31825 research outputs found
Sort by
Racial literacy development and expressions in a middle school Black Student Union
No full textSoslau, ElizabethThere is a critical need in the United States for spaces that support students with marginalized racial identities. Racial affinity groups in K-12 schools can provide safe spaces for students of color to discuss identity-related experiences, develop healthy racial identities, explore issues related to racism, and act to effect change. At the same time, the current sociopolitical climate, in which discussions about race and racism are being erased from schools and classrooms, has highlighted the need for students to be equipped with the knowledge and skills needed to become citizens who can engage in questioning and disrupting racist systems. ☐ While research in the field of education has explored both racial literacy development and racial affinity groups, there is a lack of scholarship that centers youth voices in this work. Guided by a racial literacy framework that is further informed by research on racial affinity groups, this qualitative case study explores the ways in which middle school students develop and express racial literacy in the context of a Black Student Union. Over the course of a full academic semester, data were collected through observations of Black Student Union meetings and events, individual interviews with student members and the advisor, a focus group, and document collection. ☐ Findings from this work provide insights into the functions of affinity groups for K-12 students. The affordances of school-based racial affinity groups are similar to those of affinity groups for adults, but K-12 students may have different priorities and goals for participation. Additionally, there are clear overlaps between the affordances of racial affinity groups and students’ racial literacy. Students develop and express different elements of racial literacy to varying extents in the context of affinity groups, with development and expressions of healthy racial identities as a key outcome. Findings from this study demonstrate that school-based racial affinity groups can fill a critical need for spaces that support marginalized students and provide opportunities for racial literacy development that are increasingly absent from schools and classrooms.University of Delaware, School of EducationPh.D
Differences in Executive Functioning Performance and Cortical Activation Between Autistic and Non-Autistic Youth During an fNIRS Flanker Task: A Pilot Study
No full textThis article was originally published in Brain Sciences. The version of record is available at: https://doi.org/10.3390/brainsci16010065
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license. https://creativecommons.org/licenses/by/4.0/Background/Objectives: Autism spectrum disorder is associated with executive functioning (EF) challenges, yet the neural correlates of EF challenges in autistic youth remain unclear. This study aimed to examine EF performance and cortical activation in autistic versus non-autistic youth, using functional near-infrared spectroscopy (fNIRS) during a modified Flanker task. Methods: Thirty age-matched (11.6 ± 0.8 years) autistic (N = 15) and non-autistic youth (N = 15) completed congruent and incongruent conditions of a modified Flanker task while cortical activation in prefrontal, parietal, and temporal regions was recorded using fNIRS. The Behavior Rating Inventory of Executive Function (BRIEF) was used to assess general EF impairments. Behavioral data (i.e., Flanker task mean reaction time/accuracy, and reaction time variability) and cortical activation were analyzed using ANCOVAs. Pearson correlations were used to determine the relationship between cortical activation, EF performance, and clinical measures. The significance level was set at p 1.3; e.g., Global Executive Composite Score for autistic youth = 71.3 ± 3.7; controls = 47.8 ± 2.4), indicative of delayed EF development. During the incongruent condition, compared to non-autistic controls, autistic youth showed lower left inferior parietal lobe (IPL) activation (Mean HbO2 in autistic youth = −0.02 ± 0.006 mmol.mm; controls = 0.01 ± 0.006 mmol.mm, ps < 0.001, Hedges’ g = 0.5) and a lack of left-lateralized activation (e.g., left vs. right STS activation, p < 0.001, Hedges’ g = 0.41 in the non-autistic youth). In the ASD group, lower activation in the left STS was associated with lower EF performance (r = −0.28, p = 0.007), whereas greater activation in various right-hemispheric ROIs was associated with better EF performance (r = −0.31 to −0.35, ps < 0.005), suggesting potential compensatory activation. Conclusions: The findings revealed ASD-specific differences in the neural correlates of EF performance and possible alternative compensatory