95038 research outputs found
Sort by
Mask wearing as a prosocial behavior: Proposing and testing the moral norms activation model
Get PDFThe aim of this study was to develop and test a model of prosocial prevention behavior during COVID-19, termed the Moral Norms Activation Model (MNAM). This model examines how moral norms, influenced by awareness of consequences, predict prosocial prevention behaviors, such as mask-wearing, and the role of perceived severity and collective orientation as moderating factors. We conducted a survey during the early months of the COVID-19 pandemic with a nationally representative sample of U.S. adults (N = 8,778). The survey measured awareness of consequences, moral norms, anticipated guilt, perceived severity, collective orientation, and self-reported mask-wearing behavior. A series of regressions was used to test the proposed model and interactions. Findings supported the MNAM, demonstrating that awareness of consequences was a significant direct predictor of moral norms. These moral norms, in turn, predicted prosocial prevention behavior, mediated by anticipated guilt. The moderating effects of perceived severity and collective orientation were also significant, reinforcing the strength of the association between moral norms and behavior in individuals with high collective orientation and greater perceived severity. The results highlight the critical role of moral norms and anticipated guilt in promoting prosocial health behaviors during a collective health crisis. The MNAM provides a novel framework for understanding how individual psychological processes contribute to public health behaviors. These findings suggest that public health campaigns emphasizing moral responsibility and awareness of consequences could enhance compliance with preventive measures
Diagnostic Performance of Publicly Available Large Language Models in Corneal Diseases: A Comparison with Human Specialists
Get PDFBackground/Objectives: This study evaluated the diagnostic accuracy of seven publicly available large language models (LLMs)—GPT-3.5, GPT-4.o Mini, GPT-4.o, Gemini 1.5 Flash, Claude 3.5 Sonnet, Grok3, and DeepSeek R1—in diagnosing corneal diseases, comparing their performance to human specialists. Methods: Twenty corneal disease cases from the University of Iowa’s EyeRounds were presented to each LLM. Diagnostic accuracy was determined by comparing LLM-generated diagnoses to the confirmed case diagnoses. Four human cornea specialists evaluated the same cases to establish a benchmark and assess interobserver agreement. Results: Diagnostic accuracy varied significantly among LLMs (p = 0.001). GPT-4.o achieved the highest accuracy (80.0%), followed by Claude 3.5 Sonnet and Grok3 (70.0%), DeepSeek R1 (65.0%), GPT-3.5 (60.0%), GPT-4.o Mini (55.0%), and Gemini 1.5 Flash (30.0%). Human experts averaged 92.5% accuracy, outperforming all LLMs (p < 0.001, Cohen’s d = −1.314). GPT-4.o showed no significant difference from human consensus (p = 0.250, κ = 0.348), while Claude and Grok3 showed fair agreement (κ = 0.219). DeepSeek R1 also performed reasonably (κ = 0.178), although not significantly. Conclusions: Among the evaluated LLMs, GPT-4.o, Claude 3.5 Sonnet, Grok3, and DeepSeek R1 demonstrated promising diagnostic accuracy, with GPT-4.o most closely matching human performance. However, performance remained inconsistent, especially in complex cases. LLMs may offer value as diagnostic support tools, but human expertise remains indispensable for clinical decision-making
Examining and classifying reasons for missing viral loads among adults living with HIV: An extended outcome investigation and ascertainment approach in Western Kenya
Get PDFGaps in HIV RNA monitoring persist globally impeding the ability to determine clinical progress and outcomes. This study systematically evaluated provider (e.g., guideline non-adherence), system (e.g., laboratory error) and participant-level (e.g., refusal) drivers of missed viral load (VL) monitoring measurements among people with HIV in Kenya. Adults aged 18-65 years were followed across five health facilities in Kenya as part of a clinical trial (NCT#02338739) where HIV RNA monitoring was done routinely. Instances of missed VL despite being indicated per Kenyan guidelines were identified. An algorithm for assessing root causes of missing HIV RNA was developed and generalized linear models estimated the risk ratios (RR) for participant-level characteristics associated with missed viral load. Among 1,754 participants (66% female), the