
Abstracts
Abstracts
Abstract 1: How Do AI Agents Refactor? An Empirical Study of AI-Generated Refactoring Pull Requests and Human Reviews
Presenter(s): Aymen Masmoudi
Institution: University of Michigan-Flint
Abstract—Refactoring plays a central role in maintaining and evolving software systems, yet little is known about how automated code-generation agents document refactoring changes in pull requests and how these agent-generated refactorings are reviewed and evaluated by human developers. To address this gap, we examine refactoring pull requests generated by five AI coding agents, focusing on both the intentions conveyed in the pull request descriptions and the review responses they receive. So, first we extract refactoring descriptors (lexical terms describing any refactoring activity) to identify similar transformation patterns among AI agents then we classify the pull requests into three categories: internal quality attributes, external quality attributes and code smells. Second, we use the Refactoring Review to qualitatively analyze human-reviewed refactoring PRs to better understand reviewers’nd reviewer concerns. Human reviewers typically expressed concerns about code quality, behavioral correctness, and adequacy of the tests. By integrating PR-level refactoring trends with review-level evaluation patterns, this study illustrates how AI-generated refactorings are phrased, how developers interpret them, and why they are not integrated. Our findings highlight the limitations of AI-generated refactoring and underscore the importance of testing and behavioral clarity in collaborative development. Index Terms—Refactoring, Code Review, Large Language Models, Autonomous Coding Agents, Pull Requests, Software Quality
Faculty Advisor(s): Mohamed Wiem Mkaouer, University of Michigan-Flint
Abstract 2: Enhancing GitHub Actions Failure Explanations: Log Preprocessing and Prompt Optimization with LLMs
Presenter(s): Belhassen Khefacha
Institution: University of Michigan-Flint
Large Language Models (LLMs) have become a promising tool for automating CI/CD debugging. Specifically, LLM-based failure diagnosis in GitHub Actions (GA) workflows often takes longer than expected. This study examines how well different LLMs perform at diagnosing GA workflow failures. We start by introducing a novel approach for log processing that looks for error tag filters and structures raw log data, improving the quality of input for LLM. Additionally, we conduct a detailed evaluation of several enhanced prompt engineering strategies, including zero-shot, few-shot, and self-refinement, to identify the most useful methods for generating accurate and useful failure log descriptions. Our experiments with 100 GitHub Actions workflow logs show our approach works, showing significant improvements in the correctness, relevance, and clarity of failure diagnoses.
Faculty Advisor(s): Mohamed Wiem Mkaouer, University of Michigan-Flint
Abstract 3: What does a growing supermassive black hole look like? Part I
Presenter(s): Daniel Abou Khzam
Institution: University of Michigan-Flint
Supermassive black holes, with masses from millions to billions of times the Sun, are found at the centers of most galaxies. Understanding how these black holes grow over time is an important problem in modern astrophysics because their evolution is closely connected to the stellar and gaseous content of galaxies. The goal of our research is to better understand the physical processes that allow supermassive black holes to gain mass from their surrounding environment. Our work focuses on modeling the physical structures that exist around actively growing black holes. The central structure is the accretion disk, a rotating disk of gas and dust that slowly spirals inward toward the black hole. As material in the disk falls inward, it heats up and releases large amounts of energy. Studying the structure and behavior of this disk helps us understand how matter is transported toward the black hole. Another important structural component is the disk wind, an outflow of gas that is launched from the surface of the accretion disk. These winds can remove mass and angular momentum from the system and influence how efficiently the black hole grows. To study this process, we use computational fluid dynamics (CFD) simulations to model how gas flows and accelerates away from the disk under the influence of radiation and gravity. In addition, this research considers the role of a corona, a hot region of energetic plasma located above the accretion disk. The corona produces high-energy radiation that interacts with both the disk and the surrounding gas. By modeling the accretion disk, disk winds, and corona together, our research aims to provide a clearer picture of the physical processes that control the growth of supermassive black holes and the environments in which they exist.
Faculty Advisor(s): Rajib Ganguly, University of Michigan-Flint
Abstract 4: Corporate Interlocks in the Healthcare Ecosystem : A Preliminary Analysis
Presenter(s): Dante Jackson
Institution: University of Michigan-Flint
The modern-day healthcare system is shaped by cross-sector interconnectivity. Corporations are tied together through their governance structures by means of interlocking directorates, which occur when the same individual sits on two or more corporate boards simultaneously. The preliminary results of this project map the incidence of corporate interlocking in the healthcare ecosystem for the year 2023. These classifications include corporations listed as providing or producing biological products, except diagnostic substances, electromedical and electrotherapeutic apparatus, home health care services, and laboratory analytical instruments to name a few. The initial results of the project create a relational data map of these corporations, linked through their shared board members.
Faculty Advisor(s): Roy Barnes, University of Michigan-Flint
Abstract 5: NEO Detroit Revealed | 20 Years Later
Presenter(s): Dichondra Johnson
Institution: University of Michigan-Flint
This year marks the twentieth anniversary of Marketing Neo-Detroit, a longitudinal research project first developed in my 2006 capstone predicting that Detroit’s comeback would be led by Black cultural innovation, grassroots leadership, and community-defined revitalization. Although I presented a visual representation of initial findings last year, this is the first time since 2006 that I am formally delivering the full presentation of the work’s two-decade evolution. Using updated demographic data, archival research, community narratives, and field observations, this project analyzes what has—and has not—come to pass since the original blueprint was written. The findings reveal a complex reality: Detroit’s resurgence is visible, yet inequitable. Black population loss, stagnant household income, civic disengagement among youth, and developments disconnected from longtime residents contrast sharply with the cultural innovation and community models that were predicted to lead the comeback. This 20year reflection challenges dominant revitalization narratives by asking who benefits, who is excluded, and what an authentic, community-centered comeback should look like. The presentation concludes with data-driven recommendations and highlights community-led models demonstrating that equitable redevelopment is both possible and necessary.
Faculty Advisor(s): Maria De Villa, University of Michigan-Flint
Abstract 6: Testing Ceramic Glazes
Presenter(s): Elizabeth Huggler
Institution: University of Michigan-Flint
This research presentation investigates the science behind the art of glazing. Research methods explored how firing temperature and materials influenced the surface, color, and texture of ceramic glazes. Ceramic glazes are chemical mixtures composed of natural earth materials that react and transform into a glass-like state during the firing process, producing a wide range of visual and material outcomes. Temperature and ingredients are two of the most significant variables in this transformation, affecting how glaze materials melt, flow, and interact with the clay body. Through a series of systematic glaze tests, this project examines how glazes respond when mixed with different materials. Test tiles were created using consistent clay material and firing temperatures in order to isolate glaze ingredients as the primary variable. The tiles were then fired at multiple temperature ranges to observe variations in surface quality, color development, opacity, and glaze movement. Research at this point is still ongoing for further explanation of the abstract and our results.
Faculty Advisor(s): Cristen Velliky, University of Michigan-Flint
Abstract 7: OMG, Where Did You Get Those Genes?: An intersectional analysis of the Sydney Sweeny American Eagle ad campaign and its implications for woke culture
Presenter(s): Emmaya Mendez
Institution: University of Michigan-Flint
In July 2025, American Eagle released an ad campaign featuring Sydney Sweeney, a blue-eyed, blonde-haired, White American woman, where she recited a script using the wordplay of “genes” vs “jeans.” American Eagle and Sweeney received significant backlash on social media for the ads’ undertones of misogyny and eugenics. However, official media coverage intentionally reduced the controversy to internet discourse, minimizing fault for the company and Sweeney. While often dismissed, this sort of coverage underpins current antiwoke political campaigns which threaten diversity and equality within the US. This framing also raises questions about how and why celebrities like Sweeney are able to escape accountability for their harmful actions. I conducted a content analysis of official media coverage to identify overarching trends in narrative construction. Three key narrative strategies arose in my research. Some outlets focused on the incriminating political context in which the ad was released, stating it was a bad look for the company. Others suggested the ad may have been an intentional strategy to draw attention with controversy. While some minimized the event to merely online drama. These strategies diverted attention away from the harmful effect of the company’s actions as well as Sweeney’s involvement. I used an intersectional lens to analyze how Sweeney’s race and gender are crucial in explaining why her involvement was framed this way. Companies and celebrities are often allowed to bounce back from controversy despite the systems of oppression they have reinforced through their actions. In Sweeney’s case, her gender allowed her to escape taking full responsibility for her actions despite how they directly contributed to the normalization of White supremacy. These actions come as a result of long standing structures of inequality. If left unchecked, they further reinforce these structures, embolden current political movements, and reinforce a hierarchical system.
Faculty Advisor(s): Heather Laube, University of Michigan-Flint
Abstract 8: Political Violence and Intimate Partner Sexual Violence in Low-and-MiddleIncome Countries: Regional Comparisons
Presenter(s): Hailey Owens
Institution: University of Michigan-Flint
Low-and-middle-income countries (LMICs) face disproportionately high rates of terrorism and political violence. The rate of sexual violence- perpetrated by both combatant groups and civilians alike- has been shown to increase during wartime, and studies have been conducted to explore the dynamics involved with this type of violence. However, in-depth research examining a possible relationship between intimate partner sexual violence and other types of political violence, including acts of terrorism, is not abundant. The current study aims to explore a possible association between country-level terrorism rates and increased rates of sexual violence between intimate partners in LMICs across various regions. Data from UNICEF’s Multiple Indicator Cluster surveys via IPUMS-MICS related to intimate partner sexual violence was utilized, and data related to terrorism was sourced from Our World in Data. This research project builds off of prior research conducted on low-and-middle-income countries in SubSaharan Africa and expands the analysis to all available regions within IPUMS-MICS.
Faculty Advisor(s): Julie Ma, University of Michigan-Flint
Abstract 9: Engine Order Components Derived from Inertial Forces
Presenter(s): Jeong Hun Han
Institution: Kettering University
This study analytically derives the inertial force based engine order components generated by the slider crank mechanism in a reciprocating internal combustion engine. While the primary combustion excitation is relatively intuitive, I found the mathematical origin of higher order harmonic components—such as the 4th and 6th orders in a four cylinder engine— less straightforward. Because these components can coincide with structural resonances and lead to significant NVH issues, I sought to clarify their generation mechanism through a detailed analytical derivation. By formulating piston kinematics using geometric relationships and binomial expansion, the connecting rod angle and resulting inertial forces are expressed as a harmonic series, revealing how even order components naturally arise from the nonlinear geometry of the slider crank mechanism.
