36214 research outputs found
Sort by
Cycloparaphenylenes and Related Contorted Aromatics as Building Blocks for Multifunctional Nanomaterials
No full textThesis (Ph.D.)--University of Washington, 2025Polycyclic aromatic hydrocarbon (PAH) ligands containing catechols (1,2-benzenediols), and their derivatives such as tetraoxolenes (1,2,4,5-benzenetetraols), have been intensely studied within coordination chemistry due to their redox non-innocence and capacity for conjugation with metal d orbitals. As such, materials containing catechol-functionalized ligands display a host of unique electronic and magnetic behavior. For example, tris(catechol)-substituted nanographenes (e.g., 2,3,6,7,10,11-hexahydroxytriphenylene) self-assemble with metal ions into extended sheets that can mimic the hexagonal structure, extended conjugation, and even metallic bandgap of graphene. However, small changes to ligand design, metal source, solvent, and other synthetic parameters can dramatically modulate the architecture of resulting materials and their properties—often in unpredictable ways. In this dissertation, I describe the use of catechol-bearing PAH ligands to build multifunctional materials across dimensionalities, and explore their emergent structure-property relationships, with a particular emphasis on [n]cycloparaphenylenes ([n]CPPs or ‘carbon nanohoops’). In doing so, I establish new chemistries that could facilitate the formation of crystalline, extended, nanotubular metal–organic, covalent–organic, and supramolecular materials
Healthcare Resource Utilization and Costs in Patients with AML Treated with Post-Transplant Maintenance Therapy
No full textThesis (Master's)--University of Washington, 2025AbstractHealthcare Resource Utilization and Costs in Patients with AML Treated with Post-Transplant Maintenance Therapy
Rachel Kneitel
Chair of the Supervisory Committee:
Noemi Kreif
Pharmacy
Background: Allogeneic hematopoietic cell transplantation (allo-HCT) has improved survival in patients with acute myeloid leukemia (AML); however, post-transplant relapse remains the most common cause of treatment failure and death. There is limited data regarding the economic and clinical burden associated with the progression from diagnosis to post-transplant maintenance therapy, which is intended to maintain remission in patients.
Objective: To describe healthcare resource utilization (HRU) and quantify costs among commercially insured, Medicaid, and Medicare beneficiaries who have and have not received maintenance therapy after allo-HCT.
Methods: We conducted a retrospective cohort study on commercially insured, Medicare, and Medicaid beneficiaries with AML who were and were not prescribed maintenance therapy following allo-HCT between October 1st, 2015 and March 31st, 2024 using claims from the Merative MarketScan® database. The final analytic cohort included 373 patients, of whom 43 were prescribed maintenance therapy following allo-HCT. To account for observed differences between those who received maintenance therapy and those who only received allo-HCT, we employed inverse probability weighting (IPTW) using propensity scores estimated from baseline characteristics. We assessed differences in all-cause monthly HRU by summarizing the number of emergency department (ED), inpatient (IP), outpatient (OP) visits, and hospital length of stay throughout the 12 month follow up period. We also evaluated monthly supportive therapy utilization patterns, including therapies that stimulate neutrophil production to reduce infection risk and promote red blood cell production to manage anemia, to assess treatment-related care requirements. Poisson regression models were used to estimate monthly event rates for each HRU category. Total healthcare costs were reported as the sum of ER, IP, OP, and outpatient pharmacy costs to provide insights into the costs associated with maintenance therapy. Transfusion burden was evaluated by counting the number of transfusions for each cohort.
Results: The maintenance therapy group demonstrated significantly higher HRU across multiple service types. Office visits were more than doubled (IRR = 2.11, p < 0.001), with significant increases in IP visit rates (IRR = 1.39, p = 0.015), hospitalization rates (IRR = 1.37, p = 0.024), and overall OP utilization (IRR = 1.79, p < 0.0001). Specialist clinic and ED visit rates were higher but not statistically significant (IRR = 1.5, p = 0.151; IRR = 1.75, p = 0.309), respectively. Supportive therapy utilization peaked within the first three months post-transplant in the maintenance therapy group before converging with controls by month eight. Healthcare costs were driven primarily by IP expenses, with the maintenance therapy group incurring substantially higher costs four to ten months after allo-HCT. Pharmacy costs were expectedly higher in the maintenance therapy group throughout follow-up. Blood transfusion requirements were minimal in both groups, with 93% of maintenance therapy patients and 97.6% of allo-HCT only patients requiring no transfusions during the 12-month follow-up period. The mean monthly length of stay was significantly longer in the allo-HCT only group compared to the maintenance therapy group (33.77 ± 31.49 days vs. 20.97 ± 15.36 days), with a mean difference of 12.72 days (95% CI: 11.14,14.31; p = 0.001).
