 160 times) faster than current state of the art techniques, and achieves comparable predictive accuracy. That being said, more research is needed to further reduce the quantity of labelled data used for training, while also increasing its predictive accuracy.
Data Types:
  • Other
With the significant improvement in IoT technology and smart devices, data collection and distributed computation have led a foundation for Mobile crowd-sensing (MCS). MCS utilizes the capabilities of embedded sensors in smart devices for gathering data. MCS benefits both data provider (participant/user), and data requester, i.e. data providers via incentives/rewards, data requesters by delivering required data. Apart from the benefits gained through acquiring data, confronting challenges such as participant privacy, data trustworthiness, malicious attacks (from illegitimate users) need to be addressed to build robust and reliable data solicitation. In addition to that, it is necessary to consider user motivation and user preference, comfort during its engagement in crowd-sensing. User preferences/constraints can be due to privacy concerns in terms of location, the sensitivity of data or energy usage and many more. With this in mind, the main contributions of the thesis can be listed as follows. 1) We design user selective trustworthy data acquisition frameworks. We introduce a variety of user selection criteria to form participant communities based on participants reliability and income. To evaluate the trustworthiness of our selective reputation-based data acquisition, we consider malicious users in the environment and calculate the total rewards given to malicious users. Simulations results show that community formation based on the acquired income of participants ended up with a substantial loss to the cloud platform as well as participants. Contrary to that, reputation-based community formation has shown nearly equal platform utility (profit), negligible loss of user utility compared to benchmark Non-selective data acquisition with 7% malicious probability. 2) Moreover, we attempt to enable users to modify (allow/deny access to) their builtin sensor set according to their comfort levels. We formulate three comfort levels high (only allow access to sensors that would not directly reveal personal identity such as accelerometer, light sensor, etc.), moderate (obstruct access to sensitive data, e.g. camera), zero comfort (allow access to all users). We introduce Static modification, where users pre-arrange their sensor set before the start of data collection. Our feasibility study shows that pre-arrangement of the sensor set favours user comfort, user utility at the cost of loss in platform utility and performs better than selective reputation-based recruitment for the considered settings. 3) We apply Adaptive sensor modification on top of pre-arrangement of sensor set through which participants are authorized to re-arrange their sensor availability based on reliability scores. Simulation results show that the Adaptive comfort-aware approach performed better than static in terms of platform utility and achieved comparatively better user comfort with reasonable loss in user utility.
Data Types:
  • Document