Faculty Advisor(s): Javad Baqersad, Kettering University
Abstract 10: The optimization and analysis of graphene reinforced epoxy structures using statics and dynamic testing
Presenter(s): Joshua Lamb, Brody Nichols
Institution: University of Michigan-Flint
This study makes use of experimental methods to determine the damping characteristics of epoxy and graphene nanoplatelet(GNP) nanocomposites. Material fabrication of epoxy-GNP composite into specimens was accomplished by following experimental epoxy fabrication procedures to produce a nanocomposite plate which was then split and routed to ASTM D638 Type 2 standard. This study examines epoxy reinforced by GNP concentrations of 0.2%, 0.5%, 0.8%, and 1.2% produced in this manner. Static testing of the specimens is close to being complete which will allow us to examine how strain rate, and GNP concentration affect the damping capacity and loss factor(η). Dynamic testing will commence soon using free vibration testing to show a direct relationship of η to the GNP concentration. In addition, we will also be conducting Vickers micro indentation hardness tests on the composites to examine the hardness. This study will then examine effects of both GNP concentration and strain rate on the damping capacity and η through static testing, as well as how GNP concentration alone influences η through dynamic testing. The findings of this study should point to an optimal GNP concentration to enhance the mechanical properties of graphene nanoplatelet reinforced epoxy nanocomposites.
Faculty Advisor(s): Olanrewaju Aluko, University of Michigan-Flint
Abstract 11: What does a growing supermassive black hole look like? Part 2
Presenter(s): Josie Kraemer, Daniel Abou Khzam
Institution: University of Michigan-Flint
Supermassive black holes grow by accreting gas and dust from their surrounding environments, forming an accretion disk. As a portion of this material is driven outward as disk winds, they produce broad absorption lines in spectroscopic observations, revealing highvelocity gas pushed from the accretion disk. However, observations also show narrower absorption features that originate from clouds of gas around the black hole. These clouds play an important role in shaping the observed spectra and provide valuable insight into the structure of gas surrounding growing supermassive black holes. Narrow, or associated, absorption lines trace gas that moves with a smaller spread in velocity, allowing us to study the physical properties of individual gas clouds, including their density, composition, and distribution around the black hole. This work focuses on modeling the physical structure of these gas clouds and understanding how they produce the absorption lines observed in the spectra. To model this, we analyze simulated cloud data to develop models that describe the internal structure of the gas clouds and investigate how changes in physical parameters influence that structure. By examining how different profiles influence the resulting absorption lines, this work aims to connect the physical structure of the clouds with the absorption lines they produce.
Faculty Advisor(s): Rajib Ganguly, University of Michigan-Flint
Abstract 12: The Creation of a True Monster: A Comparison of Mary Shelley’s Frankenstein and Guillermo del Toro’s Adaptation
Presenter(s): Madisyn Philo
Institution: University of Michigan-Flint
The presentation will be over my paper completed via an honors extracurricular course. The paper is about comparing and contrasting the original Frankenstein novel by Mary Shelley and the 2025 Guillermo del Toro film adaptation of it. The paper is comparing the differences made by the film, their impact on the story, and the overall main theme of generations trauma the film implements. Along with discussing the religious themes of both medias and how they differ from one another.
Faculty Advisor(s): Mary Jo Kietzman, University of Michigan-Flint
Abstract 13: How immigrants experience and perceive Genesee County
Presenter(s): Maria Alasbahi, Julie McNamara, Hannah Fuentes, Rifat Ananta
Institution: University of Michigan-Flint
The overall population of Michigan is declining, yet immigrants are a demographic that is increasing, including in Genesee County. But who is coming to Flint, and what are they experiencing here? In collaboration with the International Center of Greater Flint and their Global Flint Initiative, we set out to answer these questions using the survey and focus group data we collected from immigrants and international students from 2024 to 2026. Together, we’ll discuss the different ways newcomers contribute to the vibrancy of this community and the economic value and impact they bring. We’ll share examples of lived experiences, challenges, opportunities, and ideas for building a more inclusive community.
Faculty Advisor(s): Emily Feuerherm, University of Michigan-Flint
Abstract 14: Combinations of Physical Punishment and Verbal Reasoning and Their Associations with Children’s Socioemotional Difficulties: A Literature Review
Presenter(s): Monica Hnatiuk
Institution: University of Michigan-Flint
Physical punishment remains widely used in many parts of the world despite growing evidence linking it to negative developmental outcomes for children. At the same time, caregivers often combine physical punishment with non-violent disciplinary strategies such as verbal reasoning. Much of the existing research examines these disciplinary approaches independently rather than exploring how they operate together within caregiving practices. This literature review synthesizes existing research on caregivers’ use of physical punishment and verbal reasoning and their associations with children’s socioemotional outcomes, including aggression, distractibility, and peer relationship difficulties. By examining how these disciplinary strategies are discussed across studies, the review highlights the ways in which mixed discipline practices may influence child development. The review also identifies gaps in current research, particularly the limited attention given to how combinations of physical and non-physical discipline strategies interact within caregiving environments. Understanding these patterns may help inform parenting guidance, child welfare practice, and future empirical research examining the complexity of caregiver discipline practices across different social and cultural contexts.
Faculty Advisor(s): Julie Ma, University of Michigan-Flint
Abstract 15: (VIRTUAL) A Quality Improvement Project for Improving Colorectal Cancer Screening Compliance Amongst Primary Care Providers
Presenter(s): Neha Bakshi
Institution: University of Michigan-Flint
Colorectal cancer (CRC) is a leading cause of cancer-related morbidity and mortality in the United States, with increasing incidence among adults younger than 50. Despite guidelines recommending screening beginning at age 45, adherence among adults aged 45–49 remains suboptimal. A primary care clinic identified low CRC screening compliance among patients aged 45–49 due to inconsistent provider adherence to updated guidelines. A quasi-experimental project evaluated a provider-focused educational intervention using pre- and post-intervention electronic medical record review. A 15-minute educational session on CRC screening guidelines was delivered to primary care providers, followed by a 3-month implementation period. Fiftyfive patients were included (25 pre-intervention, 30 post-intervention). Mean age did not differ between groups (p = .281). Screening acceptance remained 100% in both periods (p = .242). A brief provider-focused educational intervention did not support consistent, guideline-adherent CRC screening practices in a primary care setting. However, if consistency from the provider is taken to adhere to CRC guidelines then there is potential for CRC to be caught early and give patients a positive prognosis.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 16: Disability Type and Discipline: Assessing Disparities Among Children in Vietnam Using Multiple Indicator Cluster Surveys
Presenter(s): Shannon Krueger
Institution: University of Michigan-Flint
Children with disabilities face a higher risk of harsh discipline. This study examines whether cognitive and behavioral disabilities—such as difficulties with learning, concentration, or emotional regulation—are linked to greater exposure to severe physical punishment compared to physical or sensory disabilities, including problems with walking, seeing, or hearing. Using data from Vietnam’s Multiple Indicator Cluster Surveys (MICS), the analysis focuses on children aged 5 to 14 and examines whether disability type predicts exposure to severe physical punishment, while controlling for child demographics and household characteristics.
Faculty Advisor(s): Julie Ma, University of Michigan-Flint
Abstract 17: A Wife, Witch, Mother, and Maid
Presenter(s): Zoe Mayo
Institution: University of Michigan-Flint
My presentation, “A Wife, Witch, Mother, and Maid”, discusses how the novel “Wuthering Heights”, by Emily Brontë, showcases the power of women given to them by their traditional, though usually weak, roles in society. It goes over how the novel uses literary devices, such as symbolism, mood, and foreshadowing, to illustrate this. It focuses on topics such as love, knowledge, society, control, the supernatural, and how these things affect the story. More specifically, how they affect the power women in this story hold by allowing them to have, influence, or use such aspects.