Discussion: Our findings show that those who received maintenance therapy following allo-HCT were associated with having significantly greater healthcare resource utilization and costs, as shown by higher rates of office visits and hospitalizations. IP visits accounted for the largest share of monthly healthcare expenditures, with the maintenance group incurring notably higher IP costs from months four to ten after allo-HCT. Pharmacy costs were the second largest component and were consistently higher in the maintenance group. Supportive therapy use was substantially higher in the maintenance group during the early post-transplant period, peaking within the initial three months. These findings highlight the need to balance the clinical benefits of maintenance therapy with its increased demands on healthcare resources
Symbolic Reasoning as a Library
No full textThesis (Ph.D.)--University of Washington, 2025The increasing complexity of software systemsdemands robust methods for ensuring correctness and performance.
As systems scale, traditional approaches are insufficient for uncovering subtle bugs or optimization opportunities,
and automated reasoning has emerged as a critical but challenging solution.
%Building effective automated reasoning tools is challenging.
However, advances in constraint solvers have shifted the engineering burden to
symbolic compilation systems that translate program semantics into efficiently solvable constraints. This dissertation introduces \grisette,a symbolic compilation framework
designed as a statically-typed, purely functional, monadic Haskell library
for building domain-specific symbolic compilers.
Its Ordered Guards (ORG) representation
enables all-path symbolic evaluation while merging symbolic values into a compact normal form.
This approach significantly reduces
symbolic evaluation time ( speedup),
constraint size ( reduction),
and solving time ( speedup) compared to traditional representations.
Beyond performance, \grisette avoids complications of lifting host languages into the symbolic domain.
Its functional design enables memoization,
while its monadic interface handles computational effects.
We evaluate \grisette on diverse benchmarks,
including \textsc{Rosette} applications.
\grisette's monadic design further enables symbolic reasoning for benchmarks
beyond existing tools,
like those with continuations and coroutines.
It also serves as the core engine for \tensorright,
an automated verification system for tensor graph rewrites. To demonstrate \grisette's versatility,this dissertation also presents \hierasynth,
a parallel framework for super-optimization built upon \grisette.
Super-optimizers synthesize high-performance code
but face a trade-off between program length () and instruction set size ().
\hierasynth{} introduces a decomposition strategy that adaptively, hierarchically partitions
the search space along dimension rather than .
It employs component-based synthesis with embedded instruction choices,
translated into SMT constraints via \grisette,
using parallel divide-and-conquer with efficient unrealizability pruning.
A RISC-V Vector super-optimizer built with \hierasynth
synthesizes programs of greater length ()
for larger instruction sets () than previously feasible,
discovering programs that are both provably optimal under a cost model
and empirically superior to human-designed code,
while achieving substantial synthesis scalability
Modular Design and Synthetic Tuning of Ternary Molecular Nanoclusters
No full textThesis (Ph.D.)--University of Washington, 2025The understanding of the nature of the metal/support interaction in catalytic systems is central todesigning better and more sustainable catalysts. The characterization of heterogeneous interfaces
most relevant to catalysis is challenging and relies on techniques that provide limited atom-level
information on processes at the surface. To better understand this complex chemistry, atomically
precise nanoclusters can be used as a molecular analogue to study such surface chemistry with
high precision. Here the chemistry of Co6Se8L6 (L = Ph2PR, Ph = phenyl) is studied as a means of
reaching this goal by facilitating the formation of ternary nanoclusters with metal edge sites which
sit upon the cluster core and are akin to single-atom catalysts on a surface. In the first chapter, the
synthesis of a series of Cu/Co/Se molecular clusters is described, and their properties probed
extensively through a bevy of techniques. Four highly isostructural clusters varying in overall
charge from –3 to 0 are synthesized and found to exhibit robust charge separation between the
copper edge sites and Co/Se core. The electronic interactions and redistribution of charge between
the copper(I) edge sites and the electronically flexible core glean insights into the nature of the
dopant/support interaction.
