22689 research outputs found
Sort by
LABYRINTH A COLLECTION OF SHORT STORIES AND PERSONAL NARRATIVE
Get PDFIn today’s cultural and media landscape, a palpable undercurrent of unease exists, a sense of impending doom that permeates our collective consciousness. Whether it’s the sudden onset of a global pandemic, the looming specter of environmental collapse, the fragility of political stability, or the existential threats posed by technological advancements, a pervasive sense of foreboding remains. Against this backdrop of uncertainty and apprehension, “Labyrinth” seeks to explore and interrogate the complexities of life, death, and purpose through the lens of three short stories
CHALLENGES AND OPPORTUNITIES OF CHILDREN AND ADOLESCENTS LIVING WITH HIV IN THE AGE OF UNIVERSAL ANTIRETROVIRAL THERAPY
No full textBackground: The introduction of Universal Test and Treat (UTT), initiating ART immediately after diagnosis regardless of CD4 count, has greatly improved the survival of children and adolescents with HIV globally. This dissertation sought to describe their experiences throughout the life course during this modern era of universal antiretroviral therapy availability.
Methods: In Aim 1, we described the HIV care continuum among children and adolescents with HIV in a region where two-thirds of them reside, Eastern and Southern Africa, using a systematic review and meta-analysis. In Aims 2 and 3, we used data from the North American Cohort Collaboration on Research and Design (NA-ACCORD) to describe the incidence of AIDS-defining conditions (ADC) after transition to adult HIV care (Aim 2), and the incidence of non-AIDS defining comorbidities (NADC) by age 30 among young adults with perinatally acquired HIV (PHIV).
Results: Among children and adolescents in Eastern and Southern Africa in the UTT era, only three-fourths were aware of their HIV status, two-thirds were on ART, three-fifths were retained in HIV care, half were adherent to ART, and half were virally suppressed. In North America, one in five people with PHIV had an ADC within three years after transitioning to adult HIV care, and burden was similar to non-PHIV; and by age 30, one-fifth of young adults with PHIV had diabetes mellitus Type 2 (T2DM), two-fifths had hypercholesterolemia, half had hypertriglyceridemia, one-fourth had hypertension, and one-fourth had chronic kidney disease (CKD).
Conclusions: Children and adolescents with HIV experienced old and new challenges in the UTT era. Children and adolescents with HIV in Eastern and Southern Africa continued to experience challenges across the HIV care continuum, although proportions of being on ART may have reached Joint United Nations Program on HIV/AIDS (UNAIDS) targets. Children with HIV in North America, upon transitioning to adult HIV care, continued to experience ADCs and may have had early onset NADCs relative to the general population. There are several opportunities to intervene, such as strengthening efforts in HIV care engagement and early screening for NADCs, and future research to better describe HIV-related mechanisms in NADC development
NOVEL BIOSENSOR ENABLES IDENTIFICATION OF POTENT CAMKII INHIBITORS
No full textCardiovascular diseases remain as leading killers worldwide. For the past two decades, Ca2+/Calmodulin-dependent protein kinase II (CaMKII) has been repeatedly shown to be a driver of cardiac injury. Our group and others have identified CaMKII hyperactivity as pathogenic in ischemia/reperfusion, myocardial infarction, heart failure, and arrhythmias. Thus, there is a pressing need for translational CaMKII-inhibiting therapies. Despite proven benefits of CaMKII inhibition in numerous preclinical models of heart disease, translation of CaMKII blockers into humans has been stymied by low potency, toxicity, and an enduring concern for adverse effects on cognition. Small molecule inhibitor discovery has been hampered by a lack of a CaMKII biosensor suitable for high throughput screening. Existing CaMKII have a myriad of limitations including low dynamic range and dependence on cellular machinery to function. Here, we developed a novel CaMKII activity biosensor that overcomes all of these obstacles and offers unprecedented sensitivity and specificity. We provide feasibility studies that demonstrate the utility of this tool in reporting the spatiotemporal dynamics of CaMKII in multiple tissues. Then, we identified CaMKII inhibitors among drugs that are safe for human use that are more potent and less toxic than current gold standard compounds. Lastly, we demonstrate that our top compound can inhibit CaMKII in cultured cardiac cells and in live mice, and is capable of ameliorating a CaMKII-dependent arrhythmia
San Mateo County Wastewater Treatment Facilities: An Assessment of Adaptation Readiness and Pathways to Resilience
