12466 research outputs found
Sort by
On 'Peri-peace' Photography: The Nexus Between Images and Empathy
Get PDFThis work is embargoed by the author and will not be publicly available until May 15, 2025.Photographic images have power. They can enflame, elicit, evoke and enthuse. While photographs of war and violence inundate society, images of peace are not as easily recognizable. This research project engages the medium of photography to navigate the field of peace and conflict resolution and illuminate creative aspects not widely utilized, so that concepts like empathy and compassion are not just prosocial behaviors but are seen as tangible stepping stones on the path to peace. Additionally, this study amplifies the language surrounding peace to include the concept of ‘peri-peace’ (2019), which I coined to elucidate the mercurial space between negative and positive peace. It also expands analysis to include the innovative practice of ‘peri-peace’ photography, which I developed to allow scholars and practitioners to recognize images of peace and enhance fragmented relationships. The aim is to synthesize various photographic approaches, via a qualitative and interpretative case study, and demonstrate that visual analyses can provide a more nuanced lens through which to ‘see’ peace and conflict, encourage empathy amongst adversaries and propel parties towards resolution and reconciliation.2025-05-1
Simple Synthetic Data as Source Domain for Transfer Learning to Remote Sensing as a Target Domain
Get PDFDeep Learning continues to grow as a prevalent toolset among multiple disciplines, including Remote Sensing and image analysis. Correspondingly, to more easily apply the deep neural networks to different subject matter domains, Transfer Learning, from natural image datasets, including ImageNet, has become a de-facto method for many Deep Learning applications, including Remote Sensing. However, such an approach may have limitations related to the differences on the characteristics of natural photographic image datasets and the characteristics of Remote Sensing. This study aims to determine if a fairly arbitrary, easily produced set of synthetic datasets can be iteratively developed and used for Transfer Learning for a typical Deep Learning task. We found this is readily and surprisingly feasible
Ground-based light curve follow-up validation observations of TESS object of interest TOI 3737.01
Get PDFContext: The study focuses on TESS Object of Interest (TOI) follow-ups, by light curve analysis of data taken by ground based observatories. TESS has provided a wealth of photometric data, but ground-based observations are essential to refine parameters and confirm planetary candidates.
Aims: Our primary goal is to enhance the understanding of a TOI by conducting a meticulous ground-based follow-up. Through precise light curve analysis, we aim to confirm the planetary nature of TOI and refine its orbital parameters, ultimately contributing to the characterization of a possible exoplanet.
Methods: We collected high-quality ground-based photometric data using observatory telescopes and instruments. By reducing and analyzing these data, we carried out a light curve analysis to validate TOI's planetary status
Problematic Substance Use and the Pretrial Period: Risk- and Needs-Based Supervision Strategies
Get PDFThe United States currently leads the developed world in incarceration rates, driven primarily by pretrial populations. Pretrial reform advocates have argued for the implementation of evidence-based pretrial strategies—such as pretrial risk assessments and pretrial supervision—to curb high rates of pretrial detention. Reform efforts have focused less on alternative strategies to address defendants’ criminogenic needs. Substance use remains one of the most common criminogenic needs in the criminal-legal system, with rates far exceeding those of the general population. Research on pretrial reform efforts focuses primarily on strategies to reduce risk, limiting our understanding of needs-based interventions during the pretrial period. Monitoring strategies for pretrial defendants with problematic substance include drug testing and mandated treatment. Few studies to date have examined the effectiveness of these strategies on pretrial outcomes or examined alternative approaches for pretrial defendants with problematic substance use. This dissertation addresses these limitations using a multi-method approach across three studies. In the first study, I examined the effectiveness of pretrial drug testing on pretrial outcomes for defendants with problematic substance use. Findings showed that pretrial defendants with a drug testing condition during their supervision period were at a higher risk of pretrial failure compared to defendants without a drug testing condition. Further, defendants with a drug testing condition showed a higher likelihood of rearrest, rearrest on drug-specific charges, and any failure compared to defendants without drug testing. In the second study, I conducted qualitative interviews with a judge, pretrial services officers, and pretrial defendants to examine perceptions of drug testing and needs of defendants with problematic substance use. Using an inductive content analysis strategy, findings showed that the judge and defendants appreciate the accountability drug testing provides. However, drug testing raised several barriers for defendants’ successful completion of pretrial supervision, including issues related to finances, childcare, transportation, and lack of information. Despite these barriers, the judge did not see a feasible alternative to pretrial drug testing. In the third study, I conducted a pilot randomized controlled trial of a brief assessment and referral to treatment strategy for defendants with problematic substance use. The intervention was based off SAMHSA’s Screening, Brief Intervention, and Referral to Treatment (SBIRT) model, and included a 10-item substance use assessment and personalized referral to treatment. The intervention aimed to increase participation in the decision-making process surrounding treatment and thus improve the likelihood the defendant would seek out treatment. Findings showed no differences in pretrial failure outcomes, potentially due to the small sample size and low outcome rates. However, descriptive explorations of the data showed promise for testing the intervention on a larger scale. Overall, findings point to the lack of existing suitable strategies for managing defendants with problematic substance use during the pretrial period. In order to address the substance use-related needs of pretrial defendants, pretrial agencies may need to reassess policies surrounding drug testing. Future research is needed to examine the feasibility and effectiveness of such policy changes and to further examine the effectiveness of other needs-based strategies that may improve the experiences and outcomes of pretrial defendants with problematic substance use