activation patterns. These potential neural correlates of EF performance highlight the utility of fNIRS-based neural measures to better understand the neural bases of EF differences in autism. Study Registration: This study was approved by the Institutional Review Board (IRB) at the University of Delaware (Protocol #: 1947455) on 4 October 2022.The last author, A.B., thanks the Delaware Biotechnology Institute and Delaware’s Economic Development Office for the Applied Research Competition Grant (PI: Bhat) and the Maggie Newmann Health Sciences Fund (PI: Bhat). Co-author, J.C., thanks the National Institutes for Health (NIH) for T32 doctoral funding (Grant #: T32HD007490, Predoctoral Training in Physical Therapy and Rehabilitation Research, Recipient: Corey; PI: Reisman) and the Foundation for Physical Therapy Research through the Promotion of Doctoral Studies (PODSs) I Scholarship for supporting J.C.’s efforts on this project. J.-M.T. would like to thank the Taiwan Ministry of Education for the International Doctoral Scholarship for supporting her doctoral studies. Last but not least, this work was supported by the National Institutes of Health through a shared instrumentation grant awarded to the University of Delaware (Grant #: 1S10OD021534-01, PI: Bhat)
Setting the record straight: digital collage as a practice of Black memory
No full textCallier, Durell M.Hicks, Cheryl D.This thesis paper serves as the supportive document in a thesis project that addresses the role of police violence in the persistence of anti-Black violence alongside the historic persistence and contemporary relevance of Black resistance to anti-Black violence. The Water Remembers: Correcting the Archive Through Black Joy, Memory, and Resistance is a digital exhibition that examines the police response to the Chicago Race Riot of 1919 as a microcosm of police behavior nationally. ☐ As a supportive paper Setting the Record Straight argues that art can fill gaps left in the archive by creating a more complete picture of both police behavior and Black Resistance. Viewers of The Water Remembers are welcomed to explore a digital exhibition that consists of 6 collages that address police violence, emboldened vigilantes, and Black resistance. Each issue has two corresponding photographs that help the viewer to connect the police violence of the Chicago Race Riot of 1919 to the anti-Black violence that remains prevalent today. ☐ Critical fabulation is used as a methodology to make connections between what we can easily discern from the historical photographic archives and what scholars have uncovered overtime. Police refusal to protect Black communities and their willingness to engage in violence emboldens white supremacy and anti-Black violence; yet Black Americans have always fought back against it. ☐ The Water Remembers and Setting the Record Straight offer a bold and intentional declaration that police violence is at least partly to blame for anti-Black violence.University of Delaware, Department of Africana StudiesM.A
LIRTS Viewer: A Web-Based Resource to View the Transcriptional Response of Lens Epithelial Cells to Injury
No full textThis article was originally published in Investigative Ophthalmology & Visual Science (IOVS). The version of record is available at: https://doi.org/10.1167/iovs.66.9.53.
Copyright 2025 The Authors
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.Purpose: Residual lens epithelial cells (LECs) respond to injury after cataract surgery, leading to posterior capsular opacification (PCO). Transcriptomic profiling of lens capsule–associated cells (CACs) post-cataract surgery (PCS) revealed that LECs quickly alter their transcriptome, producing numerous pro-inflammatory cytokines within a few hours PCS. In contrast, the significant activation of TGFβ signaling and fibrotic extracellular matrix deposition related to PCO only begins 1 to 3 days later. However, the global changes in gene expression in CACs, following the establishment of robust TGFβ signaling, remain unknown.
Methods: Lens fiber cells were removed from wild-type mice, and CACs were isolated at 0, 72, or 120 hours PCS to perform bulk RNA sequencing (RNA-seq) to obtain estimates of RNA abundance. These data were combined with existing RNA-seq datasets to create a web-based visualization resource to explore the expression dynamics of most protein coding genes in CACs.
Results: At 72 hours PCS, CACs differentially express genes consistent with a surge in proliferation and changes in actin filament organization while also robustly expressing fibrotic marker genes by 120 hours PCS. We developed a data visualization resource, the Lens Injury Response Time Series (LIRTS) Viewer, which integrates all data to gather valuable insights from gene expression in CACs over the first 5 days PCS.