In Chapter 2, synthetic explorations are undertaken to expand the scope of the supporting
phosphine ligands on the cluster, elucidating the approaches and challenges associated with
developing new ligand platforms of this type. By varying the identity of the R-group on the
phosphine Ph2PR used in cluster syntheses, a library of homoleptic clusters Co6Se8L6 (L = Ph2PR)
with various heteroatom functionalities is created. It is found that many functional groups will
tolerate the conditions of cluster formation, allowing for the facile synthesis of several clusters
with contrasting functionalities. The utility of these clusters is limited by their solubility profiles
and challenges with metalation, but they serve as a promising example of synthetic versatility of
Co6Se8 nanoclusters and hybrid phosphine ligands while providing important lessons for the future
design of ternary nanocluster systems
ICU Delirium Prevention for Patients Who Speak a Language Other Than English
No full textThesis (Ph.D.)--University of Washington, 2025Background: Delirium acquired in the intensive care unit, ICU delirium, is a common neurologic sequela of critical illness, affecting up to 70% of patients. The incidence and duration of ICU delirium is associated with increased mortality, increased hospital length of stay, and increased healthcare costs. After hospital discharge, having experienced ICU delirium is independently associated with long-term cognitive impairment and post-traumatic stress disorder. ICU delirium prevention depends on effective communication to complete complex interventions. Due to a language barrier adding to communication challenges, critically ill patients who speak a language other than English (LOE) may be at higher risk for ICU delirium. Gaps in our knowledge perpetuate inequitable access to evidence-based interventions and disparities in outcomes for LOE patients. Purpose: This dissertation aims to examine how language barriers experienced by LOE critically ill patients influence adherence to ICU delirium prevention interventions. Methods: The first study (Chapter 2) describes and compares ABCDEF bundle fidelity for critically ill LOE patients to English-speaking patients by performing secondary data analysis of the Society of Critical Care Medicine (SCCM) ICU Liberation Dataset of ABCDEF bundle completion. The second study (Chapter 3) describes facilitators and barriers to implementing delirium interventions among critically ill LOE patients using qualitative thematic analysis. The third study (Chapter 4) assesses the acceptability, feasibility, and appropriateness of the VidaTalk™ application as an implementation strategy to increase the reach of the ABCDEF bundle to LOE ICU patients using convergent a mixed-methods study design. Results: In Chapter 2, We found that the odds of adherence to the A, E, and F elements were statistically significantly higher for LOE patients compared to English speaking patients, while the odds of adherence to the C and D elements were significantly lower for LOE patients compared to English-speaking patients. Additionally, LOE patients had a longer length of stay, longer length of mechanical ventilation, and higher odds of mortality compared to English-speaking patients. These findings indicate a disparity in the ABCDEF bundle those results in worse healthcare outcomes for LOE patients. In Chapter 3, The analysis revealed barriers on the institutional, unit, and individual levels. The institutional level barriers included limited access to interpreter devices, technological issues with the interpreter devices, and communication lags during interpretation. Unit level barriers included time constraints due to staffing and unit culture. Individual barriers included biases reducing use of interpreters and taking shortcuts, healthcare worker inexperience, lack of knowledge regarding the ABCDEF bundle and cultural humility. The facilitators on the institutional level included the resources and training provided to improve cultural humility and interpretation. The unit level facilitators include adequate staffing, easy access to interpreters, and high expectations of adherence to ABCDEF bundle elements. Individual facilitators were experienced health care workers (HCWs) who had knowledge of the benefits of the ABCDEF bundle elements and therefore prioritized them for patients regardless of language. In this chapter, we highlight major challenges to communication with LOE patients experience by HCWs and LOE ICU survivors alike. These challenges contribute to reduce quality and adherence to the ABCDEF bundle for LOE patients. In Chapter 4, the VidaTalkTM device was found to be acceptable, feasible, and appropriate for use with critically ill LOE patients. Healthcare workers liked the device because it was easy to use, enabled fast basic communication, empowered patients to express their needs, and had many preset phrases that were targeted toward the ABCDEF bundle elements. However, HCWs were concerned that patients who were already delirious would not be able to use the device. They also wanted more pre-set caregiver responses, categorized by bundle element, that would enable them to easily speak back to the patient. Conclusion: LOE patients experience health disparities in care and outcomes in the intensive care unit. The barriers prevent HCWs from being able to implement ABCDEF bundle measures with fidelity. The facilitators address the barriers, but do not overcome them entirely. Future research on generating interventions to enhance effective communication at the beside between LOE patients and English-speaking HCWs is necessary