THE PRODUCTION AND PERCEPTION OF EMPHASIS IN QASSIMI ARABIC
Get PDFThis work is embargoed by the author and will not be publicly available until December 2025.This dissertation explores emphasis effects (EE) in Qassimi Arabic (QA), examining whether EE functions as a phonetic or phonological process. EE is a well-documented phenomenon in Arabic linguistics, involving the influence of emphatic consonants on neighboring segments (Ghazali, 1977; Card, 1983; Davis, 1995; among others). The study also investigates emphasis perception in QA, specifically whether EE cues assist native QA listeners in identifying preceding or following consonants as emphatic or plain.As prior research exploring EE in various Arabic varieties has revealed variation among them, and limited research exists on emphasis perception by native Arabic listeners, this dissertation addresses these gaps by examining EE production and emphasis perception in the understudied variety of QA. In the production experiment, dynamic aspects of leftward and rightward EE on QA vowels were examined by analyzing second formants (F2) at 11 temporal points. Results indicate that leftward EE had a categorical effect on non-high vowels [a] and [aː], as well as the high front vowel [i], impacting them throughout their duration, providing evidence for it being a phonological process in QA. In contrast, rightward EE primarily affected the vowel onset, suggesting it as a gradual phonetic process rather than a categorical phonological one. In the perception experiments, the perceptual correlates of emphasis in QA were investigated using the gating paradigm (Grosjean, 1980). Native QA listeners accurately identified the following consonant using leftward EE cues, even within the shortest gate containing one-third of the vowel, indicating proficiency in using leftward EE cues throughout the vowel. However, for rightward EE cues, accuracy in identifying the preceding consonant as emphatic or plain improved significantly only when the entire vowel duration was audible. These findings align with the production experiment, confirming leftward EE as a phonological process and rightward EE as a phonetic process. The dissertation’s results have implications for understanding EE and emphasis perception in QA, emphasizing the importance of considering both phonological and phonetic processes when investigating EE and highlighting the significance of coarticulatory information in rightward emphasis perception. This nuanced understanding advances research into emphasis across Arabic varieties and Semitic languages.2025-12-1
Performance-Aware Coarse-Grained Reconfigurable Logic Accelerator for Deep Learning Application
No full textDeep neural networks (DNNs) are widely deployed in various cognitive applications including computer vision, speech recognition, and image processing. The surpassing accuracy and performance of deep neural networks come at the cost of high computational complexity. Therefore, software implementations of DNNs and convolutional neural networks (CNNs) are often hindered by computational and communication bottlenecks. As a panacea, numerous hardware accelerators are introduced in recent times to accelerate DNNs and CNNs. Despite effectiveness, the existing hardware accelerators are often confronted by the involved computational complexity and the need for special hardware units to implement each of the DNN/CNN operations.To address such challenges, a reconfigurable DNN/CNN accelerator is proposed in this work. The proposed architecture comprises nine processing elements (PEs) that can perform both convolution and arithmetic operations, through run-time reconfiguration with minimal overhead. To reduce the computational complexity, we employ Mitchell's algorithm, which is supported through low-overhead coarse-grained reconfigurability in this work. To facilitate efficient data flow across the PEs, we pre-compute the dataflow paths and configure the dataflow during the runtime. The proposed design is realized on a field-programmable gate array (FPGA) platform for evaluation
Understanding Lineage Plasticity in Non-Small Cell Lung Cancers Using Monoclonal Populations of Cells
No full textThis thesis has been embargoed for 5 years. It will not be available until December 2026 at the earliest.Lung cancer diagnoses account for roughly 1.6 million deaths worldwide. It is the leading cause of cancer deaths in the US for men and the second leading cause for women. Due to late detection lung cancers have a high mortality rate, with survival rates of less than twenty percent after 5 years. While the introduction of targeted treatments has impacted survival for subgroups of lung cancer patients, especially those affected by Non-Small Cell Lung Cancer (NSCLS), the effect of these anti-cancer drugs is often temporary, and resistance is routinely acquired through different mechanisms. Lineage plasticity or the ability a tumor cell has to undergo transformation from one histology type to another is emerging as a potential mechanism of a resistance to treatment. For example, tumors of epithelial origin, like lung adenocarcinomas, can transform into neuroendocrine tumors and even acquire histological characteristics of Small Cell Lung Cancers (SCLC). However, understanding the molecular events associated with this transformation still remains an unmet need in oncology. Using a commercially available lung cancer model, this work aimed at identifying morphological and molecular traits associated with neuroendocrine transdifferentiation in lung cancer. We use single cell cloning and expansion techniques to establish monoclonal populations of cells from a heterogeneous cell line model. A total of 61 clones were established and successfully expanded and observed over time to capture morphological characteristics, signal transduction events, and assess level of expression of neuroendocrine markers. Lastly, using previously established protocols, for selected clones we assessed their ability to undergo lineage change after perturbation of their growth conditions. While clonal evolution drives acquired resistance in cancer, the establishment of monoclonal population of cells from heterogenous models may provide novel insights on the mechanisms of resistance to anti-cancer treatments and for understanding lineage plasticity in cancer.2026-12-0
