30921 research outputs found
Sort by
Sea ice transport and melt, and the loss of multiyear sea ice in a changing Beaufort Sea
No full textSea ice is a complex medium that covers up to 15 million km^2 of the northern hemisphere annually; presenting a vast highly reflective surface that cools the global climate, providing a variety of habitats for different Arctic species, and both facilitating on-ice travel for local Inuit while also limiting maritime access to Arctic waters. However, anthropogenic warming is amplified by four times in the Arctic which has driven a dramatic reduction in the extent and thickness of the Arctic ice pack and is projected to render the Arctic Ocean seasonally ice-free by the middle of this century. Underlying the loss of sea ice has been a dramatic transformation in the composition of the ice pack from a predominantly multiyear ice (MYI) cover to an inherently thinner and less resilient seasonal ice cover. Historically, the anticyclonic Beaufort Gyre retained sea ice for years, allowing it to thicken while aging, and distributing MYI throughout the Arctic Ocean. However, increasing melt rates in the Beaufort Sea have interrupted the once continuous journey of sea ice through the Beaufort Gyre, cutting off the redistribution and retention of MYI and thereby significantly contributing to the pan-Arctic transition towards a seasonal ice cover. Within this thesis, I use a combination of in situ and remotely sensed observations of sea ice to examine sea ice loss in the Beaufort Sea and its impact on MYI transport and retention within the Beaufort Gyre. In particular, I will examine how the state of the Beaufort Gyre during winter preconditions the regional ice pack for the melt season, and how preconditioning is playing a greater role as the now thinner ice pack is more mobile. I will then use a novel box model to quantify MYI loss in the Beaufort Sea and examine the associated changes in MYI transport through the Beaufort Gyre. Finally, I will set the broader context of MYI loss in the Beaufort Sea by examining MYI loss across the Arctic Ocean and the relative contributions of export, melt and replenishment, the three factors which collectively dictate the balance of MYI. From this I can speculate on how MYI in the Arctic Ocean will evolve to the point where it will one day cease to exist in a seasonally ice-free Arctic Ocean.May 202
Limitations in effective treatment of Parkinson’s Disease: neuroanatomical substrate of L-Dopa induced dyskinesia and cognitive impairment
No full textParkinson’s disease (PD) is the fastest growing neurodegenerative disease, and in an increasingly aged population the care and management of PD is an increasing global concern. PD symptoms are managed well by L-DOPA treatment in the early stages of the disease, however the efficacy is reduced as the disease progresses due to an increase in troublesome L-DOPA induced dyskinesias (LID), as well as the development of cognitive decline.
We investigated the neurophysiological and neuroanatomical substrate of LID in a 6-OHDA lesioned rat model of PD, showing that chronic L-DOPA treatment induces an exaggerated vasomotor response in LID animals. Remodeling of the microvasculature, on the other hand, is dose dependent and not evident in animals with low-dose progressive onset LID. An L-DOPA induced increase in relative cerebral blood flow (rCBF) in the dorsolateral striatum is evident in LID animals, but not in stable L-DOPA responding animals. When we measured L-DOPA induced changes in relative cerebral metabolic rate (rCMR), we observed key differences in LID and non-LID animals. L-DOPA reduced striatal rCMR in non-LID animals both from the first dose and after 21 days of treatment, consistent with the theory that L-DOPA therapeutically reduces striatal hyperexcitability. Conversely, L-DOPA failed to show consistent reduction in rCMR in LID animals, and after symptoms had developed, L-DOPA markedly increased rCMR in the striatum. Our findings support the idea that plastic changes in striatal excitability underly the expression of LID symptoms, and that these changes may be initiated in L-DOPA naïve animals.