prevalence of missed viral load in year one and two was 24.4% and 29.4%, respectively. Drivers for missed viral load measurements included loss to follow up (51.5% in year one and 57.8% in year two), clinician non-adherence with guidelines (36.7% in year one and 32.2% in year two), unknown (10.3% in year one and 8.6% in year two), and requested but not collected (1.5% in year one and 1.3% in year two). Participants aged < 24 years (RR 2.27, 95% CI: 1.66-3.12), those with higher socioeconomic status (RR 1.47, 95% CI: 1.03-1.91), receiving HIV treatment at a rural clinic (RR 1.22, 95% CI: 1.02-1.46) and with advanced HIV disease (RR 2.39, 95% CI: 1.52-3.73) were more likely to miss VL monitoring. Missed routine viral load monitoring remains high, primarily due to loss to follow-up, and may substantially alter suppression estimates. Sustainable approaches to keep people with HIV engaged in care, alongside strengthening providers' clinical practices and alignment with national guidelines, are necessary for optimizing viral monitoring and accurately assessing viral suppression within public health systems
Geriatric ocular trauma and mortality: A retrospective cohort study
Get PDFPurpose The objective of this investigation is to evaluate the 5-year mortality of geriatric patients who have sustained eye injuries. Design This retrospective cohort study included patients aged 65 years or older who had histories of either ocular trauma or age-related nuclear cataracts. Subjects and controls: Patients with ocular trauma constituted the study group, while those with a history of cataracts served as controls. Methods Data from the I2B2 Carolina Data Warehouse were analyzed. Patient demographics were collected, and the outcomes of interest were the overall mortality rate and annual mortality rates over a 5-year period. Chi-squared tests were utilized for the comparison of mortality data. Main outcomes and measures The primary outcomes were overall mortality rates and annual mortality rates expressed as percentages. Results The study group consisted of 602 patients who had suffered ocular trauma. The control group included 1066 patients of similar age who had been diagnosed with age-related nuclear cataracts at some point in their lives. Among the study group, 74 patients died within 5 years, while 69 patients in the control group died within the same timeframe, resulting in a study group mortality rate of 11.30% and a control group mortality rate of 6.47%. For patients with ocular trauma, the annual mortality rates were 4.15%, 2.60%, 1.96%, 2.54%, and 0.56%, respectively. For the control group, the annual mortality rates were 1.03%, 1.70%, 1.64%, 0.88%, and 1.38% respectively. Conclusion The study suggests that geriatric patients who have experienced ocular trauma are at a higher risk of mortality compared to age-matched controls without such injuries. These findings highlight the necessity of identifying the causes of geriatric periorbital trauma and underscore the importance of close patient follow-up to improve outcomes
Understanding PrEP misconceptions and their impact on PrEP initiation and use among pregnant and lactating women in Malawi
Get PDFMother-to-child transmission (MTCT) of HIV remains a challenge in Eastern and Southern Africa. Oral pre-exposure prophylaxis (PrEP) is a powerful tool to reduce MTCT, but women face barriers to effective use including those related to inaccurate comprehension of PrEP To understand women’s misconceptions about PrEP and their potential impact on PrEP use, we conducted interviews with 33 pregnant and lactating women in Malawi using PrEP, and ten PrEP counselors and clinicians. The results indicate that, although pregnant women generally understood PrEP’s features and functions, many held misconceptions that persisted over the course of their PrEP use and impacted their perceptions and use of the medication. Some women erroneously believed that PrEP could treat and prevent sexually transmitted infections other than HIV, which motivated some to keep taking PrEP while motivating others to stop using PrEP once their STI had resolved. Some were confused about PrEP’s function, with some believing it was the same as antiretroviral therapy for HIV treatment, and others believing that PrEP could be used for overall enhancement of health. Rarer misconceptions included fears that PrEP was connected to satanic practices, could cause cancer, or was solely for individuals engaged in sex work. These misconceptions stemmed from a mix of prior knowledge, societal influences, and miscommunications during counseling sessions. Ensuring accurate knowledge and addressing common misconceptions head-on is crucial to promote continued PrEP use among women. Both clinic- and community-based communication efforts with a particular focus on the difference between PrEP, STI treatments, and ART are needed
