36214 research outputs found
Sort by
Exploring Quantum Machine Learning-Enhanced Models for EEG Data Classification
No full textThesis (Master's)--University of Washington, 2025Electroencephalography (EEG) records brain activity linked to both executed and imagined movements, but separating true motor signals from background noise in high-dimensional EEG data remains a challenge. Reliable classifiers are therefore vital for accurately tracking patient progress over time. This work is part of a larger initiative, the Smart NeuroRehab Ecosystem, which has two primary goals: (1) to propose innovative physical-rehabilitation strategies for neurologic conditions such as stroke using emerging technologies that make therapy more accessible, and (2) to collect and analyze EEG data using machine learning (ML) models that classify movement-related brain signals. EEG data are complex and often difficult to interpret. In this research, we explore the use of quantum machine learning as an alternative approach for EEG signal classification. Compared to classical ML strategies, quantum methods may offer a fundamentally different way of representing and processing data, potentially improving classification performance or computational efficiency. We implement and analyze a ten-qubit Variational Quantum Classifier (VQC), and compare its performance to a tuned Random Forest baseline using EEG data from a publicly available 64-channel dataset. The task involves classifying each EEG time-window as either a movement or rest condition. Across 40 preliminary runs, the VQC achieves a macro-F1 score of approximately 0.75, accuracy of 0.76, and AUROC of 0.83, outperforming the Random Forest (macro-F1 ≈ 0.71, AUROC ≈ 0.79). In addition to higher macro-F1 and AUROC scores, the VQC also demonstrated significantly better precision and recall on the movement class, based on paired statistical tests. Most experiments were conducted on a quantum simulator, with a subset tested on a cloud-based quantum processor. These findings suggest that hybrid quantum-classical models can match or exceed the performance of tuned classical pipelines without increasing computational complexity. Within the scope of the Smart NeuroRehab project, this work demonstrates that quantum approaches may offer a practical path to continuous monitoring of EEG in clinical settings. Future improvements in quantum hardware may expand the range of practical applications in biomedical signal analysis
Chronic Pain and Social Pain: Loneliness, Healthcare Utilization, and Associated Characteristics in a Cross-Section of Patients Receiving Opioids for Chronic Pain
No full textThesis (Master's)--University of Washington, 2025Introduction: Loneliness is detrimental to several domains of health and quality of life, and is sometimes linked to higher healthcare utilization (HCU). Individuals prescribed opioids for chronic pain experience loneliness more frequently than the general population. Patients with chronic pain utilize more health services than the general population. Discovering characteristics associated with loneliness in this patient population could help prioritize the delivery of appropriate interventions and reduce unnecessary healthcare utilization.Methods: This is an analysis of cross-sectional survey data collected as part of the screening process for a clinical trial testing a social navigation intervention. The population was patients being prescribed long-term opioids for chronic pain. Characteristics associated with loneliness were identified via Ꭓ2 tests and logistic regression. Associations between loneliness and number of primary care visits, emergency room visits, and behavioral healthcare visits were measured using negative binomial regression. Regression models were adjusted for predictors of health service use selected from the Andersen Model of Health Service Utilization.
Results: The sample size was n=278. Characteristics associated with loneliness were social isolation( OR=1.33, 95%CI=1.18-1.51), pain severity, (OR=1.04, 95% CI=1.01-1.07), and living alone (OR=1.21, 95% CI=1.06-1.39). Being aged 65+ was associated with lower odds of loneliness (OR=0.768, 95% CI=0.650-0.908). Loneliness was not associated with increased primary care or ER visits. Loneliness appeared to have a moderate association with increased behavioral health visits (IRR=1.99, p=0.030), but this result was not significant under a Benjamini-Hochberg correction.
Conclusion: Given the prevalence of loneliness and chronic pain, there is a need for more research. Future researchers should consider revisiting this topic using longitudinal study designs with larger sample sizes
Impact of Cachexia on Resource Utilization and Costs in Patients with Pancreatic, Lung, and Colorectal Cancers
No full textThesis (Master's)--University of Washington, 2025Introduction: Cancer cachexia or cancer anorexia-cachexia syndrome (CACS) is a wasting disease characterized by inadequate food intake, loss of muscle mass, weight loss, inactivity, and changes in metabolism. Cachexia is estimated to affect 70-80% of pancreatic cancer patients and approximately 50% of lung and colorectal cancer patients. There is limited literature describing the impact of this condition on healthcare resource utilization and costs in the US.
