95038 research outputs found
Sort by
Does a Flywheel Muscle Activation Protocol Enhance Sprint Swim Performance?
Get PDFThe purpose of this study was to determine if flywheel muscle activation (FLYMA) can enhance sprint swimming performance. Fifteen swimmers meeting standards of <23.5s (males)/<25.5s (females) for best 50-yard freestyle race were familiarized before undergoing two testing sessions which involved a 50-yard freestyle trial preceded by a 30-minute rest interval and swimming only or swimming and FLYMA. Time splits and velocities at each 5-yard increment, starting performance, and technique were determined using touchpads and video recordings. There were no significant differences in overall time or any time split, velocity, or starting metric between conditions. Subjects that improved following FLYMA performed +1.5 underwater undulatory swimming kicks (UUS) and -1.32 strokes following the turn compared to those that worsened. Exploratorily, simple linear regression yielded body fat percentage as a significant predictor of improvement following FLYMA (b=0.03, p=0.015, r2=0.38). Greater physiological maturity and maximizing UUS following the turn may increase successful outcomes.Master of Art
CARE MANAGEMENT AND RACIAL DISPARITIES IN HYPERTENSIVE DISORDERS OF PREGNANCY AMONG MEDICAID ENROLLEES IN NORTH CAROLINA
Get PDFRacial and ethnic inequities in hypertensive disorders of pregnancy persist, with low-income people of color experiencing preeclampsia or eclampsia at 69.8 per 1000 deliveries, compared to 43.3 per 1000 for White birthing individuals. The reasons for these differences are unclear, but social factors tied to historical marginalization and unequal access to services may drive inequities. Since 2011, North Carolina's Medicaid has implemented a pregnancy medical home program, incentivizing providers to conduct standardized risk screenings for clinical and social needs and provide individualized care management. However, little focus has been placed on understanding the utilization and impact of care management services on hypertensive disorders of pregnancy among racial and ethnic groups. To address these gaps, I utilized a dataset spanning 2017-2019, which included North Carolina Medicaid claims, birth certificates, risk screening and care management data, and the US Census Bureau’s American Community Survey (2016-2020). I chose this period because the program began new risk scoring in 2017, and the pandemic disrupted in-person non-emergency services after February 2020. Also, North Carolina’s Medicaid program moved to managed care in 2021, preserving universal risk screening and care management but ending the data reporting system. In Aim 1, I examined the association between neighborhood deprivation, measured by the Index of Concentration of the Extremes, and the timeliness of patient entry into care, risk screening completion, and face-to-face care management among Medicaid enrollees with 1 preeclampsia risk factor using multivariate logistic regression models. In Aim 2, I examined the effect of 5 face-to-face care management visits on preeclampsia onset among Medicaid enrollees with 1 preeclampsia risk factor using a bivariate probit model. In Aim 3, I examined racial disparities in face-to-face care management services for those with 1 preeclampsia risk factor using generalized linear and probit models, implementing the Institute of Medicine (IOM) model for measuring disparities. This research demonstrates that neighborhood deprivation restricts access to prenatal services, informing targeted interventions. It highlights the effectiveness of care management services in reducing racial disparities in hypertensive disorders of pregnancy, illustrating how programs like pregnancy medical homes can integrate equity measures in evaluations to improve health outcomes for everyone.Doctor of Philosoph
Defining Novel Mechanisms of E3 Ligases in Response to DNA Replication Stress
Get PDFDNA replication stress is a major driver of genome instability. The DNA damage response (DDR) mechanisms that allow cells to tolerate DNA replication stress are critically important for genome stability and cell viability.In this thesis, I focus on identifying novel roles of E3 ubiquitin ligases in tolerating DNA replication stress. E3 ligases are known to play important roles in DDR signaling. However, as a class, E3 ligases are understudied proteins and there are many E3 ligases with unknown functions. Using an unbiased genetic screen, we identify a role for the RING finger E3 ligase RNF25 in promoting DNA replication stress tolerance. RNF25 protects reversed DNA replication forks from nucleolytic degradation. Mechanistically, RNF25 interacts with the replication fork protection factor REV7 and recruits REV7 to nascent DNA after replication stress. The role of RNF25 in protecting replication forks is fully separable from its canonical functions in ubiquitin conjugation. This work reveals the RNF25:REV7 signaling axis as an important protective mechanism in cells experiencing replication stress.DDR pathways are often pathologically reprogrammed in cancer cells. Cancer cells have high levels of intrinsic replication stress and depend on DDR mechanisms for survival. The genome maintenance