Get PDFWastewater treatment and resource recovery facilities provide essential services for the protection of human health and the environment. Within San Mateo County, California, six low-lying wastewater treatment facilities face current and future flood vulnerabilities due to uncertified flood protection and the projected impacts of sea level rise-related hazards, including daily high tides, storm surge, groundwater rise, and coastal erosion. An adaptation readiness assessment was conducted through an integrated literature review, analysis of sea level rise-related hazard exposure, and review of vulnerability assessments, adaptation plans, and other local documents. Based on the review, adaptation projects were classified, and the adaptation approaches for each WWTF were characterized. Key findings reiterate the significant vulnerability of the County’s wastewater treatment facilities, regional reliance on levee and seawall approaches to adaptation and adaptive action gaps. A wastewater-specific adaptation planning toolbox is provided to support wastewater treatment facility managers and planners in fortifying vulnerability assessments and climate change adaptation plans
Three Essays on Energy Markets
Get PDFIn this dissertation, we focus on a traditional asset in the commodity and energy market, the crude oil refinery, and a relatively new asset, Distributed Energy Resources (DERs). In addition, electricity and renewable energy sources are studied. In Chapter 2, our objective is to find the optimal transitioning time for the crude oil refinery into a greener business - the renewable diesel facility. Under the current global environment policies, a significant amount of fossil fuel reserves like crude oil will remain underground to reduce pollution. As a
result, we believe crude oil refineries, which use crude oil as feedstock, will
be stranded or restructured in the future. We employ a two-factor stochastic differential equation system to model the commodity price (Schwartz and Smith (2000) and Mirantes, Población, and Serna (2012)) and a novel stochastic
model that incorporates the information of global temperature for carbon credits. Kalman filter is used to calibrate all model parameters. Furthermore, a new Least Squares Monte Carlo (LSMC) framework is designed specifically for our problem. Various regression techniques, such as OLS, supervised learning, and deep learning, are used to approximate the conditional expectations for the LSMC.
In Chapter 3, we continue the previous study by Geman and Ma (2023) on the problem of Distributed Energy Resources (DERs) and flexibility options, with a focus on heating, ventilation, and air conditioning (HVAC) systems in the state of Texas. We propose different types of options that can be used for various flexibility-acquiring purposes. We design time series models for energy consumption, temperature, and electricity spot price. Options are priced by the Monte Carlo method and are further validated by our evaluation metrics, such as consistency, sensitivity, and reliability. Lastly, we apply the clustering method with our model-based features to divide thousands of HVACs into different fleets or groups and potentially rank them based on
their overall performance.
In Chapter 4, we aim to analyze the renewable generation in ERCOT and
perform an empirical study of the influence of renewable energy sources (RES) on the electricity spot price. Our defined Renewable Penetration Index shows a significant increase in the use of wind and solar in Texas over the last four years. To evaluate the impact of RES on the electricity spot price, we apply different statistical methods to analyze their linear relationship, quantile relationship, and volatility regime-switching relationship. Our finding shows high renewable penetration might increase the volatility of electricity spot price and bring a statistically significant negative impact on the price
CHANGES IN STATE SPECIAL SUPPLEMENTAL NUTRITION PROGRAM FOR WOMEN, INFANTS AND CHILDREN PARTICIPATION AND STATE IMPLEMENTATION OF FEDERAL POLICY
Get PDFThe Special Supplemental Nutrition Program for Women, Infants and Children (WIC) is a federally funded program that provides healthy foods, nutrition education, and breastfeeding support to eligible low-income women and children in the United States. Over 50% of infants, 25% of children under 5 years old, and 25% of pregnant and postpartum women receive WIC benefits. Despite the WIC program’s reach, participation began to decline in 2009 for unknown reasons.
In this observational study, the objective was to examine if state implementation of three Federal policies, the 2009 WIC Food Package Changes, WIC Electronic Benefit Transfer (EBT), and REAL ID, were associated with changes in state WIC participation from 2005-2017. The study database was comprised of annual state-level measures of WIC participation in 50 states and Washington DC and covariates obtained from publicly available government sources. Two quasi-experimental statistical methods, interrupted time-series (ITS) and difference-in-differences (DID), were employed in analysis.
