Universities at Shady Grove

Digital Repository at the University of Maryland
Not a member yet
    33760 research outputs found

    A Multi-Method Study Investigating the Biopsychosocial Mechanisms Linking Racial Discrimination to Type 2 Diabetes Risk and Self-Management Behaviors Among Black Adults

    No full text
    INTRODUCTION: In the United States (U.S.), Black adults have a higher incidence of type 2 diabetes (T2D) than White adults and are more likely to die from T2D. Some of the risk factors for T2D include chronic stress; being physically inactive, obese or overweight; living with prediabetes; being older (over 45); and having a family history of the disease. Black Americans have higher levels of exposure to psychological stressors that are rooted in racial discrimination, as exemplified at the height of the COVID-19 pandemic with the murder of George Floyd, and are consequently at an increased risk of developing stress-related illnesses like T2D. Racial discrimination has been historically neglected in the assessment of stress, and the role it plays in the onset and self-management of T2D remains understudied. Additionally, the mechanisms that link racial discrimination and T2D are not clear. METHODS: In this study, the Healthy Aging in Neighborhoods of Diversity across the Life Span (HANDLS) data set was used (Waves 1, 3, and 4) to examine and analyze the relationship between racial discrimination-related stress and T2D risk (measured as having a T2D diagnosis). The HANDLS data set was also used to evaluate the mediating effect of diet quality and moderating effect of adaptive coping on the relationship between racial discrimination-related stress and T2D risk. HANDLS is a longitudinal study that consists of a fixed cohort of 3,720 socioeconomically diverse African American and White adults between 30-64 years of age who reside in the Baltimore City area. Research questions were centered on understanding the natural history of disease in the sample, capturing early biomarkers that influence health disparities in prevention and disease mitigation efforts. This dissertation study aimed to: 1) examine and analyze the effect of racial discrimination-related stress on diet quality and T2D risk among a cohort of Black adults 30-64 years of age, and analyze the effect of coping as a buffering effect, while also examining the influence of nativity status (U.S. born vs. foreign-born) on T2D risk; and 2) explore the role of stress, which may stem from experiences of racial discrimination, in the self-management of T2D among Black adults 18 years of age or older living in Prince George’s County, Maryland and the Metro Detroit, Michigan area. Structural equation modeling (SEM) was performed to assess the causal pathways linking racial discrimination-related stress to T2D risk, while semi-structured in-depth interviews (N=17) shed light on the barriers to and facilitators of participant engagement in behavioral activities (e.g., physical activity, dietary patterns, and medication taking). The interviews explored participants’ access to nutritious and culturally appropriate food options in their communities and the methods used to cope with chronic stressors, such as racial discrimination. The interviews, which focused on T2D self-management, directly explored stress stemming from racial discrimination as well as other identified stressors and were conducted to understand the gravity of experiences of racial discrimination. In-depth interviews were conducted until thematic saturation was reached. Those with a clinical diagnosis of T2D were recruited to participate in the interviews via both The Health Advocates In-Reach and Research (HAIR) Program at the Maryland Center for Health Equity and the Healthier Black Elders Center (HBEC) Participant Resource Pool (PRP). RESULTS: Using SEM with survival analysis to assess T2D risk, the direct effect of racial discrimination on T2D risk was positive (coeff: 0.674, p<0.001), meaning increased experiences of racial discrimination were linked to a later T2D diagnosis. Adaptive coping moderated the relationship between racial discrimination-related stress and T2D risk (p<0.001), while diet quality did not mediate the relationship (p = 0.972). As a secondary approach using generalized SEM with a logistic regression framework, more reported experiences of racial discrimination were directly associated with a decreased odds of not developing T2D during the study period (coeff: -1.319; aOR: 0.27; p<0.001; Cohen’s d = -0.73), meaning that increased experiences of racial discrimination were linked to a higher likelihood of T2D development. The direction of the association was expected. Using latent thematic analysis to explore T2D self-management behaviors, six themes were identified: (1) A change in family dynamics influences diet; (2) Standing firm and unshaken; (3) With age comes increased knowledge; (4) Change comes from within; (5) Partnership with healthcare provider; (6) Relying on God and community. In managing T2D, spousal support and familial loss were key drivers of food choices for male participants. Male participants tended to engage in avoidance coping, while female participants endorsed the superwoman schema and the added stress it brings, all of which contributed to changes in dietary habits for some. CONCLUSIONS: Our study demonstrates that increased use of adaptive coping techniques blunts the response to racial discrimination-related stress among those with fewer reported experiences of racial discrimination. Although different measures of racial discrimination have been developed, validated, and widely used, future research should support improved methods of capturing experiences of racial discrimination, as experiences are often suppressed. In this study, the unique combination of semi-structured in-depth interviews and quantitative analyses using secondary data, allowed for a deeper understanding of the mechanisms through which racial discrimination affects diabetes development and diabetes-related complications, helping to close the gap in health disparities for Black Americans. Findings may contribute to the development of future discrimination measurement tools for Black adults. Findings may also be used to encourage knowledge sharing to reinforce advocacy and further support the allocation of resources (e.g., diabetes self-management education and support programs and mobile clinics) to inner city communities