Faculty Advisor(s): Mary Jo Kietzman, University of Michigan-Flint
Abstract 18: Improving Compliance with GLP-1 RA Guidelines in an Endoscopy Setting
Presenter(s): Andrew Krutsch
Institution: University of Michigan-Flint
Background: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) are increasingly prescribed for diabetes and weight management. While clinically effective, these agents delay gastric emptying and are associated with an increased risk of retained gastric contents (RGC), raising concern for regurgitation and aspiration during procedures requiring deep sedation or general anesthesia. In response to emerging evidence, the American Society of Anesthesiologists (ASA) released consensus-based recommendations in 2023 advising temporary discontinuation of GLP-1 RAs before elective procedures. Despite these guidelines, real-world adherence remains inconsistent, contributing to procedure delays and cancellations. Purpose: The purpose of this quality improvement project is to evaluate whether targeted education of scheduling staff and patients improves compliance with ASA-recommended GLP-1 RA withholding guidelines before outpatient endoscopy procedures. Methods: A quasi-experimental pre–post intervention design is being implemented at a local outpatient endoscopy center in North Texas. Adult patients prescribed GLP-1 RAs and scheduled for endoscopic procedures are included. The intervention consists of (1) structured education for schedulers, nurses, and medical assistants regarding GLP-1 RA–associated aspiration risk and ASA guidelines; (2) standardized patient education at the time of scheduling; and (3) automated reminder messages delivered via the electronic health record patient portal. Baseline data from the three months preceding the intervention will be compared with post-intervention data collected over three months. Outcomes: Primary outcomes include rates of procedural cancellation or rescheduling due to nonadherence with GLP-1 RA withholding recommendations. Secondary outcomes include patient compliance rates with guideline-directed medication withholding. Implications: This project aims to improve patient safety, reduce avoidable procedure delays, and enhance operational efficiency through standardized medication screening and education. Results are pending and will inform future practice recommendations for managing GLP-1 RA therapy in procedural settings.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 19: QI Project: Assessing the Impact of Shared Decision-Making in Colorectal Cancer Screening
Presenter(s): Nehali Shah
Institution: University of Michigan-Flint
ABSTRACT According to the National Health Interview Survey, the prevalence of up-to-date colorectal cancer (CRC) screening among adults aged 45 years and older was 59% in 2021 (American Cancer Society, 2023), highlighting a significant opportunity for improvement in primary care. Colorectal cancer screening modalities include stool-based testing ordered by primary care providers and referrals to gastroenterology for colonoscopy, flexible sigmoidoscopy, or CT colonography. In recent years, CRC incidence trends have shifted toward younger populations; in 2019, approximately 20% of CRC cases occurred in individuals aged 54 years or younger, compared to 11% in 1995 (American Cancer Society, 2023), representing a near doubling over 25 years. In response to this increase, national guidelines were updated to recommend initiation of CRC screening at age 45 rather than 50 years. (U.S. Preventive Services Task Force, 2021). This quality improvement project assessed the impact of shared decision-making on colorectal cancer screening rates at a multispecialty private practice in New Jersey. The project evaluated whether the implementation of written educational materials, in addition to standard verbal education, increased rates of CRC screening referrals. The educational handout supported shared decision-making by outlining screening options as well as the required preparation and procedure. Findings from this project may inform evidence-based strategies to enhance patient engagement and improve CRC screening referrals in the primary care setting. REFERENCES American Cancer Society. (2023). Colorectal Cancer Facts & Figures 2023-2025. https://www.cancer.org/content/dam/cancer-org/research/cancer-facts-andstatistics/colorectal-cancer-facts-and-figures/colorectal-cancer-facts-and-figures-2023.pdf U.S. Preventive Services Task Force. (2021, May 18). Colorectal Cancer: Screening. https://www.uspreventiveservicestaskforce.org/uspstf/recommendation/colorectal-cancerscreening#fullrecommendationstart
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 20: Losing Sight of Weight Loss Counseling: Behavioral Interventions
Presenter(s): Adrian Hair
Institution: University of Michigan-Flint
Background Obesity is an epidemic associated with increased chronic disease risk, high healthcare costs, and persistent demands for pharmacological interventions (Pati et al., 2023). Behavioral interventions, recommended by the United States Preventive Services Task Force (USPSTF), are underutilized in primary care. According to the World Health Organization ([WHO] 2022) obesity is defined as a BMI that is greater than or equal to 30. The WHO reports 43% of adults worldwide are obese, and 42.87% of adults in the United States are obese. Purpose The aim of this project was to provide a feasible behavioral treatment approach to obesity for primary care providers to follow based on the USPSTF guidelines (Curry et al., 2018). Providers need to integrate a treatment plan centered around behavioral interventions for successful treatment of obesity (Madigan et al., 2022). Utilizing referrals to dieticians, cognitive behavioral therapists, and exercise physiologists will help prevent provider burnout. Methods This quasi-experimental study included nurse practitioners and physicians in a rural outpatient clinic. An educational intervention introduced evidence-based behavioral care strategies and documentation tools, including standardized ‘dot phrases’ in the electronic health record. Chart reviews pre- and post-intervention assessed changes in provider documentation and patient acceptance of referrals and educational materials. Results Among 63 adult patients (29 pre-, 34 post-intervention), provider documentation of behavioral interventions increased from 64.7% to 100% post-intervention (p < 0.001). Patient acceptance of any behavioral intervention increased from 3.4% to 17.6% (p = 0.074), with nutrition log acceptance rising from 0% to 11.8% (p = 0.056). Other individual intervention acceptance rates, including dietician referrals, also improved, though not statistically significant. Conclusions Provider education and implementation of a structured pre-formulated smart phrase significantly improved evidence-based obesity counseling. Patient engagement showed positive trends. These findings support further implementation of behavioral intervention protocols and structured documentation in primary care settings.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 21: Mental Health Needs of International Students: Insights for Digital Solutions
Presenter(s): Alex Gurung, Chandra Kala Rai, Mahroush Chowdhury
Institution: University of Michigan-Flint
International students experience disproportionately high levels of stress and anxiety due to unique challenges such as acculturative stress, financial pressures, and social isolation. Despite these vulnerabilities, cultural stigma and the perception of mental health struggles as a personal weakness frequently deter them from utilizing traditional counseling services. To address this critical gap, this study employs a mixed-methods approach to understand these specific challenges and co-design culturally sensitive digital mental health support systems. Initially, an online survey of 148 international students revealed that a vast majority experience high academic (93.8%) and financial (93.2%) stress. The survey results also highlighted a strong reliance on self-management and a clear preference for culturally adaptive remote counseling tools. To translate these insights into user-centered digital concepts, we conducted Participatory Design (PD) sessions with nine international students. These sessions demonstrated that international students hold highly diverse mental models regarding their well-being. While some prioritize objective biofeedback and passive data tracking, others require visual metaphors, anonymous community forums, or culturally matched peer support to feel safe and engaged. Ultimately, this research provides an evidence-based framework for universities to develop more inclusive, localized, and culturally attuned digital mental health interventions.
Faculty Advisor(s): Charlotte Tang, University of Michigan-Flint
Abstract 22: Regulating Surface Chemistry for Programmable Nanoparticle Assembly
Presenter(s): Alexis Peters, Abigail Lagueux, Darren Hayden
Institution: University of Michigan-Flint
Perovskite are a class of materials with the composition ABX₃ and found increasing applications in energy conversion and storage technologies, such as solar cells. This research focuses on the synthesis and thermodynamic properties of perovskite materials, starting with bulk samples of cesium lead iodide (CsPbI₃) and cesium tin iodide (CsSnI₃) prepared via solution synthesis. We are currently investigating their thermodynamics, particularly formation enthalpies, using room temperature calorimetry. In future work, we will expand our study to include perovskites with other compositions such as bromides, chlorides, and mixed halides, both in bulk and at the nanoscale. Comprehensive characterization n will be performed using Xray Diffraction (XRD), Differential Scanning Calorimetry (DSC), and Ultraviolet-Visible Spectroscopy (UV-Vis). These measurements will provide insight into composition- and sizedependent perovskite stability and inform the development of perovskite-based devices with enhanced lifetimes.
Faculty Advisor(s): Bin Wang, University of Michigan-Flint
Abstract 23: impact of social media usage on attention spans
Presenter(s): Alina Poles
Institution: University of Michigan-Flint
This study explores the intricate relationship between social media usage and cognitive functions, particularly focusing on attention span, working memory, and cognitive control among young adults. Drawing on recent literature and case study research, it identifies a negative correlation between excessive social media use and sustained attention spans. Key findings reveal that frequent exposure to rapid, fragmented content on platforms like TikTok and Snapchat overstimulates cognitive processes, leading to decreased working memory capacity, impaired cognitive control, and challenges in maintaining focus. Additionally, the study examines the emotional implications, including heightened anxiety and diminished self-esteem, which exacerbate cognitive difficulties. Employing a mixed-methods approach combining case studies and open-ended surveys, the research underscores the dual-edged nature of social media, which offers both opportunities for connection and significant risks of cognitive overload. These findings call for moderated, goal-oriented usage patterns and provide actionable insights for educators, policymakers, and technology designers in mitigating the adverse effects of digital consumption.
Faculty Advisor(s): Amy Kirchhoff, Gregory Spencer
Abstract 24: Implementing Urine Drug Screen Prior to Starting esketamine Treatment
Presenter(s): Amaka Nwaiwu, Tatiyana Rich
Institution: University of Michigan-Flint
This quality improvement project aims at evaluating the impact of implementing a mandatory urinary drug screening protocol before esketamine treatment at an outpatient esketamine clinic in a small city in Michigan. Esketamine is a nasal spray derived from ketamine, and it is administered to adult patients with treatment-resistant mental health disorders, including Major Depressive Disorder (MDD), Post-Traumatic Stress Disorder (PTSD), Generalized Anxiety Disorder (GAD), and acute suicidality. Participants in this project include adult patients eligible for esketamine treatment, as well as healthcare providers engaged in prescribing and administering esketamine. Interventions include giving education to all medical providers and personnel whose roles are significant in the overall process of the implementation of a standardized protocol for urine drug screen for safe esketamine administration and providing post-training knowledge tests. The study uses a pre and post implementation design, measuring urine drug screen compliance and staff knowledge improvement. Statistical analysis compares pre and post implementation rates using chi square tests for categorical variables. Expected results include increased compliance with urinary drug screening (target 75%), improved identification of substance-use risks, reduced treatment-related adverse events, and enhanced provider knowledge. Findings will support integrating routine drug screening into esketamine protocols and promoting safer treatment practices. The study will conclude that standardized drug screening improves patient safety, treatment planning, and provider competency; ultimately optimizing esketamine therapy outcomes. Key search terms: esketamine, ketamine, esketamine guidelines, ketamine guidelines, patient teaching, patient education, provider teaching, substance use, clinical guideline, esketamine education, esketamine barriers
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 25: Botulinum Toxin and the Relationship Between Perceived and Experienced Emotion
Presenter(s): Amanda Alexander
Institution: University of Michigan-Flint
The beauty industry is a fast-growing market, valued at $40.6 billion in 2025. It’s no surprise preferences have shifted to covet a perfect appearance due to current social media influences. Previously so-called ‘under the knife’ procedures with dramatic changes and weeks of down time were the standard for cosmetic enhancements. Currently a shift towards nonsurgical methods is driving the aesthetic market. There were 7.8 million Botox procedures elected in 2024. Botox is a highly powerful neurotoxin that temporarily induces muscle denervation by blocking the release of acetylcholine at the neuromuscular junction, as a result, reducing afferent input from the injected area. While Botox has various other therapeutic purposes, the modern evolution of this injectable neurotoxin is to freeze facial muscles which can inhibit the onset of wrinkles. According to the hypotheses of facial feedback recognition (FFH) and embodied emotion, mimicry of expression not only aids in feeling greater emphasized emotions, but also in the recognition of emotions. Patients with treatment resistant depression have received Botox injections in the glabellar frown lines between the eyes and within two months, 9 out of 10 patients reported significantly improved mood. This is evidence that Botox injections in relation to the FFH can affect mood, and emotional processing. This poster will summarize a literature review that was conducted to explore the main areas targeted in Botox application, the effects of those targeted muscles on expression, and the exploration of how physiological and psychological responses can affect individuals’ perception and expression of emotions under interpersonal and intrapersonal social interactions.