“YOU’VE GOT THE DOOR CRACKED ENOUGH THAT MAYBE THEY CAN’T QUITE GET IT CLOSED AGAIN”: A QUALITATIVE STUDY OF TRAINERS’ PERCEPTIONS OF THE IMPACTS OF A TWO-DAY ANTIRACISM TRAINING
Get PDFObjective Diversity training is commonly used by organizations, but evidence of positive impacts is mixed. Antiracism training, which focuses on structural racism, is largely unexamined in the diversity training literature. The Racial Equity Institute (REI) offers a widely implemented antiracism training called Phase 1 which has never been formally evaluated. As a part of larger community-based participatory research partnership, the research team interviewed REI trainers to understand their perceptions of the impacts of the REI training. Design We conducted qualitative in-depth interviews (n=15) with 58% of REI trainers who lead Phase 1 to understand their views on what participants gain from the workshop. We used thematic qualitative analysis to understand the changes in knowledge, attitudes, and behaviors at the individual and organizational level that trainers hope to achieve, including the factors that influence the impact. Results The interviews revealed that REI trainers expected the training to change participants’ knowledge of and attitudes toward structural racism. Trainers anticipated specific but limited individual and collective behaviors to result from the training. The most important anticipated outcome was that participants learn to connect racial disparities to structural root causes. Conclusions While the literature on diversity training suggests many possible individual, organization, and outcome-level impacts, REI trainers shared a more limited and consistent set of benefits for how individuals and organizations conceptualize and approach inequities. Investigating the perceptions of antiracism trainers about impacts is the first step in creating appropriate criteria for evaluating REI Phase 1 and building an evidence-base for antiracism training
Preferences and willingness to use pre-exposure prophylaxis for HIV among men who have sex with men in mainland China and Hong Kong
Get PDFBackground Pre-exposure prophylaxis (PrEP) uptake remains low in mainland China and Hong Kong. We examined preferences for different PrEP modalities among men who have sex with men (MSM) in mainland China and Hong Kong. Methods We conducted a cross-sectional online survey from May to November 2022 in mainland China and Hong Kong. Eligible participants were aged ≥18years, identified as MSM and self-reported HIV-negative, or unknown HIV status. Random forest models and SHapley Additive exPlanations analyses were used to identify key factors influencing preferences for and willingness to use six PrEP options: (1) daily oral, (2) on-demand oral, (3) monthly oral, (4) two-monthly injectable, (5) six-monthly injectable, and (6) implantable PrEP. Results Among 2142 participants (mainland China: 1604; Hong Kong: 538), the mean age was 28.4 (±7.0) years in mainland China and 34.7 (±9.5) years in Hong Kong. Current PrEP use was similar between mainland China and Hong Kong (18.0% vs 17.8%, P =0.93), with an additional 10.5% and 8.0% reporting past PrEP use (P =0.11), respectively. A greater proportion of participants from mainland China preferred on-demand PrEP compared to those from Hong Kong (55.7% vs 48.1%, P P =0.02). Willingness to use non-oral options was lower, with two-monthly injectable PrEP preferred by 21.1% (19.1-23.1%) in mainland China and 15.4% (12.3-18.5%) in Hong Kong (P Conclusions On-demand and monthly PrEP options remain the preferred choices, though the monthly oral option is neither proven nor available. However, the factors influencing these preferences vary, highlighting the need for tailored and targeted approaches to PrEP implementation
A Qualitative Approach to Understanding the Acceptability and Feasibility of the Intervention STEP After Endometrial Cancer