    Scaling Parallel Full-graph GNN Training to Thousands of GPUs

    No full text
    Graph neural networks have emerged as a potent class of neural networks capableof leveraging the connectivity and structure of real-world graphs to learn intricate properties and relationships between nodes. Many real-world graphs exceed the memory capacity of a GPU due to their sheer size, and using GNNs on them requires techniques such as mini-batch sampling to scale. However, this can lead to reduced accuracy in some cases, and sampling and data transfer from the CPU to the GPU can also slow down training. On the other hand, distributed full-graph training suffers from high communication overhead and load imbalance due to the irregular structure of graphs. In this thesis, we propose Plexus, a three-dimensional (3D)parallel approach for full-graph training that tackles these issues and scales to billion-edge graphs. Additionally, we introduce performance optimizations such as a permutation scheme for load balancing, and a performance model to predict the optimal 3D configuration. Plexus is evaluated on several graph datasets and scaling results are shown for up to 2048 A100 GPUs on Perlmutter, which is 33% of the supercomputer, and 1024 MI250X GPUs on Frontier. Plexus achieves unprecedented speedups of 2.3x-12.5x over existing methods and a reduction in the time to solution by 5.2-8.7x on Perlmutter and 7-54.2x on Frontier

    EXPLORING FLUXONIUM-BASED QUANTUM COMPUTING

    No full text
    Fluxonium qubit is a promising elementary building block for quantum information processing due to its long coherence time combined with a strong anharmonicity. In this thesis, we first introduce a novel fluxonium qubit operating at zero magnetic field with high coherence. We implement and characterize single-qubit gates with an average gate fidelity of 99.93%, extracted from randomized benchmarking. This qubit serves as a ready-to-use superconducting qubit that operates in the frequency range of conventional transmons and exhibits stronger anharmonicity. Next, we implement a 60 ns direct CNOT gate on two inductively coupled fluxoniums, which behave almost exactly like a pair of transversely coupled spin-1/2 systems. Notably, the typically undesirable static ZZ term, arising from non-computational transitions, is nearly absent even in the presence of strong qubit-qubit hybridization. The CNOT gate fidelity, estimated via randomized benchmarking, reaches 99.94%. Furthermore, this fidelity remains above 99.9% over a span of 24 days without any recalibration between measurements. Compared with the 99.96% fidelity of a 60 ns identity gate, our results constrain non-decoherence-related errors during logical operations to as low as 2 × 10^−4. This work adds a simple and robust two-qubit gate to the still relatively small family of “beyond three nines” gates on superconducting qubits

    3D-PRINTED MICROFLUIDICS FOR HYDRODYNAMIC PARTICLE ARRAYING

    No full text
    Cell trapping devices have become integral platforms for performing a wide variety of biological and medical studies, and have contributed to significant advances in many fields, including cellular development, drug testing, and cancer research. By enabling the isolation and immobilization of individual cells, constantly monitored and deliberately controlled experiments can be performed on dozens or hundreds of cells simultaneously, increasing the capabilities and scale of cellular research. One critical feature of many biomedical studies is the microinjection of foreign material into cells to observe their responses and changes in development. However, the performance of microinjections can be a very difficult and lengthy process due to the mobility of the cells and the size of the microneedles and can lead to inconsistent results with low success rates. This thesis introduces the next step in broadening the scope of array trapping devices. It combines the efficiency and ease of modern commercially available 3D printers with the established principles of hydrodynamic trapping to create a straightforward device for passively arraying zebrafish embryos. This device also includes a novel sliding lid to provide access to the embryos for microinjection. Through careful design and extensive testing, this trapping array successfully immobilized embryo-sized beads with 100% efficiency. After loading the beads, the top was easily removed without disturbing the beads, which remained primed and in a conducive environment for microinjection. This trapping array will allow the microinjection of zebrafish embryos through a more straightforward process, at a faster rate, and with increased uniformity. By integrating this trapping device into biomedical studies, experiments can be performed with higher success rates, leading to advances in many fields, including biological development, drug discovery, and disease modeling

