22689 research outputs found
Sort by
Health Service Utilization Practices of Black Immigrant Women with Cumulative Trauma Experiences
No full textBackground: Cumulative trauma experiences are associated with negative physical and mental health outcomes. Black immigrant women face pre- and post-migration experiences of trauma, which increases their risk for negative health outcomes. Health services promote disease prevention and treatment, yet factors that influence health service utilization practices of Black immigrant women with cumulative trauma experiences are rarely explored.
Objectives: To explore the relationship between cumulative trauma and health service utilization and examine the health service utilization experiences of Black immigrant women with cumulative trauma living in the US.
Methods: An explanatory sequential mixed methods design was used. The quantitative phase explored the association of sociodemographic factors and cumulative trauma experiences with health service utilization of 147 women. The qualitative phase explored the experiences, barriers, and facilitators of health service utilization through semi-structured individual interviews with 17 women who were purposively sampled from the quantitative study sample. The Adapted Behavioral Model of Health Services Use guided the study design and organization of results.
Results: Cumulative trauma had a positive relationship with health service utilization and discrimination strengthened the positive association between cumulative trauma and health service utilization. Qualitative data revealed that although many women noted changes to health following trauma experiences, high trauma levels did not equate to increased health service utilization. Barriers to health services included lack of insurance, high cost of health care services, and limited knowledge of the US health care system. Motivators included increased health needs, high prioritization of health, assistance from religious/community organizations, being insured, provider referrals, and satisfaction with US health care.
Conclusion: Despite the homogenous nature of the study sample, perceptions on the impact of cumulative trauma on health service usage varied. Health care providers must practice patient-centered care that prioritizes individuals’ health needs and future research must incorporate social networks and intersectionality into study design. To increase access to health services for Black immigrant women with cumulative trauma, local and national policies must provide financial support to cover health care fees and bolster programs that increase awareness of available health services, health insurance plans, and strategies to navigate the US health system
Supports and Barriers to Experiential Learning Participation in Arts Higher Education
Get PDFHigher education arts programs are implementing experiential learning (EL) programs to address skills gaps and better prepare students for careers in an evolving arts industry. However, there is a dearth of research relating to EL in the arts. As a starting point, this research explores the supports and barriers that contribute to student participation in extracurricular EL programs. This mixed methods study employed interviews and a focus group with nine students at a Mid-Atlantic performing arts conservatory, accompanied by a brief demographic survey. The findings revealed a complex web of supports and barriers impacting students’ awareness of, participation in, and applications to EL programs, including individual, social, programmatic, and systemic factors. The framework of this small-scale study may serve as a model and starting place for other institutions to understand their own unique audiences and settings moving forward.</p
Analysis of Air Pollution Data: Spatial Filtering of Low-Cost Sensors and Estimating Causal Effects of Sharp Interventions
No full textAir pollution is a major public health concern, and considerable research efforts have been made to measure air pollution concentrations and to use air quality data to answer public health questions. However, there are many unresolved problems in the air pollution space. This dissertation focuses on three problems. First, we present a method calibrate data from a low-cost air pollution sensor network. In recent years, low-cost sensors have been deployed in dense networks to measure spatially-resolved air quality, but these sensors are not accurate and must be calibrated before they can be used. We present a spatial method for calibrating data from low-cost sensors, using spatial correlations and a model for the low-cost measurements to make predictions. The methods are then applied to a low-cost network in Baltimore, Maryland. Second, we develop an approach to combine data from multiple low-cost networks. Many cities, including Baltimore, have multiple low-cost networks across the same area, where each network may have different biases. We extend the single network calibration to simultaneously calibrate multiple networks, and apply this method to Baltimore to provide better estimates of air quality than when only a single network is used. Third, we estimate the causal effects of interventions on air pollution. Policy interventions, such as the COVID-19 lockdowns in 2020, can impact air pollution concentrations, and quantifying the effect of an intervention on air quality is a difficult problem to solve. This work presents a method to estimate the causal effect of an intervention on air pollution, as well as a validation framework which can assess, both conceptually and empirically, whether any method is suitable for a particular application. These chapters together present a collection of methods for calibrating and analyzing air pollution data, along with applications of these methods to settings of interest