Faculty Advisor(s): Nathaniel Miller, University of Michigan-Flint
Abstract 26: A Study of Open-Source Java Software Test Cases on Github.
Presenter(s): Ameen Vathimattom Ashraf
Institution: University of Michigan-Flint
Automated testing is an important part of software development. It helps developers find bugs early and maintain reliable code. In open-source projects, test cases are often used to show the quality of a project, but this can sometimes be misleading. Many tests are too simple, copied from templates, or do not properly verify the real behavior of the software. This research is an extension of the UROP study on open-source Java test cases. After reviewing related research papers, we collected a large dataset of Java repositories and analyzed different types of tests used in these projects. Our current focus is on security-related testing, especially injection-based tests and TLS/SSL related tests, to understand how developers test security features. By analyzing popular projects and studying different testing patterns, this research aims to better understand how security-related tests are implemented in open-source Java projects. The goal is to identify common patterns and eventually develop tools that can help automatically evaluate and improve security-related test cases in open-source software.
Faculty Advisor(s): Jeffrey Yackley, University of Michigan-Flint
Abstract 27: Machine Learning–Based Road Surface Identification Using Integrated TireEmbedded and Vehicle Sensor Data
Presenter(s): Anij Angdembe, Milan Kadari
Institution: University of Michigan-Flint
Road surface conditions significantly influence vehicle safety, braking performance, and overall driving stability. Variations in pavement texture, surface roughness, material composition, and environmental temperature affect the interaction between tires and the road, which directly impacts vehicle dynamics. For modern intelligent vehicles, including those equipped with advanced driver assistance systems (ADAS) and autonomous driving technologies, the ability to identify road conditions accurately and in real time is essential for improving traction control, vehicle stability, and decision-making in transportation systems. Previous research has demonstrated that acoustic and vibration signals generated during tire road interaction contain identifiable patterns that can be analyzed using machine learning methods to classify road surfaces. Earlier work developed a multimodal sensing system where microphones and accelerometers mounted on a vehicle captured these signals during operation. Machine learning models were then applied to analyze the data and distinguish between road types such as asphalt, concrete, and unpaved surfaces. Building on this prior work, the present study proposes an enhanced sensing architecture that incorporates tire-embedded temperature and pressure sensors along with external acoustic and vibration measurements. The sensing platform includes microphones positioned near the wheels, a vehicle-mounted accelerometer, and sixteen temperature-sensing channels embedded inside the tire, along with dynamic tire pressure monitoring. Sensor data are transmitted through a receiver unit and integrated into the vehicle communication network using the Controller Area Network (CAN) bus. Preliminary observations suggest that internal tire measurements exhibit distinctive patterns under different road conditions. By combining acoustic, vibration, temperature, and pressure data, this multimodal sensing framework provides a strong foundation for machine learning-based road surface monitoring and supports future intelligent vehicle safety systems.
Faculty Advisor(s): Mihai Burzo, University of Michigan-Flint
Abstract 28: Addressing Inconsistent Laboratory Monitoring Psychiatric Medications
Presenter(s): Anthonia Umelogu
Institution: University of Michigan-Flint
Inconsistent laboratory monitoring of psychiatric medications remains a significant patient safety concern in outpatient mental health settings. Medications such as mood stabilizers, antipsychotics, and antidepressants require routine laboratory testing to detect adverse effects, prevent toxicity, and ensure therapeutic effectiveness; however, guideline adherence is often inconsistent due to workflow barriers and limited provider awareness. This project proposes a quasi-experimental, pre- and post-intervention study to evaluate the impact of a targeted educational session on healthcare providers’ adherence to laboratory monitoring guidelines at an outpatient psychiatric clinic. The intervention is based on American Psychiatric Association recommendations and includes provider education, supporting materials, and follow-up reinforcement. Chart reviews will be conducted before and after the intervention to assess changes in monitoring practices, focusing on whether recommended laboratory tests were completed. By improving provider knowledge and adherence to evidence-based monitoring guidelines, this project aims to enhance patient safety, reduce preventable adverse drug events, and promote more consistent, proactive psychiatric medication management.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 29: Utilizing CMS Risk Categories to Determine the Risk of 30-day Readmission or Emergency Room Visits in Newly Discharged Patients to Home Health Care
Presenter(s): Atika Achhava
Institution: University of Michigan-Flint
BACKGROUND Accurate identification of at-risk patients in home health care is essential to utilize resources and prevent rehospitalization or emergency room (ER) visits within 30 days of discharge. Inconsistent use of a properly validated tool to determine risk for rehospitalization at the start of care can lead to inadequate scheduling of nursing visits for the initial 2 weeks by the home care agency. PURPOSE This quality improvement project aims to evaluate whether structured education for registered nurses (RNs) to learn appropriate use of the OASIS M1033 Risk Assessment Questionnaire during the first post-discharge visit at patients’ homes might improve correct patients’ risk identification, appropriate staffing and scheduled nursing visits for the initial 2 weeks of care, with anticipated decreased 30-day readmissions or ER visits. METHODS RNs will receive structured education on accurate completion of M1033 questionnaire and documentation of comorbidities. RNs will be instructed to use the questionnaire during the first home visit to determine the level of risk. The findings will be compared with baseline data reflective of current practice, where M1033 is not consistently implemented, and comorbidities are not evaluated. Inclusion criteria are all patients discharged from the hospital who are not in hospice and/or terminally ill, or a readmission to homecare. This study will take place over three months. OUTCOMES Anticipated outcomes include: increased numbers of correct risk stratification, increased numbers of correctly identified comorbidities, increased initial nursing visits allocated that correlate with the risk category, and reduced 30-day ER visits and readmissions.
Faculty Advisor(s): Roxanne Buterakos, University of Michigan-Flint
Abstract 30: Comprehensive Depression Screening in Adult Patients
Presenter(s): Azin Parsian
Institution: University of Michigan-Flint
At an outpatient internal medicine primary care clinic in mid-Michigan, adult patients are not being routinely screened for depression, or depression screening results are not being relayed to the provider due to a lack systematic management approaches. This is a recognized problem within the practice that has yet to find a solution in increasing screening, and a systematic way of doing so. This has led to a lack of consistent and thorough care where patients may not be screened, and if positive, diagnosed and properly treated. The goal of this research project is to implement systematic and routine depression screenings with results being conveyed to the provider prior to the end of the appointment in efforts to ensure those that screen positive are diagnosed and receive treatment. The clinic utilizes an electronic health record which affords the ability to document and capture screening, diagnosis, and receipt of interventions including education, medication prescription, and referrals. With the use of electronic health records, we anticipate that a practice wide change would improve the documentation associated with depression screening and interventions provided.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 31: Fabricating and Testing Single-Atom Catalyst Cathode for Zn-CO2 Batteries
Presenter(s): Bassam Jehangir
Institution: Kettering University
The transition to sustainable energy systems requires technologies capable of both storing renewable energy and addressing rising atmospheric carbon dioxide levels. Electrochemical energy conversion devices offer efficient pathways for converting chemical energy into electricity, however, many current systems rely on hydrogen fuel, which remains costly to produce, store, and distribute. These challenges motivate the development of alternative electrochemical architectures that integrate carbon utilization with energy storage. Zn-CO2 electrochemical batteries represent a promising approach by coupling zinc-based energy storage with carbon dioxide reduction at the cathode. During discharge, carbon dioxide undergoes electrochemical reduction, while charging requires the same electrode to catalyze the oxygen evolution reaction. Because these reactions involve different mechanisms and kinetic barriers, cathode catalyst design becomes the primary limiting system performance. This work investigates the fabrication and electrochemical evaluation of a single-atom catalyst cathode for Zn-CO2 systems. Single-atom catalyst provides atomically dispersed sites and tunable electronic structures that can enhance catalytic activity and selectivity. Catalyst performance was evaluated using an H-cell electrolysis setup to isolate cathodic reaction behavior. By correlating catalyst structure with electrochemical performance, this research provides insight into the design of bifunctional cathode materials for Zn-CO2 batteries and contributes to emerging strategies that combine energy with carbon dioxide utilization.
Faculty Advisor(s): Yifu Li, Kettering University
Abstract 32: Understanding the Relationship of TRPV4 and ROS production in Osteoarthritis
Presenter(s): Benjamin Bush
Institution: University of Michigan-Flint
INTRO Osteoarthritis (OA) is the most prevalent degenerative disease of the cartilage, with its largest risk being aging. Chondrocytes maintain the integrity of cartilage, via sensing the cartilage using mechanosensitive channels. However, as these cells age they can lose the ability to maintain cartilage due to build up of metabolic waste, specifically reactive oxygen species (ROS). ROS is required for cellular functions, however, if left unchecked ROS can progress aging by inducing oxidative stress (OS). This pushes chondrocytes into a senescent phenotype. This ultimately leads to chondrocyte dysfunction, and therefore OA. OS may dysregulate the mechanosensing which leads to the breakdown of cartilage. A key mechanosensitive channel in chondrocytes is the Transient Receptor Potential Vanilloid 4 (TRPV4) channel. In different cell types TRPV4 upregulation has been shown to increase ROS production. I hypothesize that TRPV4 gain of function increases ROS levels in chondrocytes, leading to OA. MATERIAL and METHODS To test this hypothesis, I will use the fluorescent ROS dye CellROX to quantify the ROS in ATDC5 cells, when they are treated with different TRPV4 agonists. To activate the TRPV4 channel I will treat with either GSK101, 4⍺PDD, or hypotonic swelling (HS), then I will measure the relative fluorescence intensity comparing to no treatment, and a positive control, menedione. RESULTS AND DISCUSSION TRPV4 activation is shown preliminarily to increase ROS in ATDC5 cells. When the TRPV4 channel was activated using GSK101 I observed a dose dependent increase in ROS at 2 hours of treatment. Activation with 4⍺PDD did not produce the same results. When activating using HS I saw an increase in ROS compared to no treatment. I will repeat these experiments using H2O2 as a positive control to have a controlled ROS concentration in my control group rather than menidione. Further studies elucidating the effects of TRPV4 sensitivity under OS are needed.