Compared to pre-policy WIC participation, ITS analysis revealed the immediate change in participation after the 2009 Food Package changes was not significant in most states. Two of the ten states whose models displayed negative level trends, or in other words the immediate change in the number of WIC participants post-policy was less than predicted by pre-policy trends. A positive level change was observed in the null ITS models of the remaining 41 states, but only five states displayed significant coefficients (p 0.05).
The average treatment effect (ATT) calculated in DID analysis indicated a possible positive but not significant effect of WIC EBT implementation on state WIC participation. The WIC EBT ATT was 8,447 participants (p=0.19) in the null model, and 6,973 participants (p=0.32) in the extended model. State REAL ID implementation exhibited a possible positive, but not significant, effect on state WIC participation. The REAL ID ATT was 8,020 participants (p=0.32) in the null model, and 924 participants (p=0.80) in the extended model.
Few researchers have examined if state implementation of Federal policy is associated with changes in state WIC participation. This novel study revealed the 2009 WIC Food Package changes were associated with long-term participation changes in the majority of states. State WIC EBT and REAL ID implementation appeared to have a positive impact on WIC participation, but study results were not significant
A MECHANISTIC ACCOUNT OF COGNITIVE FATIGUE AND ITS INFLUENCE ON MAJOR DEPRESSIVE DISORDER
No full textFatigue is one of the requisite symptoms for diagnosing major depressive disorder (MDD). Despite knowledge that fatigue alone can influence individuals' decisions regarding reward and effort expenditure, its specific role in the disrupted decision-making we see in MDD remains poorly understood. This thesis uses two studies, to begin to bridge the knowledge gap between how fatigue influences effort-based choice in healthy persons, and how fatigue symptoms may be contributing to further disruptions of effort-based decision-making in MDD.
In Aim 1, functional magnetic resonance imaging (fMRI) is employed to investigate the neurobiological mechanisms underlying how cognitive fatigue impacts decision-making. Participants undergo a cognitively fatiguing forced-choice paradigm, revealing an increase in activity in the right anterior insula (aIns) during fatigue. This heightened activity correlates with task-based brain activity, indicating that task-specific fatigue is encoded in the aIns and influences the valuation of effort-based choices.
In Aim 2, a comparable forced-choice paradigm is used alongside scales assessing depression symptoms and fatigue levels. Using factor analysis to distinguish fatigue from depression, the study finds that changes in effort-based decisions in MDD are primarily driven by fatigue-induced increases in effort cost.
These results, in combination with prior evidence of disrupted neural function in MDD of in the same brain areas brain responsible for integrating fatigue into motivation centers of the brain, suggest that the fatigue circuit may be dysfunctional in MDD. We also provide evidence that fatigue should not be ignored as a standalone factor in how those with MDD make decisions, and that further investigations of effort-based decision-making in MDD should include measures of fatigue to allow for more accurate and precise conclusions that contribute the development of a biomechanism for MDD and other psychiatric disorders
Brittle Failure of Cellular Metamaterials
No full textExploiting advances in additive manufacturing, architected materials employ the programmed distribution of matter as a means to attain superior properties compared to those of traditional monolithic solids. While there has been a large volume of studies focusing on the nonlinear mechanics associated with the elastic instabilities and plastic localization of lightweight lattices and foams, their fracture properties and failure mechanics remain relatively unexplored. Cellular metamaterials made from a brittle solid (e.g. ceramic, carbon) are used extensively in thermal protection systems, filtering, and novel energy storage devices. In each application mechanical integrity is crucial to maintain functionality and performance. This thesis introduces a comprehensive framework that integrates additive manufacturing, micro-computed tomography, experiments at different length scales, and computational modeling, to elucidate and predict the failure mechanics and associated fracture properties of a general class of ordered and disordered metamaterials. We show how the complex interplay of loading conditions, manufacturing-induced imperfections, topological characteristics, and the fracture properties of the parent solid, leads to a given macroscopic strength. High-fidelity numerical models, equipped with appropriate failure criteria at the strut level, are shown to accurately capture the compressive response of different 3D topologies including random foams and periodic lattices. We further examine which morphological characteristics govern tension-compression asymmetry in the fracture strength of brittle lattices and discuss a novel experimental protocol to measure their toughness
PRENATAL EXPOSURE TO A METAL MIXTURE AND CHILDHOOD AUTISTIC TRAITS, INCORPORATING GENETIC SUSCEPTIBILITIES
No full textBackground: Heavy metals and trace elements may play important roles in shaping
child neurodevelopmental outcomes. Moreover, the relationship between these
elements and child health outcomes may be influenced by inter-individual differences in
genetic susceptibilities. However, previous epidemiologic studies have mainly employed
a case-control design, examined individual metals, and have not considered exposure
in the context of genetic susceptibilities.