Task-based neural correlates of self-focused attention before and after cognitive behavioral therapy
No full textThesis (Master's)--University of Washington, 2025Self-focused attention (SFA), a form of self-referential processing, is maladaptive in various psychiatric disorders and may be associated with poor treatment response. This study examined SFA in individuals with social anxiety disorder (SAD) and body dysmorphic disorder (BDD), testing the hypothesis that SFA is associated with hyperactivity within default network (DN) regions and with treatment response during cognitive-behavioral therapy. Participants included 30 patients with primary SAD or BDD and 28 healthy controls, who displayed above average and below average scores (respectively) on the Public Self-Consciousness Scale, which measured trait SFA. SFA was also measured by a self-referential encoding task, which yielded both behavioral reaction time measures and task-related fMRI measures of SFA. Results indicated significantly longer reaction times at pre-treatment for self vs. other trials in patients compared to controls, with patients showing notable improvement post-treatment. Neuroimaging revealed greater activation in DN regions, including the medial prefrontal cortex, during self vs. other trials in all participants; however, there were no significant group differences at pre- or post-treatment, nor in the changes from pre- to post-treatment. Neural measures of SFA were significantly associated with treatment response, whereas behavioral measures were not. These findings suggest that activity in DN regions may serve as a transdiagnostic biomarker of maladaptive SFA that is associated with treatment response
Code in Context: Keeping Developer Context and Interfaces on Evolving Software
No full textThesis (Ph.D.)--University of Washington, 2025Despite progress in programming tools and environments, developers still struggle to maintain mental models of the systems they build and the context surrounding their code. This dissertation advances a framework for enhancing program understanding and manipulation for all agents working with a codebase—including developers and automated agents such as large language models (LLMs)—through notes on shared context and high-level representations that are semantically attached and synchronized, i.e., external notes and UIs that stay linked to sections of a program where they are meaningful, with their content updated to remain relevant. I first address the specific case of floating-point error analysis with Odyssey. Odyssey is a workbench for floating-point analysis that transforms an existing low-level, black-box expression improvement tool into a high-level, scaffolded analysis process where users can analyze, generate, and iterate on automated suggestions using their own expertise. Next, considering how contextual information from an Odyssey analysis session might be attached to actual floating-point programs, I explore maintaining semantic connections between document sections and metadata using Magic Markup. This system uses an LLM to "magically" maintain the positions of external notes on an evolving document without write access through semantic anchoring rather than brittle syntactic approaches. Finally, I formalize and implement this approach in Codetations, a VSCode extension that helps developers contextualize documents with rich, interactive notes and tools. Codetations builds on the Magic Markup method to keep notes outside the document while integrating edit-tracking and exposing a rich API for annotations to respond to buffer changes and leverage editor features. In a qualitative evaluation, developers recognize this method as enabling more extensive and useful forms of documentation, and worked examples show the additional context improving code repair performance in LLMs. Together, these systems indicate a practical path by which existing programming systems can begin to offer powerful, developer-customized tools that meet developers' needs for contextual information and high-level analysis
Graph Genera and Minors