We investigated the use of a supervised machine learning algorithm called Support Vector Machine (SVM) to retrospectively stratify patients based on brain fluorodeoxyglucose (FDG) –PET. The baseline scans were used to train a model which separated PD patients with mild cognitive impairment (MCI) as dementia converters vs. stable MCI with 95% sensitivity and 91% specificity. The model retained an accuracy of 73% in an external testing set. The SVM model was topographically characterized by hypometabolism in the temporal and parietal lobes and hypermetabolism in the anterior cingulum, putamen, insular, mesiotemporal, and postcentral gyrus. These results indicate that FDG-PET-based SVM classifier has utility for predicting the cognitive prognosis of PD-MCI patients.October 202
Near infrared spectroscopy for the non-invasive characterization of cerebrovascular reactivity in moderate and severe traumatic brain injury
No full textThe management of traumatic brain injury (TBI), in the critical care setting, centers around the reduction of ongoing secondary injury. Disruption in the brain’s ability to control its own blood flow, known as cerebrovascular reactivity (CVR), has been identified as a key contributor to ongoing secondary injury following TBI. Contemporary methods of monitoring CVR continuously rely on invasive intracranial pressure (ICP) monitoring. This limits the use of these ICP-based indices of CVR to the acute phase of TBI where ICP monitoring is already indicated. As a result, continuous examination of CVR has not previously been performed in the chronic phase of TBI or in healthy subjects.
Near infrared spectroscopy (NIRS)-based indices of CVR provide a less invasive alternative. They have been found to perform similarly to ICP-based indices, however, their examination in the clinical setting has been limited. In this thesis, the utility of NIRS-based indices of CVR following moderate-to-severe TBI was extensively examined.
The relationship between ICP- and NIRS-based indices of CVR was examined through machine learning and time series analysis techniques. Further, a novel, entirely non-invasive technique of measuring NIRS-based CVR was developed. This technique was employed to examine CVR in both healthy subjects and through the chronic phase of recovery from TBI. This allowed for insights to be gleaned into how CVR differs in these settings. Finally, the outcome association of NIRS-based CVR in the acute and chronic phase of TBI was studied.
This thesis provides a solid foundation for the further exploration of NIRS-based indices of CVR in TBI. Further, it opens the door to the continuous examination of CVR in pathophysiologic states outside of TBI.February 202
Dynamic network and data science applications in finance, security and genomics
No full textData Science studies are prolific in many application areas from health to finance and from supply chain to computer network security with varying objectives. This thesis is focusing on pattern mining and network analysis of data in some of these application areas. In the study of pattern mining, our focus is to compute waiting time in observing a desired pattern in three different application areas: patterns in DNA sequences, patterns in unauthorized access to computer systems and patterns of price rise and drops in stock prices. A novel fuzzy transition probability (TP) matrix is introduced, and a novel pattern mining algorithm is proposed for sequence data of any length. The proposed algorithm, which avoids the inversion of the pattern matrix, is applicable to Markov chains with huge state spaces. Furthermore, it is illustrated with real data that the incorporation of fuzzyness to the transition probability matrix is crucial in obtaining realistic forecasts. In the second study, we propose a novel method based on financial networks and their PageRank scores to form pairs to apply in pairs trading. In order to illustrate the practical performance of the proposed methods, algorithmic trading profits using the commonly used cointegration method are compared with the profits made by proposed correlation-based financial networks. The proposed method offers an advantage over the commonly used cointegration method by identifying more profitable trading pairs (stocks) for pairs trading.October 202
Raising disabled children: The perspectives of caregivers in - Bosomtwe, Ghana
No full textBackground: Existing literature on raising disabled children in Ghana indicates that caregivers are faced with numerous challenges, some of them have adopted strategies in dealing with challenges that come with their work.
Purpose: This thesis seeks to investigate the socio-cultural barriers faced by caregivers of children with disabilities in the Bosomtwe District of Ghana.
Method: This study utilized a qualitative descriptive approach to explore the socio-cultural barriers faced by caregivers of disabled children in the Bosomtwe District of Ghana. Purposive sampling was used to select a sample of 7 caregivers, each with a child from one of the three main impairment groups (visual, hearing, and physical). Interviews were conducted via virtual platforms such as Zoom or via telephone with the use of an interview guide. Data analysis was performed using NVivo Qualitative Data Analysis software.
Findings: This study found caregivers of disabled children in Ghana face barriers including lack of inclusive education, financial constraints, limited healthcare, transportation issues, and negative attitudes. To cope, caregivers networked, educated themselves, advocated for children, sought non-governmental organization (NGO) support, and remained resilient. Findings provide insights into caregiver challenges and resilience, informing supports needed.