    The gift of presence: exploring the relationship between mindfulness and positive affect

    No full text
    Despite the ever-increasing ease of communication, feelings of isolation and loneliness have been on the rise, especially after the pandemic (Murthy, 2023). Given this information, the need to increase positive affect, or the extent to which an individual subjectively experiences positive moods such as joy, interest, and alertness, is of increasing relevance (Miller, 2011). There is much research on the general benefits of mindfulness, the non judgemental and non elaborative awareness of one’s immediate experience (i.e., present thoughts, feelings, and sensations) with an orientation of curiosity and acceptance (Hayes & Feldman, 2004). Most research around mindfulness is focused on one’s well being particularly in regard to meditative practices. However, there is less research discussing the pathways mindfulness operates through to promote positive affect. Self esteem, or one’s overall sense of being a worthy and valuable person, may be one of these pathways (Rosenburg, 1965). The current study explores the relationship between mindfulness and positive affect with a simple mediation analysis in which self esteem is utilized as a potential mediator. The measures were assessed with a Qualtrics survey in which 95 anonymous participants engaged in a reflective or analytical activity for three minutes. The current study explores whether self esteem mediates the relationship between mindfulness and positive affect. Implications for psychology researchers and clinicians are discussed

    DYNAMIC INTERACTIONS BETWEEN SORPTION AND BIODEGRADATION OF HALOGENATED COMPOUNDS IN CONTAMINATED GROUNDWATER

    No full text
    Groundwater is one of the most important freshwater sources on Earth. Toxic dissolved chemicals from anthropogenic activities have been seeping into the groundwater, posing a risk for drinking and irrigation water resources. The contaminants form a plume that travels underground spreading those chemicals to a wider area. Permeable Reactive Barriers (PRBs) can be implemented underground to intercept the plume via sorption and biodegrade the contaminants. Traditionally, Granular or Powdered Activated carbon (GAC or PAC) were used as a fill material. Within the past decade, Colloidal Activated Carbon (CAC) has been produced, reducing the cost of PRBs installations. Despite years of research on sorption and biodegradation, some aspects of the dynamic interaction between adsorption, desorption and biodegradation have not been fully understood. Therefore, in this study, the kinetics of biodegradation in presence and absence of sorptive material, the extent of biomediated desorption, and bioregeneration efficiency were examined. Extent of biomediated desorption reached 93.5 ±4.4% on CAC materials, showing high bioavailability of sorbed TCE. This means that initially sorbed contaminants are able to desorb and biodegrade. Longevity of PRBs can be extended with the carbon acting as a repository. Moreover, biofilm morphology in presence of this novel material was characterized. It was shown that small CAC particles are taken up and embedded in biofilm. Different CAC products were compared to a reference GAC to examine the effect of particle size on kinetics, biomediated desorption extent and bioregeneration efficiency. The results showed a slower biodegradation kinetics on GAC compared to CAC. All CAC samples had a significantly (p < 0.05) higher extent of biomediated desorption than GAC samples. Additionally, bioregeneration efficiency of GAC decreased significantly post biodegradation, with GAC showing 41.5 ±3.5%, while CCAC samples exhibited an 88.4 ±11.8% efficiency.It is commonly reported that biodegradation is the rate-limiting factor for the coupled sorption and biodegradation. Therefore, an equilibrium model was compared to a dynamic kinetics model, to evaluate this assertion. The results showed that biodegradation and sorption have an interplay, and biodegradation alone is not the rate-limiting factor for the overall process. This will allow practitioners and engineers to have a more realistic prediction of the PRBs remediation efficacy. Lastly, methanogenic activity was investigated to assess the Greenhouse Gas (GHG) emissions from bioremediation of chlorinated solvents. Methane is produced, with no difference between GAC and CAC. Future studies should assess these findings in a column or field applications