Turning CAR T therapy solid tumor infiltration into reality using synthetic velocity receptors
No full textChimeric antigen receptor (CAR) T cells express antigen-specific synthetic receptors, which upon binding to cancer cells, elicit T cell anti-tumor responses. CAR T cell therapy has enjoyed success in the clinic for hematological cancer indications, giving rise to decade-long remissions in some cases. However, CAR T therapy for patients with solid tumors has not seen similar success. Solid tumors constitute 90% of adult human cancers, representing an enormous unmet clinical need. Current approaches do not solve the central problem of limited ability of therapeutic cells to migrate through the stromal matrix. We discover that T cells at low and high density display low- and high-migration phenotypes, respectively. The highly migratory phenotype is mediated by a paracrine pathway from a group of self-produced cytokines that include IL5, TNFa, IFNg, and IL8. We exploit this finding to “lock-in” a highly migratory phenotype by developing and expressing receptors, which we call velocity receptors (VRs). VRs target these cytokines and signal through these cytokines’ cognate receptors to increase T cell motility and infiltrate lung, ovarian, and pancreatic tumors in large numbers and at doses for which control CAR T cells remain confined to the tumor periphery. In contrast to CAR therapy alone, VR-CAR T cells significantly attenuate tumor growth and extend overall survival. This work suggests that approaches to the design of immune cell receptors that focus on migration signaling will help current and future CAR cellular therapies to infiltrate deep into solid tumors
Statistical Inference on Time Series of Graphs
Get PDFGraphs usually arise in many fields of study, like neuroimaging or social network analysis, to model relationships between subjects. Recently, statistical inference on graphs has rapidly developed. Much work has focused on single graph inference, while recently
there has been increased interest in multiple graphs concerning modeling and performing statistical inference. Among recent developments in statistical inference on multiple networks, MASE is a statistically principled multiple random graph embedding method for networks with latent space structure, absent dynamics, or time dependency. We aim to continue this trend by developing methods to study the properties of time series of graphs. In this dissertation, I develop feature extraction and anomaly detection methods for data as time series of graphs.
In Chapter 1, we present a comprehensive existing literature review and briefly introduce the motivation and background of our methods.
Chapter 2 considers the graph signal processing problem of anomaly detection in a time series of graphs. We examine two related, complementary inference tasks: the detection of anomalous graphs within a time series and the detection of temporally anomalous vertices. We approach these tasks via the adaptation of statistically principled methods for joint graph inference, specifically multiple adjacency spectral embedding (MASE). We demonstrate that our approach is effective for our inference tasks. Moreover, we assess the performance of our method in terms of the underlying nature of detectable anomalies. We further provide the theoretical justification for our method and insight into its use. Applied to the Enron communication graph and a large-scale commercial search engine time series of graphs, our approaches demonstrate their applicability and identify the anomalous vertices beyond just large degree change.
In Chapter 3, we consider the problem of extracting features from passive, multi-channel electroencephalogram (EEG) devices for downstream inference tasks related to high-level mental states such as stress and cognitive load. Our proposed feature extraction method uses recently developed spectral-based multi-graph tools. It applies them to the time series of graphs implied by the statistical dependence structure (e.g., correlation) amongst the multiple sensors. We study the features in the context of two datasets, each consisting of at least 30 participants, and recorded using multi-channel EEG systems. We compare the classification performance of a classifier trained on the proposed features to a classifier trained on the traditional band power-based features in three settings. We find that the two feature sets offer complementary predictive information. We conclude by showing that the importance of particular channels and pairs of channels for classification is neuroscientifically valid when using the proposed features
Understanding the Impact of Partner-Perpetrated Reproductive Coercion on Postpartum Outcomes: Longitudinal Evidence from Ethiopia
No full textBackground: Reproductive coercion (RC) undermines autonomy and is linked to non-use of contraception and elevated risk of unintended pregnancy. Little is known about determinants of pre-pregnancy RC or its relationship with postpartum outcomes.