Objective:
The primary objective of this study was to compare the healthcare resource utilization (HRU) and direct health plan costs of colorectal, lung, and pancreatic cancer patients diagnosed with cachexia to matched cancer patients without a cachexia diagnosis.
Methods:
We conducted a retrospective cohort study in commercial and Medicare Merative™ MarketScan® claims database. Patients with cachexia were identified following a colorectal, lung, or pancreatic primary cancer diagnosis between October 1st, 2016 – December 31st, 2022. Controls with no cachexia diagnosis during the study period were 2:1 matched with cachexia cases. Mean annual healthcare resource utilization and direct health plan costs were calculated using Kaplan-Meier Sample Average (KMSA) to account for patients censored before the end of the 1-year observation period. Significant differences between groups were determined by bootstrap 95% confidence intervals. A cox proportional hazard model was conducted in the unmatched cohort to determine relative hazard of cachexia diagnosis in different cancer types.
Results:
After matching the sample included 34,882 cancer patients, 11,880 in the cachexia cohort and 23,002 in the control cohort. Overall healthcare costs were significantly higher in colorectal, lung, and pancreatic cancer patients with cachexia (138,004 - 99,104, [95% CI: 101,224]). Average annual hospital admissions (cachexia: 0.98 admissions, [95% CI: 0.94 – 1.02]; control: 0.60, [0.58 – 0.62]) and annual emergency room visits, (cachexia: 2.47 ER visits [95% CI: 2.38 – 2.58]; control: 1.33 [95% CI: 1.29 – 1.36]) were significantly higher among cancer patients with cachexia compared to controls. Average annual outpatient encounters (cachexia: 61.5 encounters [95% CI: 60.7 – 62.4]; control: 44.6 [95% CI: 44.0 – 45.2]) and annual prescription fills (cachexia: 37.7 fills [95% CI: 37.1 – 38.4]; control: 30.4 [95% CI: 30.0 – 30.8]) were also significantly higher in cachexia cancer patients compared to controls. Among patients who developed cachexia, pancreatic cancer patients had the shortest median time to cachexia diagnosis of 1.2 months (IQR: 0.1 - 5.1) and had a 2.9-fold greater risk of being diagnosed with cachexia compared to those with colorectal cancer, which had the lowest hazard (HR: 2.90, [95% CI: 2.79 – 3.02, p<0.0001]).
Conclusion:
Colorectal, lung, and pancreatic cancer patients in the US, with a concurrent diagnosis of cachexia were found to have higher healthcare resource utilization and cost of care in the year following cachexia diagnosis
Repurposing City Streets: Assessing City-Scale Congestion and VMT Impacts with a System Dynamics Scenario Analysis
No full textThesis (Master's)--University of Washington, 2025Alternate street uses (ASUs) have been explored worldwide to capitalize upon public space to meet other societal needs. One widely adopted strategy catalyzed by the COVID-19 pandemic is the repurposing of roadway space from vehicular travel to other purposes, such as nonmotorized mobility, commercial use, or as social/recreational space. This thesis presents a scenario analysis of ASU policies using a system dynamics (SD) framework to investigate downstream and interacting effects of ASU policies and allocation in Seattle, WA and Phoenix, AZ. Model calibration was done with the Puget Sound Regional Council 2019 Household Travel Survey and Federal Highway Administration 2017 National Household Travel Survey datasets. It was found that increasing the existing amount of street space repurposed for commercial and social/recreational use led to decreased congestion in both Seattle and Phoenix models, while also increasing the non-motorized transportation (NMT) mode share. In higher-density cities with lower drive mode share like Seattle, increasing lane reallocation for uses that increase the capacity, safety, and reliability of NMT and transit modes can reduce both congestion and VMT. In lower-density cities with higher drive mode share like Phoenix, repurposing more street space for NMT would lower VMT; repurposing street space for transit use, however, yields little meaningful benefit. These findings support policies promoting widespread ASU placement and concerns about generating increased congestion by removing travel lanes are refuted. The thesis concludes that at least in higher-density cities, the reallocation of street space is not a zero-sum game, as those who still choose to drive experience reduced congestion, and the non-drivers receive safety and capacity enhancements. Furthermore, reduced total VMT supports broader traffic safety and emissions reduction goals
The Paradox of Resilience: Lower Mortality Among Unhoused Trauma Patients in the Pacific Northwest