Faculty Advisor(s): Randall Duncan, University of Michigan-Flint
Abstract 33: Causal Agency
Presenter(s): Billie Jo Patton
Institution: University of Michigan-Flint
Students in K-12 and college settings are regularly taught that “correlation does not equal causation”. However, people continue to conflate correlation and causation, among both college students (e.g., Seifert et al., 2022) and laypeople (e.g., Adams et al., 2017). There are typically multiple, plausible causal pathways that could explain a given correlation between two variables: [A] causes [B], [B] causes [A] (the reverse cause), and a common third variable [C] that causes both [A] and [B]. However, it can be difficult for students to consider alternative causal pathways for correlations (Seifert et al., 2022). Here we tested whether the causal symmetry of a given correlation affects the likelihood that reverse causal pathways are considered as plausible explanations for the correlation. In asymmetric relationships, one variable is more plausibly the causal variable (e.g., cookie consumption), whereas the other variable is more plausibly the outcome (e.g., weight gain). In symmetric relationships, both variables may be equally perceived as causal variables and outcomes (e.g., depression and obesity). We thus expected that participants would be more likely to endorse reverse causal pathway explanations for symmetric than asymmetric correlations. Using a sample of online adults (N=106), we first pre-tested the causal symmetry of various causal relationships. From this pre-test, we selected 4 asymmetric relationships (i.e., one directional causal relationship was perceived as the dominant relationship) and 4 symmetric relationships (i.e., both causal directions were perceived as equally plausible). We then developed 8 fictitious news articles reporting correlational studies based on these scenarios from the pre-test. 100 undergraduate students read each news article in randomized order. They first selected which causal pathways could plausibly explain the correlation: the original claim made by the journalist, the reverse causal pathway, or a common third variable pathway (more than one option could be selected). Participants were significantly more likely to choose only the original causal pathway as a plausible explanation for asymmetric than symmetric correlations and were significantly more likely to select the reverse causal pathway for symmetric than asymmetric correlations. Thus, students may have more difficulty endorsing reverse cause explanations for asymmetric than symmetric correlations both because the original causal pathway is more salient and the reverse causal pathway is less salient. More broadly, knowing which types of correlation scenarios are more difficult to reason about may have pedagogical implications for teaching the concept of “correlation does not equal causation.”
Faculty Advisor(s): Audrey Michal, University of Michigan-Flint
Abstract 34: Teaching Trauma Informed Patient Centered Care Using Mixed Reality Simulation
Presenter(s): Brendan Nelson
Institution: University of Michigan-Flint
The ability to provide trauma-informed patient-centered care is imperative for those who interact with survivors of sexual assault. However, future sexual assault nurse examiners (SANEs) have limited opportunities to learn or practice trauma-informed patient-centered care (TIPC) interview skills; yet are expected to perform effectively in highly dynamic and stressful situations upon completion of their SANE education program. The purpose of this project was to evaluate the usability and effectiveness of a mixed reality (MR) holographic patient sexual assault (SA) focused simulation-based experience (SBE) designed specifically for SANE nurses to facilitate opportunities for deliberate practice of a trauma-informed patient-centered approach.
Faculty Advisor(s): Carmen Turkleson, University of Michigan-Flint
Abstract 35: Strengthening the Front Line: Implementation of Intimate Partner Violence Screening in Mental Health Settings
Presenter(s): Brittany Coleman
Institution: University of Michigan-Flint
The perpetration of intimate partner violence (IPV) against women represents a critical public health crisis, with approximately one-third of American women experiencing sexual violence, physical violence, or stalking by an intimate partner in their lifetime. This quality improvement project aims to implement standardized IPV screening protocols for reproductive-age women in a private practice mental health setting through comprehensive provider education and systematic screening implementation. Using a quasi-experimental pre/post-intervention design, this study will evaluate the effectiveness of provider education on IPV screening completion rates and acceptance of evidence-based interventions. The project will utilize the validated HARK (Humiliation, Afraid, Rape, Kick) screening tool, incorporating electronic health record integration and clinical decision support alerts. Implementation outcomes will be measured through a retrospective chart review comparing three-month periods before and after the educational intervention. This initiative addresses a critical practice gap in mental healthcare while establishing sustainable protocols for identifying and supporting IPV survivors.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 36: Determining the Minimum Inhibitory Concentration & Mechanism of Action for Analogs of Cannabidiol
Presenter(s): Brooklyn Dean
Institution: University of Michigan-Flint
Microbial bacteria are becoming resistant to the antibiotics used to treat infections caused by these bacteria, causing a global problem. Cannabidiol (CBD) has shown antimicrobial properties against various infectious bacteria, including Streptococcus agalactiae and Streptococcus mutans. The goal of this research is to determine whether any closely related structures of CBD have the same antimicrobial properties as CBD. This is done by performing two main assays: a MIC assay and a membrane permeability assay. CBD inhibits bacteria by penetrating the cellular membrane, and the membrane permeability assay is used to demonstrate this.
Faculty Advisor(s): Emily Lavering, University of Michigan-Flint
Abstract 37: Oxidative Stress Decreases Insulin-Like Growth Factor’s Ability to Properly Arrange Cytoskeleton in Chondrocytes
Presenter(s): Bryauna Strauel
Institution: University of Michigan-Flint
Aging increases the risk of developing osteoarthritis (OA), a deterioration of cartilage that surrounds the articular joint space that results in pain and swelling of the joint. A leading cause to OA is chondrocyte function and degradation of the cell matrix. In addition, OA chondrocytes have a higher level of reactive oxygen species (ROS) leading to significant amounts of oxidative stress in the cell. However, studies have found that biochemical signals such as Insulin-like Growth Factor -1 (IGF-1) may reduce the effects of oxidative stress by enhancing polymerization of the actin cytoskeleton, increasing cell tensile strength. The question that remains is how does oxidative stress affect IGF-1’s ability to properly arrange the cytoskeleton. I hypothesize that oxidative stress will cause IGF-1 to be less effective in arranging cytoskeleton of chondrocyte cells. Using ATDC5 chondrocytes, I will initially replicate previous data from my lab, showing that IGF1 increases actin polymerization in the absence of oxidative stress. Using confocal microscopy, I will quantitate these changes in actin cytoskeleton following treatment with IGF-1. I will also determine total cellular tensile strength (stiffness) using atomic force microscopy. I will then examine changes in the cytoskeleton with IGF-1 following treating with the cells with hydrogen peroxide, a natural ROS. Again, actin polymerization and cell stiffness will be measured. I predict that as we introduce more ROS and oxidative stress to the cell, IGF will become less effective in arranging the cytoskeleton. If this is the case, chondrocyte cells with more ROS will have less cytoskeleton rearrangement causing the cells to be less stiff. These data may provide valuable insight into how OA proceeds in an oxidative stressed state and mechanisms that may prevent OA progression.
Faculty Advisor(s): Randall Duncan, University of Michigan-Flint
Abstract 38: Effect of Provider Guideline Education on Preventing Cardiovascular Disease in Patients Abstract
Presenter(s): Cara Horcha Fuwell, Cassandra Aebersold
Institution: University of Michigan-Flint
Background Atherosclerotic cardiovascular disease (ASCVD) is the primary cause of morbidity and mortality worldwide. Research shows that statin medications significantly lower the risk of ASCVD, resulting in the U.S. Preventive Services Task Force’s (USPSTF) guideline recommendation on Statin use for the Primary Prevention of Cardiovascular Disease in Adults. Objectives The goal of this quality improvement (QI) project was to increase provider awareness of the USPSTF guideline in the prevention of ASCVD. The project was intended to meet the following objectives: • Integrate the USPSTF evidence-based guidelines into clinical practice. • Educate providers on the use of ACC/AHA ASCVD Risk Estimator Plus to facilitate appropriate statin prescribing. • Increase the number of patients at-risk of ASCVD initiated on statin therapy. Methods This QI project utilized a quasi-experimental design at a rural primary care clinic in Michigan, which serves the local community. On October 9, 2025, two graduate students conducted a 20minute educational intervention to teach the providers about the guideline and use of the risk estimator to identify patients who would benefit from statin therapy. Three supporting handouts: the full guideline, instructions on the ACC/AHA Risk Estimator Plus calculator, and a patient pamphlet on statins were provided. Data was collected through medical record abstraction for three months prior to the intervention and six weeks post-intervention. All data was coded, cleaned, entered into Excel, and analyzed by a biostatistician using SPSS. Findings Two providers received the educational intervention. A total of 400 medical records were abstracted: 272 pre-intervention and 128 post-intervention. Pre-intervention, zero patients received ASCVD screening using the ACC/AHA tool and post-intervention four patients had documentation of risk screening with the tool. Two patients were appropriately treated with statin medications and two patients refused statin intervention. Conclusions Educational interventions alone are insufficient for providers to consistently integrate evidencebased guidelines into clinical practice. Additional strategies are necessary to support providers making a consistent practice change. Implications This study demonstrates the need for provider buy-in to the use of a tool prior to an educational intervention. Determining the willingness to change and adopt a new tool for screening is an important first step in the change process. Aligning the changes with system process changes may potentially increase the overall use of the new tool.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 39: Identification of fish eggs through DNA barcoding
Presenter(s): Celia Velasquez, Cassidy Vangilder
Institution: University of Michigan-Flint
We identified several different fish egg species from the Flint river via DNA barcoding. This process involves taking a sample, extracting the DNA, amplifying and sequencing it to provide an accurate and precise way of identifying the fish egg species. We compared these “barcodes” to other sequences in the Barcode of Life Data system (BOLD) to match a known species within this database. With this information, we aim to create a reference library for Michigan fishes in which we utilize reference barcode samples and photos to aid in the identification of eggs from their appearance.