mechanisms that neoplastic cells use to tolerate intrinsic replication stress can also confer resistance to therapy-induced DNA damage. Therefore, better understanding of DDR pathways will enable us to elucidate better strategies for sensitizing cancer cells to existing therapies. The cancer testis antigen MAGE-A4 promotes DNA replication stress tolerance through stabilization of its binding partner, the E3 ligase RAD18. MAGE-A4 is an attractive therapeutic target due to its cancer-specific expression. We show that MAGE-A4 is therapeutically tractable and describe first-in-class cyclic peptide inhibitors that disrupt its association with RAD18. Interestingly, we also identify RNF25 as a novel binding partner of MAGE-A4. The MAGE-A4:RNF25 interaction may eventually reveal additional ways to sensitize cancer cells to replication stress.Doctor of Philosoph
A harmonic non-parametric test of independence for high-dimensional data sets
Get PDFDetermining independence between continuous random vectors in high-dimensions (d >> 1)from noisy observations is the goal of a new framework for a class of non-parametric tests ofindependence. Recent work on non-parametric tests of independence use Hadamard matrices (Walshwavelets) to perform a frequency analysis in search of dependencies between random variables of anoisy sample. This dissertation expands on that work by presenting a framework for other typesof functions that can be used to extract dependence features within the noisy sample through aharmonic analysis. The variety of high-dimensional functions that this work explores utilise efficientalgorithms to implement them: hierarchical tree algorithms, fast wavelet transforms, randomisedalgorithms and other tools form scientific computing. Special attention is focused on wavelet theory,where Haar wavelets present an adaptive hierarchical approach to achieving dependence detectionin high-dimensions, and on k-nearest neighbour algorithms to build localised functions with specialstatistical properties.Doctor of Philosoph
Morphological Assessment of Mercenaria Shells from Warm and Cold Climates of the Plio-Pleistocene, US Mid-Atlantic Coastal Plain
Get PDFBivalve shell morphology offers insight into their evolutionary history and life habits, with morphological changes often linked to environmental changes. However, few morphometric studies address possible contributing factors to fossil shell shape variation within a single species. This study investigates whether bivalve shape changed based on seawater temperature and seasonality during the mid-Piacenzian Warm Interval compared to the early-Pleistocene. The hypothesis is that Mercenaria shells from the two time intervals display morphologic distinctions that are influenced by climatic or environmental factors related to temperature. Traditional morphometrics was used to measure and compare physical characters (n=11) in 57 Mercenaria shells from North Carolina and Virginia. Visual analyses did not appear to illustrate morphological differences, though Mahalanobis distance between centroids suggest significant differences among most populations. Despite significant results, further analyses and additional samples are required to support the hypothesis of morphological differences between shells from different climate intervals.Master of Scienc
SCALING SYMBOLIC EXECUTION FOR EFFICIENT SECURITY VERIFICATION OF HARDWARE
Get PDFEnsuring the correctness and security of hardware designs is critical in modern computing systems. Formal verification methods, particularly model checking, are widely used in industry to verify functional correctness. Recently, symbolic execution has emerged as a powerful alternative for hardware security verification. However, symbolic execution faces fundamental scalability challenges. Symbolic execution suffers from the path explosion problem, where the number of execution paths grows exponentially with the number of branch points executed. Furthermore, symbolic execution engines rely heavily on satisfiability modulo theories (SMT) solvers to compute satisfiability queries, which dominate the overall runtime and remain a bottleneck [1, 2]. This dissertation introduces three innovations to make symbolic execution practical for large-scale hardware verification. First, we present the Sylvia hardware symbolic execution engine and a new technique called piecewise composition, which leverages the modularity of hardware designs to reduce redundant path exploration. By decomposing execution into independent logical blocks and reconstructing full execution paths using SMT queries, piecewise composition significantly reduces the number of paths explored per clock cycle. Building from here, SylQ-SV extends our hardware-oriented symbolic execution engine by adding language support for SystemVerilog and SystemVerilog Assertions. Furthermore, we bring query caching to the hardware domain to reduce redundant solver invocations. Finally, we develop SEIF, a symbolic execution-based information flow analysis framework that integrates symbolic execution with a static signal connectivity graph, guiding path exploration toward security-critical flows. We evaluate our approaches on multiple open-source hardware designs, including system-on-chip (SoC) architectures and CPU cores, and contribute a new set of open-source hardware security verification benchmarks to the community. Our results demonstrate that piecewise composition reduces redundant work, query caching significantly improves execution time, and augmenting symbolic execution with static analysis unlocks information flow validation. By addressing the scalability challenges of symbolic execution, this work establishes new foundations for efficient, security-driven symbolic execution of hardware designs.Doctor of Philosoph