Conclusions: This study illuminated the challenges Ghanaian caregivers face through first-hand accounts. Findings underscore the need for greater formal and community supports. Further research on caregiver perspectives is critical to advocacy efforts.N/AFebruary 202
Experiences of Indigenous people with bariatric surgical care in Manitoba
No full textBackground
Obesity and type 2 diabetes mellitus (T2DM) are growing global health concerns associated with significant morbidity, mortality and increasing healthcare expenditures. Indigenous peoples are at higher lifetime risk of both and poorer health outcomes. This work aimed to explore the experiences of Indigenous who had undergone bariatric surgery.
Methods
We established relationships with Indigenous community leaders. A mixed methods scoping review of experiences and outcomes of Indigenous patients undergoing bariatric surgery was conducted. Guided by an Indigenous Elder, we gathered knowledge through Sacred sharing circles, ceremony and Traditional teachings in a decolonized way.
Results
Scoping review found Indigenous patients have poorer access to bariatric surgery with similar weight loss outcomes and strong motivators for pursuing bariatric surgery. Relationship building, community involvement, and honoring tradition are crucial when conducting research with Indigenous communities. Indigenous people undergoing bariatric surgery in Manitoba had positive experiences, strong motivators, and felt that more cultural supports were needed.
Conclusion
Bariatric surgery is an effective treatment for obesity and T2DM. Research with Indigenous communities to close gaps in health outcomes must be done in a good way, rooted in Indigenous methodologies. Indigenous patients have strong motivators for pursuing surgery, and have a desire for non-surgical, culturally relevant supports along the bariatric pathway. Culturally sensitive care is necessary for Indigenous patients in bariatric clinic settings.May 202
The effect of exclusive breastfeeding, exclusive bottle-feeding, mixed feeding, and oral Rothia on early childhood caries
No full textObjective: To investigate the association between feeding practices (exclusive breastfeeding, exclusive bottle-feeding, and mixed) with early childhood caries (ECC) and the association between the prevalence of oral Rothia, at the species and genus level, with ECC.
Methods: A total of 438 children (178 caries-free, 260 with ECC) were included in this case-control study. A comprehensive questionnaire that included a section on feeding history, was completed by the parent or caregiver of each participant. Based on the feeding history responses, participants were classified as either exclusively breastfed, exclusively bottle-fed, or mixed (both bottle-fed and breastfed). A supragingival plaque sample was collected from each participant and then assessed with 16S rRNA sequencing for Rothia genus and species’ relative abundances. The association between feeding practices and Rothia prevalence with ECC, accounting for confounding variables, was analyzed using multivariable logistic regressions.
Results: A low relative abundance of Rothia aeria was significantly associated with ECC (p0.05). While the odds of association between a low relative abundance of Rothia (genus) and ECC was 1.63, this relationship failed to reach the threshold of significance (p=0.054). Feeding practices were not significantly associated with ECC after adjusting for confounding variables. However, bedtime bottle use, bedtime snacking, living in a rural/remote area, being an older age, the use of fluoridated toothpaste, and having less-educated guardians were significantly associated with ECC (p<0.05)
Conclusions: Findings in this study suggest that low relative abundances of R. aeria and Rothia (genus) are associated with ECC. Feeding practices were not key risk factors for ECC in this sample, as other factors related to the social environment and bedtime habits were more strongly associated. Since the current literature is limited, prospective studies examining Rothia and feeding practices with ECC would be beneficial to understand the causal relationship between Rothia and ECC and to help reach a consensus on the effect of different feeding practices on ECC risk.October 202
Making home: a multi-family housing complex for Yazidi refugees in Winnipeg
No full textThis practicum project addresses the housing challenges that refugee populations face when settling in a foreign country after fleeing from genocide, war, and unrest in their home countries. The mixed-use transitional housing facility designed in this project, Mala, fosters Yazidi refugees in a community-like setting and it serves to encourage families to support one another as they settle into a new, and extremely different, way of life.