No full textThesis (Master's)--University of Washington, 2025There is a gap in knowledge in the role chronic stress plays in response to subsequent acute stressors, with equipoise as to whether chronic stress may be protective. To study the possibility that individuals who experience chronic stress may be more resilient to an acute stressor, we aim to measure the time-to-event differences between housed and unhoused trauma patients after an acute injury, with reduced time to recovery as a marker of resilience. We conducted a retrospective, observational cohort study of trauma patients at a single center (N=7,666) separated by housing status from 2016-2019. Cox proportional hazards modeling, competing risk, and zero-inflated negative binomial regression was used to assess in-hospital mortality, differences in time to ICU discharge, and outcome-free survival at 60 days by housing status, when adjusting for trauma-related covariates. Unhoused patients had a 34.5% lower hazard of all-cause in-hospital mortality compared to housed (HR 0.66, 95% CI: 0.44-0.98, p-value=0.04). There was no difference in the time to ICU discharge, number of days liberated from ventilation, nor number of days spent out of the ICU. However, unhoused trauma patients had longer hospital admissions. Unhoused trauma patients had significantly lower all-cause in-hospital mortality compared to housed trauma patients, despite having arrived in more critical condition
Development, optimization, and validation of quantitative lateral flow assays for accessible, at-home biomarker monitoring
No full textThesis (Master's)--University of Washington, 2025Many chronic autoimmune disorders, like Rheumatoid arthritis and psoriasis, are characterized by painful joint inflammation and extreme fatigue. Patient outcomes are significantly improved by early treatment interventions, which can be enabled through information from continuous monitoring and early detection. The current standard for diagnosis and monitoring of these diseases is laboratory-based testing for the quantification of specific biomarkers. Frequent laboratory blood tests are not feasible for all patients or universally accessible, so a low-cost, point-of-care monitoring solution could improve the monitoring of autoimmune disorders and even provide quantitative metrics of treatment efficacy. The field of paper microfluidics has become a significant source of new tools that expand healthcare accessibility, particularly when designed to be robust in various environments. Innovations in this field have included the use of lateral flow assays and freeze-dried reagents to create portable and accurate testing devices. Here, we present the development of two quantitative sensing techniques for sandwich lateral flow immunoassays, which enable sensitivity to analytes present in two different concentration regimes (one being 1 million times more dilute than the other). The results indicate the utility and readiness of two immunoassays for translation into a paper-based, multiplexed, at-home biomarker monitoring platform
Office Space and Gentrification in King County: A Machine Learning Based Approach
No full textThesis (Master's)--University of Washington, 2025In recent years, King County has seen a surge in housing costs and unaffordability, leading to a housing crisis. Local newspapers and community groups have pointed to large companies and the highly paid employees they attract as responsible for unaffordability and gentrification in the area. Though intuitively this may appear to be the case, this thesis uses newly available Machine Learning (ML) technology to quantitatively investigate the importance of offices as they correlate with gentrification. Making use of data made available through the American Community Survey and King County GIS, patterns of gentrification examined from 2010 to 2019. Following this, demographic and aggregated office-related variables are established at the block group level, and new GPU-boosted methods of performing ML and SHAP analysis are used to investigate the level to which office-related variables, such as taxable land value, office age, etc., are correlated with a prediction of gentrification within a block group
Hidden Notes: The Impact and Legacy of Flutist, Educator, Administrator, and Advocate Dorothy Antoinette Handy
No full textThesis (D.M.A.)--University of Washington, 2025Dorothy Antoinette Handy was a Black female flutist, educator, and administrator. She taught at several Historically Black Colleges and Universities (HBCUs), she was one of the first Black musicians of the Richmond Symphony in Virginia, and she wrote books on Black musicians in both the classical and jazz fields. While serving as the Director of Music for the National Endowment for the Arts, Handy was responsible for having jazz nationally recognized and funded by that organization. Though she never considered herself a composer, she also published one known piece for flute,Hommage A Haute Savoie: Five Short Impressions for Solo Flute, which stands out as an example of music by Black female composers from the 20th century.Until now, scholarship has neglected Antoinette Handy’s accomplishments, but they deserve to be celebrated. Her story is all the more extraordinary in light of her experiences as a Black woman who grew up in the south during Jim Crow and segregation. Her story also opens onto the rich history of a Black family’s life in the south and her passion for accessibility in the arts. She did what many strive to do now: to have a rich and impactful career in the arts that offers variety and fulfilment. Antoinette Handy highlights the hitherto underrepresented role of Black women in classical music and arts administration prior to the 21st century.