Objectives: We aim to 1) examine the joint and independent associations between
metal concentrations during pregnancy with developmental traits related to autism in
early childhood, 2) assess evidence that specific single nucleotide polymorphism (SNP)
and genes are associated with concentrations of lead (Pb) in human biological matrices,
and 3) investigate the associations between urinary Pb concentrations during
pregnancy and genetic susceptibility to Pb.
Methods: In Chapter 2, we utilized data from two prospective pregnancy and birth
cohorts that enrolled newly pregnant women who had already given birth to a child with
autism spectrum disorder (ASD). We analyzed spot urine samples collected from these
women one to three times during pregnancy and measured five metals (Pb, mercury
[Hg], manganese [Mn], selenium [Se], and cadmium [Cd]). The Social Responsiveness
Scale (SRS) was administered for each child when they reached 3 years of age. Using
a Bayesian Kernel Machine Regression (BKMR) model, we estimated the effect of
prenatal metals mixture as well as individual mixture component levels on prospective
child SRS scores. In Chapter 3, we performed a systematic review and meta-analysis of
the existing literature on genetic variant associations with metals concentrations. Finally,in Chapter 4, we developed a Pb polygenic score (Pb-PGS) and used linear regression
models to examine the relationship between urine Pb concentrations during pregnancy
and maternal Pb-PGS, reflecting aggregate genetic variation contributions to lead
concentrations.
Results: In Chapter 2, we show that no clear monotonic associations were observed
between the prenatal exposure to a metal mixture and children's SRS scores, collected
at around 36 months. As reported in Chapter 3, we observed some indication of positive
associations between the vitamin D receptor (VDR) BsmI genetic polymorphism
(rs1544410) and elevated levels of Pb in human biological matrix, specifically bone. No
significant statistical associations were discovered between a Pb polygenic score and
Pb concentrations in the urine of our study participants (Chapter 4).
Conclusions: Our results suggest increased prenatal exposure to a mixture of metals
does not influence children's SRS scores at approximately 36 months. Additionally,
despite evidence in Chapter 3 supporting genetic contributions to blood and urine
metals concentrations, at individual loci, an aggregate score of multiple loci does not
associate with metals levels in pregnant people
Listening to Multi-talker Conversations: Modular and End-to-end Perspectives
Get PDFSince the first speech recognition systems were built more than 30 years ago, improvement in voice technology has enabled applications such as smart assistants and automated customer support. Conversational intelligence of the future is expected to move beyond single-user applications of voice technologies to actively participate in human conversations, including in scenarios such as note-taking, fact-checking, or collaborative learning in peer groups. For such systems, recognizing free-flowing multi-party conversations is a crucial and challenging component that still remains unsolved. In this dissertation, we focus on this problem of speaker-attributed multi-talker speech recognition for the meeting transcription task, and propose two perspectives which result from its probabilistic formulation.
In the modular perspective, speaker-attributed transcription is performed through a pipeline of sub-tasks involving speaker diarization, target speaker extraction, and speech recognition. Our first contribution is a novel method to perform overlap-aware speaker diarization by reformulating spectral clustering as a constrained optimization problem. We also describe an algorithm to ensemble diarization outputs, and show that it can be used to either combine several overlap-aware systems, or to perform multi-channel diarization by late fusion. Once speaker segments are identified, we robustly extract single-speaker utterances from the mixture using a GPU-accelerated implementation of guided source separation. This eventually allows us to use an off-the-shelf ASR system to obtain speaker-attributed transcripts.
Since the modular approach suffers from error propagation, we propose an alternate “end-to-end” perspective on the problem. For this, we describe the Streaming Unmixing and Recognition Transducer (SURT) which extends neural transducers for multi-talker ASR by incorporating an unmixing component. We show how to train SURT models efficiently by carefully designing the network architecture, objective functions, and mixture simulation techniques. Finally, we add an auxiliary speaker branch to enable joint prediction of speaker labels synchronized with the speech tokens, and propose a novel speaker prefixing approach for ensuring label consistency through the recording. We demonstrate that training on synthetic mixtures and adapting with real data helps these models transfer well for streaming transcription of real meeting sessions