Transitional housing models and theories surrounding trauma-informed design practices, Topophilia, and Mise-en-scéne, have informed the programming, design, and placement of Mala in Winnipeg’s downtown neighborhood as a hub for the Yazidi community of Winnipeg. By creating an opportunity for refugee families to reside in multi-family settings where they can remain engaged socially and help one another through the assimilation process, this project adopts a progressive method for reimagining the settlement process for refugees arriving in Winnipeg, Canada.February 202
Investigating the role of ADORA2B receptors in pyroptosis of inflammatory macrophages through gene knockdown and overexpression
No full textAdenosine receptors (ARs) regulate the immune response by activating pro- and anti- inflammatory cascades through mechanisms that are not completely defined. Macrophages also play a central role in the regulation of inflammation from initiation to resolution. ARs on macrophages have been shown to direct major functions including macrophage cell death. Pyroptosis is a highly immunogenic cell death induced by inflammasomes that cause cleavage of caspases and gasdermins. This process leads to the release of inflammatory cytokines Interleukin 18 (IL-18) and Interleukin -1β (IL-1β). The canonical pathway that leads to pyroptosis involves the NLRP3 (nucleotide-binding domain, leucine-rich–containing family, pyrin domain–containing-3) inflammasome, caspase-1 and gasdermin-D (GSDMD), though other pathways exist. Recently, our lab found that knocking down the Adenosine 2B (A2B) receptor renders macrophages susceptible to pyroptosis. In this thesis, we seek to confirm the role of the NLRP3 inflammasome specifically in pyroptosis observed after A2B receptor knockdown, and whether lost receptor-protein interactions promote this pathway. Results show that in A2B receptor deficient macrophages, NLRP3 inhibition was able to prevent IL-1β secretion and GSDMD cleavage. However, addition of adenosine to macrophages induced GSDMD cleavage that was not sensitive to NLRP3 inhibition. Furthermore, preliminary experiments showed that caspase-4 was activated by adenosine. Proteomic analysis was used to study receptor-protein interactions and pyroptotic proteins were identified in immunoprecipitation assays. A2B coimmunoprecipitation in epithelial cells and macrophages show that A2B receptors enrich and interact with cellular communication, calcium signaling and gap-junction proteins as well as Toll-like receptor signaling, and serine proteases. Results from proteomics and gene ontology analysis have identified possible alternative pathways through which the A2B receptor might influence macrophage pyroptosis.October 202
Machine learning and data science application for financial price prediction and portfolio optimization
No full textThis thesis explores interconnected advanced machine learning (ML) and data science (DS) methodologies for improved predictive accuracy in financial markets and resilient portfolio optimization.
Studying the literature on ML/DS methodologies extensively led us to observe a significant lack of application of these advances, such as autoencoder (AE), recurrent neural networks (RNN), etc. in the finance industry. The novelty of this thesis is to study price prediction and portfolio optimization with RNN and AE algorithms. Furthermore, unsupervised ML strategies were studied to introduce robustness in portfolio optimization. For this purpose, two innovative encoder-decoder-based RNN architectures autoencoder-based gated recurrent unit (AE-GRU) and autoencoder-based long short-term memory (AE-LSTM) were proposed, which were shown to be effective in predictive efficacy across diverse asset types and market conditions, showcasing enhanced predictive accuracy for financial assets. Various DS concepts, such as data visualization, Bollinger bands, data-driven volatility estimates, unsupervised ML, etc. were integrated while implementing and experimenting with new architectures for price prediction and portfolio optimization.
The proposed models in this thesis showed effectiveness in price prediction and portfolio optimization under varying market conditions. The study also highlights the benefits of diversified portfolios by proposing a novel DL-based model for portfolio construction, especially when coupled with affinity propagation (AP) clustering and appropriate data-driven risk measures based on volatility estimates - with sign correlation (VES) and volatility correlation (VEV).
Traditional models optimize portfolio weights using objective functions, while recent innovations emphasize data-driven risk measures for minimum risk weights from random samples. Despite challenges with short-term data featuring negative mean returns, the proposed ML-based diversification approaches (for both traditional and data-driven PO) identified portfolios with positive returns by clustering assets with high mean returns.
In summary, this thesis introduces novel hybridized ML/DL models for price prediction and approaches to enhance diversification within specific types of portfolio optimization through advanced clustering methods like affinity propagation and DBSCAN. Diversified stock selection with clustering techniques significantly enhances profitability in data-driven portfolio optimization. These studies underscore its critical role in portfolio resilience, showing that optimal asset selection across different clusters is pivotal for portfolio performance under challenging market conditions.Funding from supervisor's NSERC Discovery grant.October 202