Get PDFEndometrial cancer survivors (ECS) are at elevated risk of cardiovascular disease, even when compared with other cancer populations. ECS also exhibit high levels of sedentary behavior in daily life, which may be one contributing factor to their increased cardiovascular morbidity. The intervention STEP After Endometrial Cancer was developed to lessen sedentary behavior in ECS, and there is a need to determine if this intervention is likely to work in a ‘real-world’ setting prior to larger-scale research trials. Therefore, this project used 1:1 semi-structured interviews with ECS to evaluate the acceptability and practicality of the STEP After Endometrial Cancer intervention. Interviews focused on specific intervention components, and overarching themes were determined and analyzed. The themes reported were (i) ECS had an increased awareness of their daily activity with a smartwatch, (ii) they valued the accountability provided by a health coach, (iii) they were willing to interact with other people’s posts on a fitness application, (iv) they were hesitant to share and post their own progress and goals, and (v) they are subject to personal health fluctuations that could impact engagement. Overall, these themes demonstrated that the intervention was perceived as acceptable and feasible, supporting its potential for broader implementation with minor adjustments. Exploring acceptability and practicality provided insight into the feasibility of this intervention on a larger scale while also permitting refinement of the intervention to strengthen any future implementation attempts.Bachelor of Scienc
MST-DCBS: A Multi-Scale Transformer and Differentiable Clustering-Based System for Neural Signal Processing and Analysis
Get PDFNeural spike sorting is a critical process in neuroscience, enabling the differentiation of action potentials originating from distinct neurons. [1] Here, we present a multi-scale transformer and differentiable clustering-based system (MST-DCBS) designed to perform robust spike sorting while modeling both temporal and spatial dependencies in electrophysiological signals. Our method includes: (1) a frequency convolution layer to capture sub-band features across hundreds of channels, (2) stacked multi-scale transformer blocks that leverage both self-attention over time and graph attention across channels, (3) a soft k-means module for differentiable clustering, and (4) a self-supervised reconstruction head. Furthermore, an optional dynamic domain adaptation decoder (DDAD) can be employed for motion-related or cross-session tasks.
We evaluate MST-DCBS on the international brain laboratory (IBL) open-source dataset, which involves Neuropixels benchmark recordings. Experimental findings indicate that MST-DCBS achieves high spike classification accuracy and F1 scores, with competitive inference times relative to mainstream deep learning baselines (e.g., Deep Belief Network, Convolutional Neural Network, Recurrent Neural Network). The integration of self-supervised reconstruction drives robust feature learning, while multi-scale attention effectively addresses complex neural activity patterns. Our results underscore the adaptability and potential of Transformer-based architectures for spike sorting, paving the way for scalable analysis and real-time neural decoding in next-generation electrophysiological research.Bachelor of Scienc
Cultivating Change: Reimagining Campus Landscapes as Integrative Spaces for Food Security and Resilience at UNC-Chapel Hill
Get PDFAs UNC-Chapel Hill strives to embody its mission as the "university of the people," its physical landscape remains rooted in outdated, exclusionary landscaping traditions. This thesis explores how reimagining campus grounds as integrative landscapes—spaces that blend ornamental beauty, native biodiversity, and food production—can directly address food insecurity, climate resilience, and student well-being. Through a case study of North Campus residence hall beds, an inventory and analysis of existing plants were conducted, identifying invasive species and assessing opportunities for edible, low-maintenance replacements. Using culturally relevant plant recommendations, this project proposes a practical framework for transforming non-functional spaces into vibrant, multifunctional environments that nourish both community and ecosystem health. By aligning the campus landscape with the university’s evolving public mission, this work argues that ecological stewardship and social resilience are not ancillary goals, but essential components of a truly inclusive university. By reclaiming the landscape as a site of nourishment, resilience, and inclusion, UNC-Chapel Hill can move beyond symbolic commitments and model transformative change for institutions nationwide.Bachelor of Scienc