    STALLED CONVERGENCE: RACE, GENDER, AND SECTOR IN AMERICAN MANAGERIAL ACCESS AND EARNINGS, 1960-2021

    No full text
    This study contributes to sociological research on inequality in managerial employment by analyzing patterns of access and compensation across race, gender, and sector using harmonized IPUMS Census and American Community Survey microdata from 1960 to 2021. The analyses model relative odds of attaining managerial positions and decompose wage disparities among managers by decade using consistent occupational definitions, examining disparities for Black women, Black men, and White women relative to White men, with separate analyses by public and private sector.Three patterns emerge. First, wage convergence stalled or reversed after 2000 for African Americans, while White women continued progress. Black women's wages stagnated at 0.48 log points below White men (62 cents per dollar) in private sector from 2000-2021, while their public sector gap remained flat at 0.29-0.31 log points. Black men experienced wage regression: private sector gaps increased from 0.33 to 0.44 log points, and public sector gaps rose from 0.19 to 0.22 log points. White women continued modest gains, with gaps narrowing from 0.36 to 0.26 log points (private) and 0.26 to 0.21 log points (public). Second, substantial sectoral differences persisted. In 2021, Black women's access gap was 1.5 percentage points in public versus 6.8 points in private; Black men's was 2.1 versus 8.5 points; and White women's was 0.9 versus 1.1 points. Wage gaps were consistently smaller in public sector: Black women earned 0.29 log points less than White men versus 0.48 in private, Black men's gaps were 0.22 versus 0.44 log points, and White women's were 0.21 versus 0.26 log points. Third, gap composition shifted fundamentally. Educational differences between groups explained access gaps from 63% to 24% for Black men and 45% to 15% for Black women in private sector. Industry segregation and marital status differences emerged as primary explanatory factors, together explaining 40% of Black women's access gap by 2021. Coefficients effects increased substantially, reaching 94% for White women, 63% for Black women, and 55% for Black men in private management

    I AM STANDING IN A ROOM I MADE

    No full text
    I am standing in a room I made. This room holds all of you. Within this space is a portrait of each one of you. I cannot disconnect you from the places you have been–we have been. In an effort to preserve the memory of you I have set out to create a space where we can be together

    Odorant responses in freely moving C. elegans: Insights into learning and the role of small RNA regulation

    No full text
    The ability to sense and respond to dynamic environmental stimuli is a fundamental property of all living organisms. The nematode Caenorhabditis elegans, with its well-defined neural circuitry and genetic tractability, serves as an ideal model for studying the molecular and cellular mechanisms underlying sensory responses and behavioral plasticity. Here, we explore the behavioral responses of C. elegans to two attractive (butanone and benzaldehyde) and one aversive (nonanone) volatile odorant using a refined assay design that minimizes confounding variables by avoiding chemical or physical immobilization and incorporates bidirectional rectangular arenas to control for external gradients. Further, we introduce an information-theory-centric measure of dispersal that is applicable broadly to other model systems. Together with a measure for the responding population, this addresses potential locomotion defects and the presence of uncharacterized gradients within the experimental arena. Using this rigorous methodological and analytical framework, we investigated the role of small RNA regulators in chemotaxis to odorants and starvation-associated memory of pre-exposure to odorants. Defective chemotaxis and learning were observed in worms lacking some of the components of the RNA silencing machinery involved in double stranded RNA (dsRNA) import, primary dsRNA processing, secondary amplification and nuclear silencing, implicating small RNA-mediated regulatory mechanisms in odorant sensation and experience-dependent behavioral plasticity. Additionally, we reveal randomness in starvation-associated memory of butanone pre-exposure even under well-controlled laboratory conditions and speculate that learning is an infrequent occurrence in the dynamic wild. Finally, we highlight the limitations of transgenic models in behavioral assays, demonstrating that NeuroPAL worms, commonly used for neuronal identity mapping, exhibit altered baseline chemotaxis responses. These findings underscore the necessity of careful experimental design and interpretation when utilizing genetically modified strains for behavioral analyses. Together, our results refine behavioral methodologies for C. elegans chemotaxis assays and provide novel insights into small RNA-mediated regulation of sensory behavior. Future research leveraging automated behavioral tracking, whole-brain functional imaging, and transgenerational analyses will further elucidate the mechanistic interplay between small RNA pathways and neural function in behavioral plasticity

    Quantum Codes From Symmetry

    No full text
    The Eastin-Knill theorem shows that the transversal gates of a quantum code, whichare naturally fault-tolerant, form a finite group G. We show that G is an invariant of equivalent quantum codes and thus can be considered as a well defined symmetry. This thesis studies how the symmetry G dictates the existence and parameters of quantum codes using representation theory. We focus on qubit quantum codes that have symmetry coming from finite subgroups ofSU(2). We examine two different methods of deriving quantum codes from these symmetries. The first method is concrete but not very general, it only applies when G is a binary dihedral subgroup BDa of SU(2). The second method is abstract but more general. Not only does it apply to all subgroups of SU(2), but it highlights the role that symmetry plays in logical errors and unveils a hidden time-reversal symmetry. From each of these methods we produce many examples of novel qubit code families

    21,810

    full texts

    33,760

    metadata records
    Updated in last 30 days.
    Digital Repository at the University of Maryland
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