This dissertation provides an overview of Handy’s life with the intention of bringing awareness to her work, so that her career can serve as an inspiration to new generations of artists. Handy’s story reveals a lifelong passion for accessibility and equity, which she channeled into a successful career in the face of myriad challenges. Throwing light on her life and career in this way also shows how Handy’s legacy remains important today
Multilingual Immigrant High School Seniors’ of Color Experiences in a Transition Math Course: Intersectional Discourses of Race, Language, and Mathematics
No full textThesis (Ph.D.)--University of Washington, 2025This qualitative case study investigated multilingual immigrant high school seniors’ of color experiences in a transition mathematics course, Pathways to Higher Education Mathematics . Transition mathematics courses are designed to prepare students, deemed not on track for graduation, for college-level mathematics. Furthermore, I employed Critical Race Theory, Intersectionality as a Social Theory and a Poststructural Theory of discourses to analyze the intersectional discourses expressed by students in relation to race, language and perceptions of mathematics capabilities with their schooling experiences. Multilingual immigrant students are often not adequately prepared to meet high school graduation requirements and thus are disproportionately placed in low-track or transition mathematics courses. There is also minimal research on transition courses or multilingual immigrant students’ experiences in these classes. Findings showed limited opportunities for mathematical reasoning within the course materials and structure. There were patterns of limited participation and collaboration along with high skipping rates. Students primarily saw the course purpose as a step toward graduating high school versus preparing them for higher education or a source of mathematical learning. Furthermore, students saw Pathways as a class for “dumb” people. I explored how students internalized deficit discourses of race, language and perceptions of a racial hierarchy in terms of mathematical abilities based on their experiences in Pathways. I also discuss how these discourses and experiences shaped their perceptions of their future mathematics trajectories. This research adds to the field of mathematics education by highlighting students’ experiences and perspectives in transition courses. It also contributes to the understanding of how students internalize racialized and intersectional discourses of the use of English in math classrooms and perceptions of race and mathematics intelligence
Entangled: Identity, Control, and Depictions of Hair in the Ancient Near East
No full textThesis (Ph.D.)--University of Washington, 2025This dissertation examines how the portrayal of hair in Mesopotamia, New Kingdom Egypt, and the Hebrew Bible expresses identity and enacts control. Hair offers a basis for comparison across these three different but interconnected cultures: it is uniquely symbolizable, being visible, malleable, and ubiquitous. This study draws on Bourdieu’s concept of habitus: the cultural, historical, and social norms that influence and produce individual practice, even without the individual’s conscious awareness. It also understands ancient Near Eastern texts and images as performative, that is, as active agents that shaped reality. This work uses a series of thematic case studies, examining depictions of the hair of foreigners, of kings and elites, and of liminal figures for each of the three cultures. It draws on both textual sources, e.g., New Kingdom love poems, and visual sources, e.g., the Neo-Assyrian palace relief sculptures. Each case study examines how these portrayals of hair, seemingly ornamental, are intertwined with and bring about three kinds of control. First, an exemplar may depict real-world control, as with a soldier grabbing an enemy by his beard. Second, these portrayals control their audience, molding their habitus and behavior into the desired form envisioned by the elitecreators of these works. Third, these depictions themselves enact cosmological control, shaping reality, whether by legitimating a king or repelling the chaos of the foreign. By examining hair across cultures, this dissertation reveals both shared strategies, such as the careful differentiation of foreign hair, and unique inflections, such as the uniformity of Neo-Assyrian hair across class and gender. Moreover, by attending to the details of how hair is depicted, this dissertation demonstrates that focus on a common feature can serve as a productive point of comparison across media, time, and cultures. While the scholarly conversations of Egyptology, Assyriology, and Biblical Studies are often distinct, this dissertation uses the motif of hair to illuminate the repetition of key themes across cultural boundaries, highlighting a shared worldview, shaped by elite values and concerns