11549 research outputs found
Sort by
IMPROVING SEGMENTED STYLE TRANSFER VIA BLENDED PARTIAL CONVOLUTION
No full textStyle transfer aims to render the content of an image in the style of another, but applying this technique to specific segments within an image poses significant challenges, particularly in achieving seamless integration between styled and non-styled regions. In this thesis, we explore potential improvements to segmented style transfer by introducing blended partial convolution into the processing pipeline. Specifically, we evaluate three techniques: replacing traditional style transfer mechanisms with partial convolution, incorporating mask dilation in partial convolution, and applying mask feathering both prior to encoding and within the decoder. Systematically assessing these methods identifies their contributions to enhancing the style adaptation within designated segments, reducing boundary artifacts, and improving overall visual coherence. Preliminary results indicate that these techniques collectively have the potential to offer a more refined tool for applications in digital art, augmented reality, and image editing. This work advances the field of style transfer by addressing key limitations in segmented applications and provides a foundation for future research in localized style adaptation
PREDICTING AND MAPPING THE GEOGRAPHIC DISTRIBUTION OF GLAUCOMA IN THE UNITED STATES: THE ROLE OF SOCIAL DETERMINANTS USING THE ALL OF US DATASET
Get PDFVision impairment and eye diseases are significant public health concerns in the United States and globally. Glaucoma, a chronic and progressive disease, is one of the leading causes of irreversible blindness worldwide. In the U.S., more than three million individuals are estimated to be affected, with projections indicating a rise as the population ages. While clinical and genetic factors influencing glaucoma onset and progression have been extensively studied, growing evidence suggests that environmental exposures, socioeconomic status, and lifestyle factors also play a crucial role. With disparities in healthcare access and outcomes based on socioeconomic factors, it is crucial to explore how these factors, alongside genetic predispositions, affect glaucoma onset and progression. Addressing these gaps could lead to more targeted interventions, improving outcomes for vulnerable populations. This study aims to bridge this gap by leveraging machine learning techniques to build predictive models for glaucoma risk. By utilizing demographic information and Social Determinant of Health (SDOH) from the All of Us dataset, this research develops a comprehensive framework for glaucoma prediction. These models allow for an improved understanding of how SDOH influences glaucoma risk, helping to inform early detection strategies. The optimized Decision Tree model, tuned with GridSearchCV, was the best-performing model for this prediction task, achieving an accuracy of 67.87%. For class 0 (Non-Glaucoma), it yielded a precision of 0.71, recall of 0.52, and an F1 score of 0.60. For class 1 (Glaucoma), the model achieved a precision of 0.66, recall of 0.81, and an F1 score of 0.73. Feature importance analysis identified age as the most significant predictor, followed by race and the affordability of seeing an eye doctor. In contrast, factors such as affordability of specialist care and copay affordability had minimal impact. The findings from this study have broader implications for enhancing glaucoma risk assessments and healthcare interventions. Additionally, the methodological approach can be applied to other complex diseases, contributing to a more equitable and informed public health approach. By emphasizing social determinants, this research takes a promising step toward reducing the burden of glaucoma and advancing the goals of precision medicine
Moral Enquiry Meets Artificial Intelligence: Considering Influences of Interactive Algorithmic-based Ethical Decision Making on Agentive Wellbeing
No full textThis version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: http://dx.doi.org/10.1007/s13347-025-00892-7
The article is has a one-year embargo and it will be available after June 2026.Artificial intelligence (AI) has become an undeniable paradigmatic influence on human beliefs, attitudes, intentions, and behaviors. This article analyzes research that offers insights into risks related to contextual interaction with AI for human well-being in morally pluralistic organizations. Building on previous research that connects AI use with an increased risk of the sociopsychological phenomenon known as moral disengagement, we argue that AI created by developers who utilize rival versions of moral enquiry can activate moral disengagement mechanisms and shift a person’s perspective regarding responsibility by sanctioning attitudes, intentions, and behaviors that would otherwise be inconsistent with the person’s moral beliefs. We specifically examine human interaction with Generative and Predictive AI and their risks for increasing cognitive dissonance, cognitive distress, and self-harm. We also intentionally clarify important differences between AI-to-human interaction with human-to-human interaction. Finally, we recommend how to identify problematic interactions with AI and recommend ways to reduce a risk to human flourishing
METAMORPHIC TESTING FOR FAIRNESS EVALUATION IN LARGE LANGUAGE MODELS
Get PDFLarge Language Models (LLMs) have made significant progress in Natural Language Processing, yet they remain susceptible to fairness-related issues, often reflecting biases from their training data. These biases present risks, mainly when LLMs are used in sensitive domains such as healthcare, finance, and law. This research proposes a metamorphic testing approach to uncover fairness bugs in LLMs systematically. We define and apply fairness-oriented metamorphic relations (MRs) to evaluate state-of-the-art models like LLaMA and GPT across diverse demographic inputs. By generating and analyzing source and follow-up test cases, we identify patterns of bias, particularly in tone and sentiment. Results show that tone-based MRs detected up to 2,200 fairness violations, while sentiment-based MRs detected fewer than 500, highlighting the strength of this method. This study presents a structured strategy for enhancing fairness in LLMs and improving their robustness in critical applications
DEVELOPMENT OF EMPATHETIC MANAGERS TO SUPPORT DISABLED EMPLOYEES UTILIZING SIMULATION-BASED LEARNING
Get PDFEmpathy is a crucial skill for effective leadership, particularly in diverse workplaces where understanding and addressing employees’ needs contribute to organizational success. This study examines the impact of simulation-based learning on developing empathy in business students. Grounded in Kolb’s experiential learning theory (1984), the research explores how immersive simulations enhance students' empathy by engaging them in real-world workplace scenarios.