Balkan Berlin: Civic and Social Integration Across Three Generations of Western Balkan Migration to Germany
Get PDFAs Germany has become an increasingly diverse nation, large numbers of migrants from the Western Balkans have found a place in the tapestry of German society. This began in large numbers after the Second World War, with Yugoslav workers filling the demand for labor in West Germany and West Berlin as guest workers. Migration to Germany from the Western Balkans accelerated in the 1990s as refugees fled the Yugoslav Wars, and today many Western Balkan nations face a brain drain as they continue to struggle with economic hardship and strive to accede to the EU. Migrants across these three generations faced unique circumstances while integrating in Germany. This analysis defines integration in both civic and social terms, and analyzes the civic and social integration of each of these three generations. The results show higher levels of integration in both of these respects for the most recent cohort of migrants from the Western Balkans to Germany.Master of Art
Expanding the Utility of Configurationally Unstable Compounds: Generating Molecular Complexity Through Organocatalyzed, Stereoconvergent Processes
Get PDFThe increasing stereochemical complexity of molecules with pharmaceutical and agrochemical relevance has drawn interest in the development of methods to obtain such molecules in high stereochemical purity. Molecules bearing configurationally unstable functionality stand as an ongoing challenge in the field of synthetic organic chemistry. Stereocenters containing an acidic proton, such as protons proximal to β-dicarbonyls, α-halo esters and pyridylmethyl ketones and acetates, embody this challenge when exposed to basic conditions. We demonstrate productively leveraging kinetic control and crystallization thermodynamics to uncover stereoconvergent pathways towards stereochemically pure molecules. Disclosed are molecules decorated with up to five contiguous stereocenters, demonstrating these methods as powerful tools for stereochemical complexity generation. Detailed mechanistic investigations unveil the factors governing the stereoconvergence exhibited in the reported examples.Doctor of Philosoph
Vehicle-to-grid as a Demand Response Strategy for Carsharing in Thailand
Get PDFElectric vehicles (EVs) are quickly being adopted as a primary solution for decarbonization. Vehicle-to-grid (V2G) is an emerging technology that allows EVs to dispatch electricity back to the grid through bidirectional charging. Carsharing businesses could optimally benefit from a V2G program while contributing to grid stability and flexibility. This study examines an empirical dataset of 135 vehicles from Haupcar, a Thai carsharing company, to assess the impacts of V2G. Using a Monte Carlo Simulation, three charging schedules and tariff designs were simulated to estimate changes to grid load, the internal rate of return, and the potential avoided carbon dioxide emissions with program implementation. The findings reveal a break-even profitability point of 24 THB/kWh (~USD 0.70/kWh). However, with reduced charging infrastructure costs and preferential tariff policy, V2G can achieve profitability at a residential electricity tariff of 5-6 THB/kWh, improving the economic viability for carsharing companies.Master of Scienc
USING SATELLITE IMAGERY AND DEEP LEARNING TOOLS TO DETECT HOG OPERATION SPRAYFIELDS AND LAGOONS IN EASTERN, NORTH CAROLINA
Get PDFSarah Marlene Rachita: Using Satellite Imagery and Deep Learning Tools to Detect Hog Operation Features in Eastern, North Carolina(Under the direction of Courtney Woods)Concentrated animal feeding operations (CAFOs) for hogs and poultry have a long history in Eastern North Carolina, where spreading hog waste on fields raises environmental and health concerns. Advancing satellite imagery and machine learning makes identifying the true ground area covered by hazards impactful for agricultural practices and may drive environmental policy. This study uses ArcGIS Pro deep learning tools and object detection models to identify hog waste lagoons in Sampson County and delineate agricultural fields in Duplin and Sampson Counties. The lagoon model achieved a precision score of 0.8013, detecting lagoons with few false positives. The field model delineated 5,045 unique fields and overlapped with 499 known Duplin sprayfields, capturing 28.2%, suggesting poor training performance. While post-model cleaning is recommended, improving training data is the most effective strategy. Overall, the study shows that deep learning applied to satellite imagery offers promise for identifying environmental hazards and informing health policy.Master of Scienc