Faculty Advisor(s): Kevin Tang, University of Michigan-Flint
Abstract 40: Phylogenetic relationships of Plesiopidae, Pseudochromidae, and other families of Ovalentaria (Teleostei) based on molecular evidence
Presenter(s): Chase Lambeth
Institution: University of Michigan-Flint
Genetic sequencing is vital for studying the evolutionary relationships among species. Many groups of fishes within Ovalentaria, such as the Plesiopidae and Pseudochromidae families, have limited genetic data available. Expanding the available genetic data will help support future research on the phylogenetic relationships within these groups. This study will extract and sequence DNA from a selected species within Ovalentaria and send these sequences to a genetic database. DNA will be extracted from tissue samples of a selected species. Target gene regions will be amplified using polymerase chain reaction (PCR), and gel electrophoresis will be used to confirm successful amplification. The DNA sequences will then be submitted to the genetic database. By contributing new genetic data, this study will support the efforts to fully sequence these genomes and expand upon the understanding of Ovalentaria and the relationships of the families within it.
Faculty Advisor(s): Kevin Tang, University of Michigan-Flint
Abstract 41: Motivations to Respond Without Prejudice: A Literature Review of Research Using the IMS/EMS Scale
Presenter(s): Christine Tyson
Institution: University of Michigan-Flint
Over the decades, white individuals’ self-reported attitudes toward black individuals have improved, leading to increased interest in how individuals regulate prejudice. In 1998, Plant and Devine developed a scale to measure internal and external motivations to respond without prejudice in interracial interactions. This literature review examines all studies published since 1998 that utilize the Plant and Devine IMS/EMS scale to investigate what motivates positive interracial interactions. Only studies including measures of internal or external motivation were retained. Studies were retrieved through a search of Proquest and Web of Science databases. The review examines effect sizes, internal and external motivation, and whether prejudice was measured implicitly or explicitly. Existing literature suggests that higher internal motivation is associated with more positive interracial interactions and lower levels of prejudice, whereas external motivation shows weaker or less consistent relationships.
Faculty Advisor(s): Jennifer LaCosse, University of Michigan-Flint
Abstract 42: Play Beyond Boundaries
Presenter(s): Christopher Williams
Institution: University of Michigan-Flint
I will be presenting a poster of the project that I have continued on since last year. I will likely have an overview of the project as well as the new areas that PBB is (or plans on) expanding to. While I will mainly be presenting a poster, I do wish to present a working ride-on car powered by a Raspberry Pi that uses facial recognition if I can get a working prototype.
Faculty Advisor(s): Shirl Donaldson, University of Michigan-Flint
Abstract 43: Bloom
Presenter(s): Claire Jess
Institution: University of Michigan-Flint
Occupational therapy capstone presentation
Faculty Advisor(s): Mara Robert, University of Michigan-Flint
Abstract 44: Assessing Primary and secondary Oxidation of Fish Oil via peroxide Value and Conjugate Diene Assays
Presenter(s): Collin Miner, Katelin Fields
Institution: University of Michigan-Flint
Our research primarily focuses on using electrochemistry and the properties of antioxidants to reduce the rate of oxidation in common lipid species such as frying oil. The primary methods we are utilizing are cyclic voltammetry to measure the resistance of the sample and correlate to the state of oxidation, Peroxide value analysis, and Conjugated Diene value analysis which are common analytical techniques used to measure the level of oxidation oil samples by comparing to standard values. Once we are able to accurate measure the level of oxidation, our goal will be to explore methods of reducing the rate of oxidation.
Faculty Advisor(s): Matthew Fhaner, University of Michigan-Flint
Abstract 45: Magnetic Field Effects on Thermal Oxidation of Polyunsaturated Fatty Acids
Presenter(s): Damian Cross
Institution: University of Michigan-Flint
Polyunsaturated fatty acids in fish oil are highly susceptible to thermal oxidation, posing significant challenges to food quality and nutritional integrity. This study investigates whether applied magnetic fields can measurably alter the rate or extent of oxidation. Stripped fish oil (endogenous antioxidants and pro-oxidants removed) will be heated under controlled conditions with and without an applied magnetic field. Key variables include oil volume, oven temperature, exposure duration, sampling frequency, and magnet position. Oxidation will be quantified internally via peroxide value (PV) and conjugated diene (CD) analysis, and externally validated by GC-FID profiling of fatty acid degradation products. Results are pending, though this research will establish whether magnetic fields represent a viable mechanism for modulating lipid oxidation kinetics. Positive findings would have direct implications for food processing and preservation, offering a novel approach to extending the shelf life of polyunsaturated fatty acid-rich oils without chemical additives.
Faculty Advisor(s): Matthew Fhaner, University of Michigan-Flint
Abstract 46: Caregiver Education and Harsh Physical Discipline in Low- and MiddleIncome Countries
Presenter(s): David Suarez
Institution: University of Michigan-Flint
This study examines whether rates of harsh physical discipline differ between households with varying caregiver education levels, even within the same household wealth level in low- and middle-income countries. Harsh discipline includes both severe and moderate forms of physical punishment, such as beating a child, hitting a child on the face, head, or ears, hitting a child with an object, or spanking. Drawing on discipline items from the UNICEF Multiple Indicator Cluster Surveys (MICS), the study aims to explore whether caregiver education level is associated with differences in the use of harsh disciplinary practices while holding household wealth constant. Specifically, the study investigates whether households with lower caregiver education levels report higher rates of harsh discipline compared to households with higher caregiver education levels within similar economic contexts.
Faculty Advisor(s): Julie Ma, University of Michigan-Flint
Abstract 47: Parenting Styles and Their Influence on Adolescent Self-esteem and Violent Actions
Presenter(s): Diane Tupper
Institution: University of Michigan-Flint
This study seeks to investigate the impact of four distinct parenting styles on adolescent self-esteem and rates of violence in middle and low-income countries. This study will also examine if there is any correlation between self-esteem levels and violence. The four parenting styles that will be examined are Authoritarian, Authoritative, Permissive, and Neglectful. Utilizing a literature review with multi-level modeling, the study will explore if there is a parenting style that most increases self-esteem and limits violent tendencies in low and middle income countries.
Faculty Advisor(s): Julie Ma, University of Michigan-Flint
Abstract 48: A Quality Improvement Project: Inter-Agency Communication Between Urgent Care and the Emergency Department
Presenter(s): Erika D’Aoust
Institution: University of Michigan-Flint
A Quality Improvement Project: Inter-Agency Communication Between The Urgent Care And The Emergency Department Introduction: Clear communication is essential in medicine and is critical in the transfer of patients from Urgent Care Centers (UCC) to the Emergency Department (ED) and directly relates to patient safety. Objective: The primary objective of this original research was to demonstrate the use of communication tools such as the SBAR report to improve ineffectiveness of current communication between the UCC and the ED. Methods: This translational quality improvement (QI) project used a quasi-experimental design method comparing data both prospectively and retrospectively with pre and post intervention surveys of ED nurses and UCC providers. These surveys assessed the strength of the current transfer method and compared it to the newly implemented SBAR communication tool over a 12 week period. Both ED nurses and UCC providers were provided brief educational sessions highlighting the importance of communication as it relates to patient safety. Results: Despite project limitations, 69.5% of nurses surveyed post-intervention agreed, or strongly agreed, that the SBAR communication tool improved communication, while 88.9% of providers surveyed pre-intervention agreed that the current practice in EMS reporting was not adequate and important information was not relayed. Conclusion: Utilizing a standardized SBAR tool resulted in improved communication between the UCC and the ED and, thereby, improved patient safety in general.
Faculty Advisor(s): Kristi Wilson, University of Michigan-Flint
Abstract 49: Autonomous Driving at 1:10 Scale – Integrated Approach using Perception & V2I within Simulation Environment
Presenter(s): Frank Tilli, Ashwin Krishnakumar
Institution: Kettering University
This work presents a modular, simulation-based framework that integrates perception and vehicle-to-infrastructure (V2I) communication to improve the reliability of autonomous navigation. The system is implemented and validated using the Quanser 1:10 scale QCar2 digital twin within the Interactive Labs (QLabs) environment, which enables the creation of realistic and customizable driving scenarios. The perception module uses a camera-based AI model to detect traffic lights, stop signs, yield signs, pedestrians, and other vehicles. The model is trained using YOLOv8. A large dataset was collected from the QLabs environment and labeled with bounding boxes in Roboflow. The labeled data was then exported and used to train the model with the Ultralytics Python package on a GPU workstation, enabling accurate and efficient object detection within the simulation. The V2I subsystem emulates communication between the vehicle and smart traffic lights using TCP/IP over WiFi. Each traffic light is assigned a unique IP address, and its signal state is accessed wirelessly through the Quanser traffic light library. To ensure context-aware operation, geofencing is implemented around each traffic light. These geofenced zones are rectangular regions centered on the light’s position and scaled to match the QCar’s coordinate system. Geofencing provides spatial awareness by allowing the vehicle to monitor traffic signals only when it is within a defined proximity. Once inside a geofenced area, the vehicle evaluates the signal state and decides whether to stop or proceed. This approach reduces unnecessary polling, prevents premature reactions, and mirrors how real vehicles approach intersections. It also acts as a fallback mechanism if vision-based traffic light detection fails, ensuring safe and reliable behavior without overloading the system. Overall, the framework integrates AI-based perception, infrastructure communication, and vehicle control for trajectory tracking within a unified simulation environment. Designed for portability between virtual and physical platforms, the system supports iterative development, retraining, and performance optimization. Its modular and scalable architecture provides a strong foundation for future enhancements, such as incorporating LiDAR to complement visionbased detection of pedestrians and vehicles.
Faculty Advisor(s): Girma Tewolde, Kettering University
Abstract 50: Teaching LLMs to Analyze Character Sentiment Based on TV Series Discussions
Presenter(s): Gabriela Haralambous, Souha Ben Hassine
Institution: University of Michigan-Flint
Online discussions on Reddit about TV series are not only a popular way for people to express opinions on their favorite shows, but also provide valuable perspectives and sentiments on characters and how they change throughout a series. However, the large number of posts available for each show makes it difficult to extract the most relevant information, distinguish between characters, and form conclusions about audience perceptions. The objective of this project is to create a database of Reddit users’ attitudes and the reasoning behind those sentiments for use with LLMs. This allows users to ask natural-language questions about characters and receive responses based on the dataset. The data from Reddit was organized hierarchically. Each TV series contained posts from its respective subreddit. For each post, the character discussed was identified, and annotators chose their associated sentiment and reasoning. These annotations were used to create character-level data, which allows LLMs to process questions and retrieve relevant information from the database. This system helps LLMs analyze how and why viewers feel towards characters throughout different TV series.