Methods: Using PMA-Ethiopia 2019-2021 prospective cohort data, which followed pregnant women through 12 months postpartum, we assessed the relationship between two summary RC variables and postpartum outcomes. Aim 1 (N=2,169) examined correlates of RC and the probability of experiencing RC within the year before pregnancy. Aim 2 and Aim 3 (N=1,935) employed semi-parametric time-to-event modeling to assess the relationship between RC and time to adoption of postpartum family planning (PPFP) and time to rapid repeat pregnancy (RRP) within the first year postpartum.
Results: An estimated 32.6% of women (18.2% less severe; 14.3% more severe) experience RC in the year before pregnancy. Residence in Amhara (aOR=1.82, 95% CI: 1.21-2.75) increased risk of any RC; having ≥4 children (aOR≥4 children=0.49, 95% CI: 0.34-0.72) or more than secondary education (aOR=0.39, 95% CI: 0.22-0.64) were protective. Residing in Afar (aRRR=0.35, 95% CI: 0.13-0.94) and having ≥4 children (aRRR≥4 children=0.42, 95% CI: 0.27-0.68) were protective against less severe RC. Living in Amhara (aRRR=2.40, 95% CI: 1.34-4.04), being in a cohabitating relationship (aRRR=2.04, 95% CI: 1.06-3.94), and having a previous relationship (aRRR=1.84, 95% CI: 1.22-2.78,) increased risk of more severe RC; having 2-3 children (aRRR2-3 children=0.54, 95% CI = 0.32-0.91) or more than secondary education (aRRR=0.16, 95% CI: 0.07-0.40) were protective. In Aim 2, relative to no RC, there was no difference in PPFP uptake among women who experienced less severe RC (aHR=0.90, 95% CI 0.62-1.07, p=0.24); more severe RC increased hazard of uptake (aHR=0.79, 95% CI: 0.62-0.99, p=0.05). In Aim 3, relative to no RC, there was no difference in the hazard of RRP among women who experienced less severe RC (aHR=1.44, 95% CI: 0.80-2.59; p=0.23); more severe RC increased the hazard of RRP (aHR=1.72: 95% CI 0.97-3.04; p=0.06).
Conclusions: Severe forms of RC may decrease uptake of PPFP and increase RRP. Addressing gender power imbalances and supporting woman-implemented safety strategies to maintain reproductive autonomy are important RC prevention and response strategies
MY ISLAND HOMES, AND OTHER STORIES
No full textMany of the articles and stories included in this collection revolve around two islands that are important to me: Long Island and Nantucket. Born and raised in Huntington, New York, I have spent most of my life on Long Island’s North Shore. In 2021, I moved to Nantucket—my grandma’s favorite place in the world. It now means a lot to me, too. Throughout much of this collection, I explore the nature of each place, as well as some of the environmental issues each island is facing
Engineering Flexible Optoelectronic Materials for Photovoltaics Applications
Get PDFSolar energy has captured the world’s attention in recent decades due to
its abundance, ubiquity, long-term sustainability and environmental friendliness. It is now regarded as a significant replacement or supplement for traditional energy sources, especially in countries and areas with high solar irradiance. Current solar energy harvesting technology, as well as related optoelectronic technologies, are poised to benefit from colloidal nanomaterials, since these materials possess unique characteristics including size-dependent optical property tunability and room temperature solution-phase processing
flexibility, making them ideal for low-cost thin film optoelectronic applications in novel scenarios, such as flexible displays, wearable devices, vehicular power, and building integrated photovoltaics. This thesis addresses new developments in the utilization of lead sulfide (PbS)-based colloidal quantum dots in the area of photovoltaics, using a combination of nanomaterials synthesis, optoelectronic modeling, experimental demonstration and advanced
characterization.
The first experimental part of this thesis focuses on the design and fabrication of diffuse light solar concentrators for enhanced solar illumination intensity on photovoltaic devices. The design based on total internal reflection accomplishes wide acceptance angle sunlight concentration without external tracking components. The experimental verification using transparent flexible
silicon polymer-based optics paves the way for the deployment of building
integrated photovoltaics and other off-grid energy applications.