Full text linkThis thesis studies embedding graphs on surfaces, with an emphasis on orientable surfaces, rotation systems, and forbidden minors. We begin by developing the historical background, from the earliest occurrences of the crossing number of a graph, to Heawood's map-coloring conjecture and the eventual invention of rotation systems as a way of encoding cellular embeddings of graphs on surfaces. We then describe and analyze \textsc{PAGE}, a rotation system based genus algorithm developed jointly with Alexander Metzger. For an arbitrary graph with vertices and edges, \textsc{PAGE} determines the orientable genus of in steps where is the \textit{girth} of . We illustrate \textsc{PAGE} with examples where it was used to determine the genus of graphs whose genus was previously unknown, most notably the cage which has genus 17. Additionally, we discuss ways in which \textsc{PAGE} can be improved and other strategies that could be used to determine the genus of graphs such as the cage which are so far out of reach for algorithmic approaches. We also give an overview of the problem of determining the full set of forbidden toroidal minors, summarize known results, and discuss possible directions for further progress on completing the set
The Environmental Impact on Facility-Treated Pediatric Asthma Exacerbation: A Secondary Study
Full text linkThesis (Ph.D.)--University of Washington, 2025Pediatric asthma is a leading chronic disease in America; however, it is best thought of as a syndrome of airway symptoms with various etiological origins. As a result, there are numerous combinations of "triggers" that can precipitate an exacerbation in individuals who have asthma. Categorically, these triggers can range from psychosocial, such as stress and anxiety, to physical, such as pollen and pet dander. Family geographic relocations are events that can expose a child with asthma to many of these triggers simultaneously. Ecological transition is a concept within Bronfenbrenner's ecological system theory of human development, where individuals and their environments interact to produce change. The ecological transition concept accounts for aspects of the individual and the broadening layers of the environmental interaction (microsystem, mesosystem, exosystem, macrosystem, and chronosystem) in a holistic approach that allows for seemingly indirect factors to be incorporated in the development process. Through the lens of ecological transition upon the foundation of ecological systems theory, two studies were devised to examine different environmental effects on facility-treated pediatric asthma exacerbations. First, a secondary cross-sectional multilevel binomial regression (n=10,401) suggested there was no significant correlation between children of veteran parents and the likelihood of facility-treated asthma exacerbation (OR 0.96, p= .452). Second, a secondary longitudinal multilevel binomial regression (n=1,055,742) identified a significant correlation between the month of relocation and the first several months following a geographic relocation and the increased likelihood of facility-treated asthma exacerbation. Month of relocation (OR 2.10, p <.001), which is consistent with predictions from ecological transition and ecological systems theory
Analysis of Test Mixture and Water Contaminants Using Advanced Chromatographic Techniques
Full text linkThesis (Master's)--University of Washington, 2025This thesis work highlights Solid Phase Extraction (SPE) coupled with Intuvo 9000 Gas Chromatography (GC), comprehensive two-dimensional (2D) gas chromatography-time-of-flight mass spectrometry (GC×GC-TOFMS), and one-dimensional gas chromatography-mass spectrometry (GC-qMS) to study the chemical components of a Test Mixture and contaminated by a diesel water sample. Diesel compounds were found during the first step of the analysis on one- dimensional GC-qMS. However, due to significant peak overlap, the compound resolution and identification were limited. GC×GC-TOFMS was used to analyze samples and its SPE extracts to overcome those limitations. During the analysis of the Test Mixture, a consistent temperature program and other conditions were adjusted on GC-qMS developed method. Nonadecane peak intensities were used as an example compound to find the limit of detection (LOD), which produces detection ability at low concentrations, and gave an instrumental practical value of 4 ppm.
To analyze changes and determine which compounds were most reduced or eluted during the extraction, chemometric techniques such as Principal Analysis (PCA) and tile-based F statistic ratio (F-ratio) analysis were used. A significant change in analytes was shown by the separation of the extracts and original neat sample by PCA. To reduce noise and correct retention time shifts, the chromatograms with more than 2500 differential areas were reduced to around 1216 significant features using F-ratio analysis.
A detailed analysis of a typical peak at mass-to-charge ratio 120 discovered that the Extracts had a weaker signal intensity than the Original Neat Diesel Sample, which shows partial compound loss after extraction. When comparing the real laboratory data with the NIST library mass spectrum, it has strong match across significant fragment ions, confirming the identification of 1-ethyl-2methylbenzene