Faculty Advisor(s): Steve Wilson, University of Michigan-Flint
Abstract 51: Examining Trends in Mammal Communities with Camera Traps in Northern Hardwood Forests, Michigan
Presenter(s): Gavin Stover, Leah Ritter, Dylan Jones
Institution: University of Michigan-Flint
Northern hardwood forests cover more than 2 million ha in Michigan and are managed for ecological and timber values. The mammal community of these forest systems provides important ecological and recreational functions. As part of a long-term study on forest regeneration techniques and wildlife interactions, we quantified site-level (12 ha) mammal occupancy using unbaited camera traps. We stratified study sites (n=48) by 3 regions (northern Lower Peninsula [NLP], eastern Upper Peninsula [EUP], and western Upper Peninsula [WUP]) in Michigan and collected data from 2017-2021. We tagged all photos to species and used standard photo verification processes in conjunction with Ai generated bounding boxes and species suggestions via the Wildlife insights photo categorization platform. We documented a wide functional range of mammals, including small mammals (deer mice, flying squirrels), mesocarnivores (marten, fisher, bobcat, coyote), and larger herbivores (deer) and a carnivore (wolves). We report on mammal occupancy and community structure in managed northern hardwood forests in Michigan. We also report on the amount of sampling needed to represent the suite of detectable species.
Faculty Advisor(s): Melissa Starking, University of Michigan-Flint
Abstract 52: What We Say & What We Sow: Variation in Seed Mix Composition Based on Search Terminology
Presenter(s): Griffin Wagner
Institution: University of Michigan-Flint
The terminology and language used in advertising ecologically friendly lawn alternatives do not give clear cut, accurate descriptions of the plant species involved. To gain a better understanding of plant species in seed mixes associated with different terminology, 6 search terms were used across 4 databases and the top results were gathered and analyzed based on native status & broad morphology. The gathered data indicate a trend of higher species richness and native species makeup for search terms including the words “Native” and “Wildflower”, while search terms containing “Lawn” yielded lower overall species richness, a higher non-native species composition, and an increased abundance of graminoids. These results indicate clear trends in the language used surrounding lawn alternatives impacting the types of species advertised under such language.
Faculty Advisor(s): Rebecca Tonietto, University of Michigan-Flint
Abstract 53: Advancing Smoking Cessation in a Primary Care Clinic Serving an Arab American Population Through Provider Education
Presenter(s): Hassan Tekko
Institution: University of Michigan-Flint
Abstract Background: Tobacco use is a leading cause of preventable disease and remains highly prevalent within Arab American communities. Despite national recommendations, smoking cessation practices in primary care are often inconsistent and limited to brief assessment. Provider education may improve guideline-based smoking cessation practices and patient engagement. Purpose: This quality improvement project evaluated whether a provider education intervention based on U.S. Preventive Services Task Force smoking cessation guidelines improved documentation of smoking status assessment and patient acceptance of smoking cessation interventions in a primary care clinic serving a predominantly Arab American adult population. Methods: A quasi-experimental pre–post design was used. Providers and medical assistants participated in an educational session emphasizing the “Ask, Advise, Refer” framework. Electronic health record chart reviews were conducted before and after the intervention. Results: Fifty patient encounters were reviewed pre-intervention and twenty-three postintervention. Smoking status assessment documentation increased from 58.0% to 91.3%, and patient acceptance of cessation interventions increased from 24.0% to 78.3%. Improvements were statistically significant (p < .001). Conclusions: Provider education significantly improved smoking cessation documentation and patient engagement in primary care.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 54: The Historical Evolution and Importance of The Criminal Justice Major in The University of Michigan-Flint
Presenter(s): Heidi Henriksen
Institution: University of Michigan-Flint
The criminal justice major at the University of Michigan-Flint started out as a minor in 1980 and evolved into a fully fledged major in 1985 according to yearbooks that listed majors and minors at the time. Now that criminal justice is offered as a major, the positives, negatives, and purpose associated within the program need to be observed. Some of these positives include how students are interested in the criminal justice courses offered at U-M Flint and how the courses offered are sufficient enough for students to graduate with a criminal justice degree. Some of the negatives explore the consequences of having only one criminal justice professor and how the criminal justice program is not advertised as much as other programs, possibly due to understaffing. The purpose of the criminal justice program is explained through connections to location, students, issues in society, and the motivation of students. With my research, I explore all of these topics and connect this information to sociology, as this is my main goal. The connection to sociology is significant because of how intertwined sociology is with the criminal justice system and how prevalent sociology is to all the topics covered in my research.
Faculty Advisor(s): John Girdwood, University of Michigan-Flint
Abstract 55: Effects of Pulse Electromagnetic Fields on Osteoblast-Like Cell Proliferation
Presenter(s): James Elford
Institution: University of Michigan-Flint
Skeletal remodeling is a continuous process in bone that is necessary for the maintenance of bone strength and integrity. As we age, the balance between bone formation by osteoblasts and bone resorption by osteoclasts shifts toward resorption. This imbalance leads to decreased bone density and a higher risk of fractures, a pathology known as osteoporosis. Females who have gone through menopause experience more rapid loss of bone due to lower estrogen levels. Males over 70 are also at high risk. Pulsed electromagnetic field (PEMF) therapy is a non-invasive and drug-free technique that is clinically approved for treating nonunions by targeting L-type voltage gated calcium channels. However, clinical trials have shown that these positive effects have been inconsistent. I postulate that delivering low-frequency magnetic fields activates L-type voltage gated channels that increases intracellular calcium signaling. This increase stimulates bone growth and reduces overall healing time. I also hypothesize that the inconsistency of PEMF treatment occurs since PEMF parameters are not optimal. In this study, I will work to optimize the parameters of PEMF, examining differences in magnitude, frequency and duration of the stimulus on anabolic response in bone. To test these hypotheses, I will generate PEMFs using Helmholz coils across MC3T3-E1 osteoblast-like cells to determine changes in proliferation of the cells with differences in magnitude. My lab’s preliminary data has indicated that proliferation is increased when a 2.2 mT field is applied. I will expand the magnitude response to determine if 2.2mT is optimal. Understanding the relationship between L-VGCCs, PEMF magnitude, frequency, and duration can help better understand the effects of PEMF, and potentially lead to improved effects of PEMF therapy and perhaps even toward development of wearable technology to reduce bone loss.
Faculty Advisor(s): Randall Duncan, University of Michigan-Flint
Abstract 56: Hepatitis C
Presenter(s): Jayla Garth, Emily Klintworth
Institution: University of Michigan-Flint
The aim of this quality improvement project was to implement the United States Preventative Task Force (USPSTF) 2020 guideline at an urban primary care practice in Michigan. It states that screening for Hepatitis C should be given to adults aged 18 to 79 years old. According to the USPSTF, hepatitis C is the most common chronic blood-borne pathogen in the US and a leading cause of complications from chronic liver disease. This guideline is in place to address the rise in Hepatitis C and begin steps of evidence-based practices to educate and eliminate further risk. The guideline recommends screening all asymptomatic adults, including pregnant women, without known liver disease. The ultimate goal is that patients that fit criteria are screened and treated early, if necessary. Hepatitis C screening in the early stages gives patients a higher opportunity at avoiding liver failure. Providers should include this screening into routine screening, if patients agree. Outcome data was collected before and after provider education. Two months of data were obtained looking to see if there was an increase in testing being completed as well as a plan of care and treatment for any patients that tested positive for Hepatitis C.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 57: Biomimetic Robot Herd Integration for Wildlife Monitoring
Presenter(s): Jenna Yasin
Institution: University of Michigan-Flint
This project is an interdisciplinary study between wildlife biology and robotics that aims to enhance wildlife monitoring while advancing research in animal-robot interaction. Wildlife data collection is currently largely human-centered, which can limit the amount of data collected and disrupt natural animal behavior. This project proposes the use of a biomimetic robot designed to integrate into the ecosystem alongside white-tailed deer in Michigan in order to overcome these limitations. The robot contains technology that enables the monitoring and collection of behavioral and environmental data with minimal disruption to the ecosystem. Reinforcement learning from human feedback allows wildlife experts to guide and improve the robot’s behavior. This approach seeks to improve methods for studying wildlife while advancing our understanding of animal-robot interaction.
Faculty Advisor(s): Mark Allison, University of Michigan-Flint
Abstract 58: Influence of Infill Density and Raster Angle on the Cyclic Torsional Performance of 3D-Printed ABS
Presenter(s): Jeremy Sarpong, Grayson Lumsden
Institution: University of Michigan-Flint
This study investigates the fatigue behavior and failure mechanisms of 3D-printed ABS subjected to cyclic torsional loads, with a focus on the effects of infill density and raster angle on torsional fatigue performance. A total of 50 test specimens representing 25 unique combinations of infill density (20%, 40%, 60%, 80%, 100%) and raster angle (0°, 25°/-65°, 45°/45°, 75°/-15°, 0°/90°) were fabricated and tested using a dynamic torsion system. Fatigue failure was defined as 50% reduction in torsional strength, recorded through cycle-by-cycle torque monitoring. The twist angle was cyclically varied between ±10° at a frequency of 5 Hz until failure occurred. The results demonstrate that both infill density and raster angle substantially influence fatigue life, with higher infill generally improving performance but showing diminished gains between 60% and 80% density. Raster angle exhibited an even stronger effect, with orientations of 25° and 75° consistently achieving the highest fatigue lives across all densities, while 0° and 90° produced the poorest performance. When normalized by mass, the most efficient set of parameters was 100% infill printed at 25° raster angle. These findings highlight the importance of proper selection of 3D-printing infill parameters such as infill density and raster angle in optimizing 3D-printed ABS parts. This study provides practical guidance for the print design of additively manufactured parts for cyclic bending purposes, and contributes to a better understanding of the material properties of FDM in fatigue applications.