A mixed-methods approach was used to assess the effectiveness of simulation-based empathy training. Undergraduate business management students participated in a simulation addressing mobility challenges, reflecting on accommodations for employees with disabilities. Pre- and post-tests measured changes in empathy levels, while qualitative responses provided deeper insights into students' perspectives. The results showed a significant increase in self-reported empathy scores (p = 0.015), with participants demonstrating an improved ability to recognize and respond to employees’ emotional needs. Qualitative findings further indicated a heightened awareness of inclusive leadership and a stronger commitment to fostering supportive work environments.
This study affirms the value of simulation-based learning in developing empathy among future managers. Findings highlight how Kolb’s experiential learning cycle—encompassing concrete experience, reflective observation, abstract conceptualization, and active experimentation—effectively enhances emotional intelligence in management education. While the simulations fostered short-term empathy growth, future research should investigate long-term skill retention. Additionally, expanding simulations to include other diversity factors, such as race, gender, and socioeconomic status, could further promote empathy and inclusivity in leadership training
Executive Summary: PRESS ON Hands-Only CPR: A Community Education Project
No full textPRESS ON Hands-Only CPR:
A Community Education Project
There were 139,822 out-of-hospital cardiac arrests (OHCA) captured in the 2023 Cardiac Arrest Registry to Enhance Survival (CARES) report, and sadly, only 10.2% of those people survived to hospital discharge. The CARES registry collects data from 37 states and 11 additional community-based sites, which represent approximately 56% of the United States population. Overall, approximately 350,000 OHCA cases occur annually in the United States (Blewer et al., 2024). Poor survival rates for OHCA can be attributed to several factors, including pre-existing health conditions, the time of first chest compressions, whether an automatic external defibrillator (AED) is used, emergency medical response times, and the availability of a receiving hospital to provide evidence-based care. There are opportunities to improve outcomes along the spectrum of care; however, the most important factors are the early activation of 911, the rapid initiation of chest compressions, and the use of an AED if available. Studies have shown that bystander CPR can double the odds of surviving OHCA (Cheng et al., 2020; Dainty et al., 2022). Currently, only 40% of adult victims of OHCA receive cardiopulmonary resuscitation (CPR), and 29% had an AED applied before paramedics arrived (RACE CARS Trial, n.d.). Those rates are lower in rural areas, low-income neighborhoods, and minority populations (Ashburn et al.,2021; Pu et al., 2023). Despite extensive funding and education initiatives, the United States is still falling short of the Healthy People 2030 (Office of Disease Prevention and Promotion, n.d.) goal that 45.1% of all OHCA victims receive bystander CPR. We must continue to create opportunities to empower bystanders with the skills and confidence to act when cardiac arrest occurs, thereby improving this measure and its related outcomes.D.N.P
The Separation and Fragmentation of 15-deoxy, Δ12,14-prostamide J2 as an Anti-Tumor Therapeutic Using ESI-MS/MS
No full textAccording to the American Cancer Society, colon cancer is the third most diagnosed cancer in people of all ages and the second leading cause of death due to cancer, in the United States. Most cancer treatments involve chemotherapy, which kills the rapidly growing cancer cells, along with healthy cells including blood-forming cells. A class of prostaglandins and prostamides was found to have the capability to cause endoplasmic-reticulum stress-induced apoptosis in cancer cells with minimal effects on non-cancerous cells. This study focuses on 15-deoxy, Δ12,14-prostamide J2, or 15d-PMJ2, an effective molecule shown to eliminate cancerous colon cells. The 15d-PMJ2 will be delivered through micelles due to its nature of being hydrophobic and simply injecting the drug intravenously would be problematic due to solubility issues. To better characterize the drug for future studies, mass spectrometry is being used to analyze and assign the fragments of the molecule when subjected to collision-induced fragmentation into smaller ions. These fragments can be used as a fingerprint for 15d-PMJ2 which provide high selectivity for detecting 15d-PMJ2 within a complex mixture
Types of Physical Activity Pre-Pregnancy Women Engage In
No full textSignature Honors Project 2024
East Carolina University
Types of Physical Activity Pre-Pregnancy Women Engage In
Aubrey C. VanWynsberg, Bhibha M. Das
Being physically active leads to a healthier lifestyle. However, there is a stigma that to be physically active, you must participate in vigorous workouts. Due to this stigma, many groups of people are driven away from physical activity, and a very prominent one is pregnant women. Physical activity is beneficial for pregnant women before, during, and after pregnancy. In fact, physical activity can also be beneficial for the fetus during pregnancy. Purpose: This study aimed to identify types of physical activities pre-pregnancy women engaged in, as well as physical activity recommendations for pregnant women during and postpartum. Methods: Existing data was analyzed to identify what types of physical activities were most frequent among women pre-pregnancy. Furthermore, existing literature was also reviewed to determine the benefits, risks, and recommendations of exercise before, during, and after pregnancy for the mother and the fetus. Results: A total of 425 participants from the Southeastern, rural United States