The next experimental section discusses engineering the photonic band structure in strongly absorbing materials enabling spectral selectivity in potential multi-junction photovoltaic applications. The experimental demonstration using PbS colloidal quantum dots serves as a novel method for spectral tuning
in optoelectronic devices.
The final experimental section demonstrates the possibility of solar cell fabrication solely via solution-phase techniques. A semi-automated spray-casting system is built for demonstration of fully spray-cast colloidal quantum dot solar cells, including the electrode materials, expanding potential applications
of photovoltaics onto a larger variety of substrates and infrastructures.
The broader impacts of the work described in this thesis rely on the involvement of flexible optoelectronic materials in photovoltaics. This could
lead to renewable energy deployment for new applications, eventually benefiting the movement towards a world powered solely by clean and sustainable energy
Obtaining Energetics and Dynamics of Proteins Using Adaptive Steered Molecular Dynamics
No full textAdaptive steered molecular dynamics (ASMD) is an enhanced sampling method developed to address the convergence challenges encountered by steered molecular dynamics (SMD), with the usage of the latter often hampered by the calculation of average work through the non-equilibrium trajectories of events, e.g., protein unfolding, via Jarzynski’s Equality. ASMD, on the other hand, overcomes this issue by dividing the overall reaction coordinate into smaller stages, constraining the spread of work values with contractions performed at each stage. In this dissertation, several contraction criteria are compared and discussed with Ala10 as an example. Then, the most straightforward version of ASMD — Naıv̈e-ASMD — is applied to reveal additivity and other properties in the helix-coil transition among several alanine-rich polypeptides by the determination of energetic profiles and hydrogen bond changes. To adapt the ASMD approach to larger solvated systems, the telescoping box scheme is introduced to adjust the solvation box size to exactly immerse the partially unfolded proteins along the stretch progress. Lastly, we investigate the mutagenesis study on Enterocin 7B protein with several single-site Arg mutations through both experimental and computational approaches.
In summary, ASMD could serve to explore the energetics and dynamics of protein unfolding events efficiently via non-equilibrium trajectories. In practice, the integration of the telescoping box scheme developed here should limit the number of solvent molecules needed in such simulations
Exploring the relationship between statistical and physics-based free energies within the protein sequence-structure paradigm
Get PDFIn the field of protein stability prediction, computational resources remain a major
limitation. Modeling approaches using all-atom force fields provide a wealth of information
about the energetic determinants of sequence-structure compatibility, but require significant
amounts of processing power, memory, and data storage. Here we utilize the COREX algorithm
to develop a thermodynamic framework to evaluate the stability of a protein through the lens
of compatibility of any sequence for any fold. The heuristics in this approach are based on the
naturally occurring frequencies of different amino acids in defined thermodynamic
environments that encompass all possible energetic profiles. Our hypothesis is that these
frequencies are directly related to the energetic impact of a residue on the stability of the
protein, given the residue’s environment. If true, the frequency-based scores could provide
information that is comparable to validated physics-based approaches that directly evaluate the
energy of a mutation and therefore the effects on the compatibility between the sequence and
structure. To test this hypothesis, we compared the sequence compatibility scores obtained
from COREX analysis with the computed energy determined using the well-known physics-
based software embodied in the Rosetta platform. Importantly, we find a substantial correlation
between the energetic scores obtained from each approach, when applied to a large database
of proteins, ranging in size from 50 to 250 amino acids. To further investigate this correlation
and control for the well-known size dependence of protein stability, we investigated the ability
of each algorithm to identify which of two potential structures would be adopted by an
engineered set of high identity sequences of identical length. The demonstrated agreement
between COREX and Rosetta suggests that the thermodynamic heuristics provide a fast and
efficient means of evaluating the compatibility of a sequence for a particular fold while
maintaining important information often not captured in structure-based heuristic methods.
Because the COREX-based scores are derived from efficiently additive position specific
frequencies, they represent a 104
-fold decrease in computational resources necessary to
exhaustively evaluate the effects of sequence variation in a 200 amino acid protein. The
intentional combination of these methods provides a unique opportunity to investigate
previously elusive research questions with an accessible, computationally reasonable approach