Bridging the Gap: Rhetoric of Telecommunications Compliance Policies
No full textAs the telecommunications industry expanded and grew since the invention of the telephone, the U.S. Congress has passed several privacy laws to protect consumer’s personally identifiable information that telecommunications providers collect during the regular course of their business, but they have also passed several laws and regulations that require compliance with legally requested live and stored personally identifiable information from law enforcement. Telecommunications providers struggle with communicating their legally required compliance with law enforcement requests while also communicating their measures to protect their customers’ privacy and personally identifiable information. To evaluate the balance between these two conflicting requirements, this study analyzes the rhetoric of telecommunications providers’ legal compliance policies and discusses the convoluted and often misleading language used to explain these compliance policies to their customers
Negative Urgency, Emotion Regulation, and Stress Generation: An Experience Sampling Study
Get PDFThis dissertation is a two-part ecological momentary assessment (EMA) study that focuses on two transdiagnostic elements surrounding negative affect and avoidant emotion regulation in a clinical analog sample: negative urgency (NU) and stress generation. NU is the tendency to engage in impulsive behavior in response to strong negative emotions and has been shown to characterize various mental health problems. Emerging prospective evidence suggests that NU predicts avoidance of negative affect, which itself is a well-established risk and maintenance factor for a wide range of psychopathology. However, only a few EMA studies have examined the relationship between NU and avoidance, and they have only evaluated trait-level NU as a predictor of specific avoidance strategies. Investigations of trait- and state-level NU and the ways in which both may influence the relationship between NA and broad avoidant emotion regulation are needed to advance our understanding of the role of NU in psychopathology and to facilitate the potential development of related interventions. Stress generation is the process by which one’s own thoughts and behaviors directly contribute to the development of stressors. Prospective studies show that people’s experience of symptoms of emotional disorders precede the experience of more dependent stressors, and avoidant emotion regulation is one proposed mechanism of this process. A few longitudinal studies show that avoidance predicts dependent stressors months to years later, but only one EMA study has evaluated these associations on a more momentary basis. That study found that NA did not interact with avoidance strategies to predict dependent stressors; however, NA, avoidance, and stress generation were evaluated at different time points, spanning several hours, and in a healthy sample, which limits the conclusions that can be drawn. The present dissertation project used an EMA study to evaluate these issues involving NU and stress generation in relation to avoidance of negative affect in a sample of 84 adults who endorsed high levels of emotion dysregulation, relative to the general population. Participants completed a screener and background survey at baseline, followed by an EMA study involving four surveys per day for two weeks. Background surveys included a measure of emotion regulation, and EMA surveys included state measures of NA, NU, avoidant emotion regulation strategies, and dependent stressors. Hypotheses were evaluated with multi-level generalized linear models. Study 1 evaluated the interactions of trait and state NU with state NA on avoidance strategies. Results demonstrated main and interaction effects on same- and next-time-point avoidance that varied by state versus trait assessment of NA and NU, and suggested that a recently introduced state measure of NU may be more predictive of avoidance than an established trait measure. Study 2 investigated the interaction between NA and avoidance strategies on stress generation. Findings showed that NA and avoidance differentially predicted dependent stressors assessed at the same timepoint and next timepoint. Results indicate that distress and avoidance may be a consequence and a generator of stressors. The strengths and limitations of findings, and future directions are discussed
A Proteomic Profile of Imidacloprid and its Active Metabolites Desnitro-Imidacloprid and Imidacloprid-Olefin in Human Neural Cells
Get PDFNeonicotinoids are a popular class of pesticides around the world and have been shown to bind both insect and mammalian nicotinic acetylcholine receptors (nAChRs). Imidacloprid (IMI), is amongst the most commonly used neonicotinoids in agriculture and domestic, non-industrial, applications but shows strong toxic effects in many organisms. In addition, the potential for toxicity from its two main metabolites desnitroimidacloprid (DN-IMI) and imidacloprid-olefin (IMI-olefin), are not yet defined despite recent evidence that they can bind to the mammalian nAChR. In this study, we used Lund Human Mesencephalic (LUHMES) cells, as a human cell line model of dopaminergic neurons to test the effects of IMI and its metabolites. Cells were treated with 50μM IMI, DN-IMI, and IMI-olefin for 48 hours, and then examined for proteomic change using liquid-chromatography electrospray ionization mass spectrometry (LC-ESI MS/MS). Bioinformatic analysis using Gene Ontology (GO) and enrichment analysis were performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome, and WikiPathways databases. Our results provide novel insight into convergent as well as differential molecular effects by IMI, DN-IMI, and IMI-olefin in neural cells. These studies begin to explore pathways of neonicotinoid neurotoxicity leading to neuro disease in exposed individuals