self-reported physical activities using a questionnaire. Out of the participants, 76% were White, 19% were Black, 4% were Asian, and 1% were categorized as other. The average age of the participants was 30 ± 4.44 years old with an average pre-pregnancy BMI of 25.92 ± 4.91 kg/m2. The most common activities were moderate to high intensity as walking (21%) and jogging (15%) were the two highest activities among pre-pregnant women in the study. Furthermore, while walking and jogging were found to be common aerobic exercises, strength training was the third highest activity (13%). Literature suggests the continuation of these physical activities during pregnancy as they can help to counteract the emotional and physical side effects of pregnancy. For example, post-partum depression, anxiety, weight gain, gestational diabetes, birth complications, and fetal neurocognitive development. It was found that there is a lack of information on what types of physical activities are safe for pregnant women as well as a lack of education and prescription from healthcare providers. Conclusion: The majority of pre-pregnancy activities were moderate to high-intensity exercises with walking, jogging, and strength training being the most common. There are numerous benefits associated with physical activity before, during, and after pregnancy. Therefore, healthcare providers should recommend physical activities and prescribe exercises specific to the individual they are working with. Physical and emotional risks associated with pregnancy can be positively influenced by the type, duration, and frequency of exercise. Exercise can be catered to the individual’s needs such as those with a predisposition to pregnancy complications. Future research would compare what types of physical activities these women engage in during pregnancy as well as the benefits they experience
FROM SCARS TO STRENGTH: UNVEILING THE INTERSECTION OF ACES, SOCIAL DETERMINANTS, AND RESILIENCE IN ADULT INPATIENT REHABILITATION
No full textThis dissertation explores the complex relationship between adverse childhood experiences (ACEs), protective and compensatory experiences (PACEs), and their influence on resilience in primary care contexts and among adult inpatient rehabilitation patients. While ACEs are known to be linked with negative health outcomes, little attention has been given to the care for patients following a screening of ACEs in primary care, and protective factors are rarely if ever, considered in healthcare contexts. Furthermore, the role of ACEs, PACEs, social determinants of health (SDoH), and resilience in adult inpatient rehabilitation contexts are relatively unknown. The primary objective of this research is to deepen the scientific understanding of how ACEs, PACEs, and SDoH interact to shape resilience when individuals face medical conditions requiring inpatient rehabilitation. By examining this dynamic, this dissertation investigates clinical practices, research methodologies, and health equity policies, ultimately enhancing strategies for screening, triage, and intervention in rehabilitation units. The dissertation employs a dual approach, combining systematic reviews with original quantitative empirical analysis. It is organized into six chapters: the first chapter introduces the concept of resilience in rehabilitation and its relationship to ACEs and PACEs. The second chapter provides a systematic review of ACE screening practices in adult primary care settings, focusing on the types of clinical responses (e.g., resources, referrals, interventions) that follow positive ACE screenings. The third chapter systematically investigates additional psychosocial factors, both protective and adverse, that are assessed alongside ACE screenings in primary care, framed through a socio-ecological resilience model. The fourth chapter details the methodology of the original quantitative research. The fifth chapter presents the results of the empirical study, which examines the interaction between SDoH, protective factors, resilience, and behavioral health in adult inpatient rehabilitation patients with a history of ACEs; and the sixth chapter synthesizes the key findings and their contributions to the field, focusing on the integration of protective experiences in ACE screening interventions. It employs data visualization techniques to analyze health disparities across North Carolina counties based on the study participants' sample, offering recommendations for improving patient care and health equity. The research ultimately advances an understanding of the interplay between childhood adversity, protective factors, and health outcomes, offering strategies for improving resilience and health equity in diverse inpatient rehabilitation populations
DIVING INTO THE REALM OF STEM AND POLITICS
No full textAccording to the 2018 National Study of Learning, Voting, and Engagement (NSLVE) report, Science, Technology, Education, and Mathematics (STEM) majors do not vote at rates as high as non-STEM majors at East Carolina University and around the country. We believe this is a pressing issue that should be addressed. Going back to the fundamentals of education, we created the first ever East Carolina University Honors College Seminar developed by students, for students, to address the lack of voting among STEM majors. The seminar is currently being taught this semester (Spring 2023) by Dr. Mosier, a professor in the Political Science Department. The course emphasizes the importance of scientifically sound civic engagement and political participation for STEM majors and other STEM interested students.
As a result of new research, we are shifting our efforts to the East Carolina University student body, but still with a focus on STEM-related topics. By the Fall of 2023, we plan to have this topic implemented into the COAD 1000-Student Development and Learning in Higher Education course.