Conclusions: The LIRTS Viewer is useful for generating hypotheses related to PCO pathogenesis, as it reveals CAC gene expression dynamics, gene correlations, and biological pathways during the first 5 days following lens injury in an in vivo cataract surgery model.The initial framework for the LIRTS Viewer was designed in collaboration with a University of Delaware Computer Science Senior Capstone Design Team (CISC498/499) during the 2022–2023 academic year. We are grateful to Jaysheel Bhavsar for setting up the hosting of the LIRTS Viewer on the University of Delaware Biomix computational cluster and Karol Miaskiewicz, PhD, for assistance with the SSL certificate for the site. We are also thankful to Salil Lachke, PhD, and his team for linking the LIRTS Viewer to the iSyTE website. Deep and sincere gratitude goes to members of the Duncan lab for generating the data underlying the LIRTS and for beta-testing the resulting website.
Supported by a grant from the National Eye Institute, National Institutes of Health (EY015279); by grants from the Delaware INBRE (RRID:SCR_017696, NIGMS P20GM103446) providing access to the University of Delaware CBCB Bioinformatics Data Science Core Facility and use of the BIOMIX and BioStore computational resources; by a Shared Instrumentation Grant from the National Institutes of Health (S10OD028725); by the State of Delaware; and by the Delaware Biotechnology Institute.
Disclosure: S. Gorai, None; A.P. Faranda, None; M.H. Shihan, None; Y. Wang, None; M.K. Duncan, Non
OPENING THE BLACK BOX: AN ACTIVE APPROACH TO EXPLANATION METHODS IN MACHINE LEARNING
No full textWith the popularity of machine learning methods ever on the rise, it is more important than ever to create models that can be trusted. This dissertation presents an integrated active framework for explainable AI (XAI) that enhances the transparency of machine learning models, which are especially critical in high-stakes decision-making scenarios. The importance of model transparency is emphasized, as opaque decision-making processes in AI can lead to mistrust and ethical concerns. In this research, we include an in-depth exploration of the current landscape of explanation methods within the XAI field, paying particular attention to feature-importance-based methods, which provide insights into model decision-making by highlighting the relevance and impact of different features on model predictions. Through this exploration, we discuss key challenges and opportunities for improving explainability in machine learning.
In addition to the introduction of two novel feature-importance-based explanation methods, the core contributions of this dissertation are: (1) our investigation into the nature and solvability of the weakly supervised object localization (WSOL) problem and (2) an introduction of an integrated active framework for active XAI, wherein models are trained with an active awareness of their own explainability in relation to a chosen explainer. This approach ensures that models trained within this framework are not only accurate but also maximizes interpretability and transparency in its decision-making process (i.e., attain minimum risk with maximum explainability). This framework contributes to the ongoing efforts to enhance the trustworthiness and accountability of AI systems, paving the way for more responsible and ethical AI deployment in high-stakes domains
The effect of apiaceous vegetables on acrolein-related cardiovascular disease models
No full textTrabulsi, Jillian C.Atherosclerotic cardiovascular disease (ACVD) is the leading cause of death worldwide, affecting 26 million adults in the United States. ACVD is characterized by lipid plaque buildup within the arterial wall and comprises conditions such as coronary heart disease, heart attack, and stroke. Risk factors for ACVD include unhealthy diet, lipid dysregulation, hypertension, and smoking. Acrolein (ACR), a reactive aldehyde present in cigarette smoke, damages DNA and proteins, contributing to inflammation and endothelial dysfunction—both central to ACVD. Since smoking cessation has low success rates (10%), dietary interventions may offer a more viable risk reduction strategy. ☐ Apiaceous vegetables (API) have been used for centuries in traditional medicine and show potential for reducing inflammation, hyperlipidemia, and endothelial dysfunction. This dissertation examines API’s effects on ACR-related cardiovascular disease in vivo and in vitro. ☐ The first aim examined the effect of fresh celery and parsnip consumption on ACVD markers in ApoE-/- mice exposed to ACR. Thirty-three 12-week-old male and female mice were allocated to one of four groups: genetic control (GEN), ACR-only (ACR), and ACR groups with diets containing 21% or 42% API. ACR groups received intranasal ACR for four weeks. Mice in the GEN group received saline intranasally, while mice in the ACR and ACR + API groups received 0.5 mg/kg ACR intranasally for 4 weeks. Aortas, livers, and plasma were harvested at the conclusion of the intervention for analysis and showed API mitigated ACR-induced decreases in liver detoxification enzymes (GSTs), reduced triglyceride synthesis protein (SREBP-1), and modulated plasma cytokines, favoring an anti-inflammatory profile. ☐ The second aim evaluated the effect of 80% methanolic extract of fresh celery and parsnip on ACR-exposed human aortic endothelial cells (HAECs). HAECs were allocated into four groups: negative control group (NEG) which received saline only, positive control group (POS) which received 0.4 mM hydrogen peroxide, acrolein group (ACR) which received 30 μM ACR for 2 hours, and API group (API) which received 200 μg/mL API extract for 24 hours followed by 30 μM ACR for 2 hours. Exposure was performed three separate times with three dishes per group, and protein was extracted for analysis. Western blotting showed no significant changes in endothelial dysfunction or inflammasome-related proteins. ☐ The findings suggest that API may counteract ACR-induced inflammation and detoxification impairments in mice. Future studies using optimized ACR exposure conditions could clarify API’s role in ACR-induced ACVD.University of Delaware, Department of Health Behavior and Nutrition SciencesPh.D
Larval development and parasitism of emerald ash borer (Agrilus planipennis) in Oregon ash (Fraxinus latifolia) and European olive (Olea europaea): implications for the West Coast invasion
No full textThis article was originally published in Journal of Economic Entomology. The version of record is available at: https://doi.org/10.1093/jee/toaf008.
Published by Oxford University Press on behalf of Entomological Society of America 2025.
This work is written by (a) US Government employee(s) and is in the public domain in the US.The invasive emerald ash borer (Agrilus planipennis Fairmaire) (EAB) has been devastating North American ash (Fraxinus spp.) resources for over 2 decades. In its native range, EAB attacks and kills primarily stressed ash trees. In North America, however, EAB also attacks healthy trees of every Fraxinus species encountered, most recently Oregon ash (Fraxinus latifolia Benth.). Successful EAB development has also been reported in European olive (Olea europaea L.). The recent detection of EAB in Oregon puts the future of these 2 hosts into question, as little is known about EAB’s development in these species or how introduced biocontrol agents will respond. We conducted laboratory and field infestations of olive and ash in Delaware and Oregon to assess EAB development and associated parasitoid responses. We found no difference in the net population growth rate of EAB developing in Oregon ash versus green ash. However, these species supported significantly more population growth than olive, in which EAB net population growth rate was zero, with most larvae dying prematurely. Artificially infested olives were small, which may have negatively impacted phloem availability and larval survival. Future studies should be conducted investigating EAB development on larger olive material. Although no parasitism was observed in infested olive, as EAB larvae seldom reached life stages (third or fourth instars) susceptible to larval parasitism, late-instar larvae developing in Oregon ash were attacked by both Tetrastichus planipennisi Yang and Spathius galinae Belokobylskij and Strazanac, suggesting that biocontrol is a suitable option for this newly invaded region.This work was supported by USDA appropriated base funds (8010–22000-031D) and the Farm Bill (PPA 7721) Program (USDA APHIS Award # 23-8130-0971). This research was also conducted in part by an appointment to the Agricultural Research Service (ARS) Research Participation Program administered by the Oak Ridge Institute for Science and Education (ORISE) through an interagency agreement between the U.S. Department of Energy (DOE) and the U.S. Department of Agriculture (USDA). ORISE is managed by ORAU under DOE contract number DE-SC0014664. Opinions expressed in this paper do not necessarily reflect the policies and views of USDA, DOE, or ORAU/ORISE
The hydrogeological drivers of marsh migration
No full textMichael, Holly A.The Delmarva Peninsula is a hotspot for sea level rise making it especially vulnerable to coastal ecosystem change. The Peninsula is fringed by salt marshes and experiencing inland movement of marshes (marsh migration) as rising sea levels create wetter and saltier conditions, displacing upland freshwater vegetation. This dissertation investigates how sea-level rise and other hydrogeological mechanisms drive marsh migration by examining how terrestrial and oceanic forces interact to shape salinity dynamics and hydrological conditions at the marsh-upland boundary. Across three forest and three agricultural field study sites, space-for-time transects were instrumented with shallow wells, soil moisture sensors, and redox probes to capture fine-scale spatial and temporal variability in salinization, flushing, and flooding. Three years of high-resolution observational data reveal that terrestrial groundwater decline is a key, and often overlooked, driver of marsh migration. Inland groundwater declines reverse hydraulic gradients, inducing inland movement of saline groundwater and resulting in lateral salinization that exceeds storm salinization. In fact, storms often deliver enough rainfall to rapidly flush and freshen groundwater, challenging the assumption that storm events mainly contribute to salinization by overland saltwater flooding. However, storm surges remain important contributors to vertical salinization in shallow soils, where recovery can take months and groundwater flooding can produce prolonged anoxia. To further untangle the relative roles, timing, and interactions of compound drivers, a numerical model was developed to quantify salinization under both isolated (e.g., storm surge or drought alone) and sequential forcing scenarios (e.g., drought before storm surge or rain before storm surge). Findings show that the order of events significantly effects salinization; for instance, a storm surge following a drought leads to deeper salinization than the storm surge following a rain event. This work advances our understanding of the spatial and temporal complexity of hydrogeological stressors and provides new insight into the mechanisms driving marsh migration under a changing climate.University of Delaware, Department of Civil, Construction and Environmental EngineeringPh.D
Urban lighting variability: assessing energy end-use and policy impacts in New York City with remote sensing / ǂc by Lan Yu.
No full textDobler, GregoryThe policymaking cycle, as a nonlinear process, consists of four key components: agenda setting, policy design, policy implementation, and policy evaluation, with policy outputs feeding back into the design phase (De Marchi et al., 2016; Hill & Varone, 2021; Jann & Wegrich, 2017). With the development of evidence-based policy, this cycle has placed greater demands on the use of data, science, and research. Specifically, in the context of urban studies, remote sensing observation facilities have been widely used at varying spatial and temporal resolutions to support scientific research and provide evidence for policymaking. Meanwhile, Artificial Light at Night (ALAN), as an important component in urban studies (e.g., Globally, electricity accounts for about 20% of total energy consumption (International Energy Agency, 2020), nearly 20% of which is used for lighting and illumination (Pode, 2020; United Nations Environment Programme, 2013; Zissis, 2016), is related to studies on energy consumption, carbon emissions, ecology, public safety. However, policymakers have not fully recognized the impacts of lighting, and policy development in the lighting sector remains immature. ☐ Therefore, in this dissertation, we use the simplified policy cycle as a foundation and add an extended scientific evaluation layer to guide our exploration of lighting policymaking and evaluation in New York City (NYC). We first examine the LightsOut policy advocacy process in NYC, highlighting how scientific evidence has supported coalition efforts. We then use the Urban Observatory (UO) facility to remotely monitor ALAN, assessing the extent to which proximal remote sensing can serve as a reliable evaluation tool for lighting policies under two distinct research objectives. ☐ The first objective is to develop a real-time power outage detection model to support disaster management across three stages: (1) Pre-event prevention — identifying recurring issues; (2) Real-time identification — pinpointing the location, timing, and extent of power outages; and (3) Post-event planning — quantifying grid performance. The second objective is to apply a machine learning method to evaluate the Light-emitting diode (LED) changeover policy of the Bloomberg administration implemented in 2013 by identifying lighting types in the same scene across two different years (2013 and 2018) in Manhattan. ☐ Collectively, these three research projects contribute to a broader understanding of building compliance with NYC’s Lights Out policy, offer a complementary approach to enhancing grid resilience and response strategies, and provide a robust method for future lighting policy evaluation. This includes empirical insights into rebound effects, environmental justice implications, and policy interaction. ☐ In summary, our findings demonstrate the potential of integrating scientific methods, particularly proximal remote sensing, as an extended scientific layer to support a continuous policy learning cycle, specifically for lighting-related policies in NYC.Ph.D.University of Delaware, Energy and Environmental Policy Progra
Process optimization and online product quality analysis for solid-based continuous pharmaceutical manufacturing
No full textIerapetritou, Marianthi G.At present, the pharmaceutical industry is undergoing significant transformation. The production process of small molecule drug products, which is the most widely used type of drug, needs to become more efficient, flexible, agile, and intelligent to meet the stringent quality requirement and improve global healthcare benefits. Many new technologies emerged, such as continuous production, process analytical technology, real-time release testing, which provide an effective guarantee for efficient production and quality control of small molecule drug products. Based on the intersection of small molecule drug production and process systems engineering, this thesis proposes and solves a series of current problems and challenges. ☐ First, simulation-based optimization for continuous pharmaceutical manufacturing processes has been challenging. Surrogate-based optimization approaches have been widely adopted in industrial problems due to their potential to reduce the number of simulation runs required in the optimization process. The surrogate-based optimization framework has been extended to feasibility analysis in pharmaceutical manufacturing to characterize the design space. Most surrogate-based approaches for feasibility analysis are limited to the construction of a regression model for the feasibility function. In Chapter 2, we developed a framework with the feasibility problem considered as a classification problem, and additional stages introduced to improve local exploitation and global exploration. We illustrate the efficiency of the proposed framework on three test problems and implement it in a realistic case study describing the production of solid-based drugs using wet granulation, aimed to reduce the operation cost, improve product quality, and increase process flexibility and robustness. ☐ Second, residence time distribution (RTD)-based material tracking and quality control has been a promising tool in continuous pharmaceutical manufacturing. A system’s RTD is obtained through tracer experiments. However, RTD measurements are accompanied with uncertainties because of process fluctuation and variation, measurement error, and experimental variation among different replicates. Due to the strict quality control requirements of drug manufacturing, it is essential to consider RTD uncertainty and characterize its effects on RTD-based predictions and applications. Towards this end, Chapter 3 is focused on developing approaches for RTD uncertainty quantification and propagation to analyze the effects on downstream processes. Results depict probability intervals around the upstream disturbance tracking profile and the funnel plot, facilitating better decision-making for quality control under uncertainty. ☐ Third, despite thorough efforts in tracer selection, data acquisition, and calibration model development to obtain tracer concentration profiles for RTD studies, there can exist significant noise in these profiles. This noise can make it challenging to identify the underlying signal and get a representative RTD of the system under study. Such concerns have previously indicated the importance of noise handling for RTD measurements in literature. However, the literature does not provide sufficient information on noise handling or data treatment strategies for RTD studies. To this end, we investigate the impact of varying levels of noise using different tracers on measurement of RTD profile and its applications in Chapter 4. We quantify the impact of different denoising methods (time and frequency averaging methods). Through this investigation, we see that Wavelet Transform turns out to a good method for denoising RTD profiles despite varying noise levels. The investigation is performed such that the key features of the RTD profile (which are important for RTD based applications) are preserved. Subsequently, we also investigate the impact of denoising on RTD-based applications such as out-of-specification (OOS) analysis and RTD modeling. The results show that the degree of noise levels considered in this work do not significantly impact the RTD-based applications. ☐ In summary, feasibility-driven process optimization enables economical process operation while ensuring product quality through accurate characterization of process variability and feasibility, uncertainty quantification for RTD-based material diversion accounts for and mitigates risks due to variability, statistical data pretreatment ensures noise is accurately captured and addressed. These three works equip Pharma 4.0 with tools to manage product quality, supporting flexible yet high-quality manufacturing.University of Delaware, Department of Chemical and Biomolecular EngineeringPh.D