Faculty Advisor(s): Khalil Khanafer, University of Michigan-Flint
Abstract 59: Cognitive Responses to Acute Activation of the Immune System
Presenter(s): John Miner, Brandon Albright
Institution: University of Michigan-Flint
Acute infections, whether viral or bacterial, activate the immune system and result in a constellation of responses. Animal studies show that lipopolysaccharide (LPS) impairs spatial navigation. Humans with upper respiratory tract infections (URTIs) report cognitive deficits, but links between acute infection and spatial navigation in older adults remain unclear. We report on two studies examining the effects of acute infections, such as urinary tract infections (UTIs) or URTIs, on cognition, including spatial navigation, in older adults. We hypothesize that acute infections will negatively affect wayfinding and cognitive functions. One study focuses on the data from the Health and Retirement Study. We examined several codebooks, including the Aging, Demographics, and Memory Study Wave; Biomarker Data; and the 2019 Health Survey, to identify variables for inclusion in the study. Variables related to cognition and acute infections were identified for inclusion in the analysis. In a second study, with a focus on older adults residing in an independent living facility, we conducted a review of the literature to update the study protocol. The study design is repeatedmeasures. Eligibility criteria include adults aged 50 years or older, without visual or hearing deficits, no neurological/psychiatric conditions, or any condition that could impair the use of a computer keyboard or arrow keys. Subjects must be healthy for at least one week before testing. Subjects with URTI symptoms will be tested within 96 hours of symptom appearance. Subjects will be interviewed using PASE (physical activity), cognitive measures (cognitive status), and 3D-CAM (delirium screening). Spatial navigation will be tested using the ComputerGenerated Arena, Morris Water Maze analog. Spatial navigation indices include directional heading error and time-to-platform completion. Data is collected using REDCap, which automates scoring and integrates validated tools. Data collection occurs over 1-3 days. Findings could inform early interventions for infection-related cognitive decline in aging populations.
Faculty Advisor(s): Barbara Kupferschmid, University of Michigan-Flint
Abstract 60: A Survey of the Introductory Cybersecurity Landscape in Undergraduate Higher Education within the U.S.A.
Presenter(s): Johns Hopkins, Aidan Gitschlag
Institution: University of Michigan-Flint
This survey is designed to understand how introductory cybersecurity is taught across the United States of America (USA) at the collegiate, undergraduate level. The U.S. Cybersecurity and Infrastructure Security Agency (CISA) National Initiative for Cybersecurity Education (NICE) Framework which defines cybersecurity competency areas in addition to the knowledge and skills necessary to complete tasks in defined work roles is designed to be used by educators to create an improved curriculum that meets the needs for the modern cybersecurity workforce. However, it is currently poorly understood exactly how introductory cybersecurity is taught across the country at the college-level and which subjects are taught within these courses and how they map to the NICE framework. This research study examines 2,541 colleges and universities across the United States to determine how many teach cybersecurity at the introductory level, what topics are actually covered in introductory courses, and how the content of these courses maps to the NICE Framework.
Faculty Advisor(s): Jeffrey Yackley, University of Michigan-Flint
Abstract 61: Using RandomForest Regression for 24hr Weather Forecasting in Flint, Michigan
Presenter(s): Joshua Tewolde
Institution: Kettering University
The weather forecasting industry is projected to reach $4.1 billion by 2030, but much of the available capital is spent on expensive equipment such as satellites, weather balloons, ground/sea stations, and supercomputers capable of forecasting. Though a large quantity of quality real-time weather data is freely available from state and federal agencies, no widelyadopted method exists for localized weather forecasting. Thus, the research goal is the development of a 24-hour weather forecasting model for Flint, MI, solely based on publicly available Automated Surface Observing Station (ASOS) real-time weather data. Because the westerlies (wind patterns) play the most prominent role in continental US weather patterns (30°N-60°N), ASOS data from 83 total stations was considered, from seven states west of (and including) Michigan. Eleven years of temperature and dew point data were collected from these weather stations, a choice informed by feature importances, and used—in conjunction with various preprocessing and feature engineering techniques—to train a Random Forest Regression (RFR) tree-based model. Additionally, a k-Nearest-Neighbors (kNN) model was used as a baseline comparison. In all, correlation between input weather in the chosen locations and 24-hr weather in Flint was very high, with the coefficient of determination R^2 > 0.93. Nearly 70% of forecasted temperatures lay within 5°F of the correct values, 50% of forecasts within 3°F, and all metrics of the RFR exceed the kNN. The relevance of this project is twofold: first as a proof of concept of accessible weather forecasting based on publicly available data, where local groups can easily create their own quality weather models, and secondly as a concept-validation for the use of RFR and other tree-based models for weather forecasting.
Faculty Advisor(s): Saroja Kanchi, Kettering University
Abstract 62: Assessing the Antimicrobial Properties of Analogs of Cannabidiol against Staphylococcus Aureus and Streptococcus pyogenes
Presenter(s): KayLee Robinson
Institution: University of Michigan-Flint
The steady rise of antibiotic-resistant bacteria demonstrates the need for new antimicrobial compounds. Staphylococcus aureus and Streptococcus pyogenes are Grampositive bacteria responsible for many clinical infections. Cannabidiol (CBD), a non-psychoactive compound derived from Cannabis sativa, has shown promising antimicrobial activity. This study examines analogs of CBD and their effectiveness against Gram-positive bacteria, including S. aureus and S. pyogenes, using minimum inhibitory concentration (MIC) assays. Results indicate that several CBD analogs inhibit the growth of these pathogens.
Faculty Advisor(s): Emily Lavering, University of Michigan-Flint
Abstract 63: Non-White People’s Approach to Interracial Interactions: A Literature Review
Presenter(s): Kaylie Randolph
Institution: University of Michigan-Flint
There is a lot of research on White people’s approach to interracial interactions; however, there is not a lot of research involving people of color and their approach. To discover what research does exist on non-White people’s approach to interracial interactions a literature review was performed. While analyzing the literature six common themes emerged. The first and second themes are expectations / perceived responsiveness. People of color will have expectations of how the interaction is going to go based on cues from the white person. If the white person engages in conversation, smiling, nodding, and having open body language; these cues will help give the notion of perceived responsiveness. Helping to reduce anxiety surrounding stereotypes, prejudice and racism. The third and fourth themes are identity threat / identity safety. Words, non verbal cues, and behaviors can help someone know if the interaction is going to cause identity threat or identity safety. The literature examines if basic cues like smiling and nodding help to give identity safety. More explicit cues that show empathy and openly disapproving of racism increase identity safety. In addition, previous research suggests that if a person of color knows a white individual that has a diverse network of friends, they will assume the interaction will have more identity safety. The fifth and six themes are ingroup and outgroup. The literature examines the impact of being ingroup or outgroup in interracial interactions. People of color experience fear, anxiety and metastereotype concerns when interacting with a white person. Research explores the different social norms within ingroups and how the differences impact interracial interactions.
Faculty Advisor(s): Jennifer LaCosse, University of Michigan-Flint
Abstract 65: Adolescent perceptions of telehealth access and quality: Comparing video visits and digital messaging in Genesee County
Presenter(s): Lily Meyer, Melisa Gerard, Phani Davu
Institution: University of Michigan-Flint
The goal of the study is to describe the perceived benefits of pediatric synchronous and asynchronous telehealth overall and by Medicaid status based on interviews with adolescents. Flyers advertising the study were placed in Genesee County, MI locations including pediatrician offices. Parents/caregivers provided consent and adolescents provided assent to participate. Parents/caregivers completed a socio-demographic survey related to their adolescent. Two researchers interviewed each adolescent virtually. Audio recordings were transcribed verbatim. Themes were developed by two independent coders who read the transcripts. Of the 81 adolescents, 16 (19.8%) were 14-years old, 20 (24.7%) were 15-years old, 23 (28.4%) were 16-year old, and 22 (27.2%) were 17-year old. A total of 45 (55.5%) were nonHispanic or non-Latino White, 20 (24.7%) were non-Hispanic or non-Latino Black or African American, 5 (6.2%) were White Latino, and the remaining were from other racial/ethnic groups. Of 81 participants, 37 (45.7%) had Medicaid coverage in the last 12 months. Adolescents expressed that video visits contributed to a sense of comfort and convenience. Adolescents stated the ability to review health information on an application or via email allowed for better adherence to care plans. Adolescents stated the option to electronically schedule their own appointments contributed to a greater sense of independence. While participants with and without Medicaid noted benefits of telehealth, they expressed differences in priorities and access-related issues. Adolescents on Medicaid viewed telehealth as a way to overcome transportation barriers. Those not on Medicaid perceived it as a means of managing extracurricular activities and healthcare appointments. Understanding the perceived benefits of telehealth could improve elements of youth-focused healthcare such as increased availability of and access to telehealth.
Faculty Advisor(s): Gergana Kodjebacheva, University of Michigan-Flint
Abstract 66: Developing a Mobile-Application for Fall Risk in Older Adults: Validating Balance with Eyes Closed
Presenter(s): Lily Sullivan, Keera Barnhill, Zeriah Barnhill, Matthew Hardesty
Institution: University of Michigan-Flint
While the increase in the number of aging individuals means that advances in healthcare are helping people live longer, it also means that there is likely going to be an increase in age-related health issues that negatively impact older individuals’ quality of life (Jugran, 2025). Several studies show that the risk of falling increases with age (Álvarez, et al., 2023; Jehu et al., 2021; Sun & Sosnoff, 2021). Researchers agree that the increased risk of falls amongst older adults decreases their autonomy (Álvarez, et al., 2023). Additionally, falls amongst older adults have been considered to be one of the leading causes of severe injury or death in people 65 and older (Jehu et al., 2021; Sun & Sosnoff, 2021). One approach to limit falls in older adults is to regularly measure fall risk through technology (Hsieh et al., 2019; Roshdibenam et al, 2021). Our research team has been developing a mobile application to measure fall risk. The current study assessed the validity of our mobile app by comparing the output of our app to a gold-standard measure of balance (Mobility Lab Software). Both our app and the gold-standard software measure balance through small Inertial measurement units (IMUs), that contain accelerometers, gyroscopes, and magnetometers, which are common in smart phones. The current study assessed the validity of our measure, through a Bland-Altman analysis, by comparing measurements while participants maintained balance on both feet, with their eyes closed. The results of these analyses will be discussed. Key Words: Fall Risk, Technology, Balance, Mobility Lab
Faculty Advisor(s): Nathaniel Miller, University of Michigan-Flint
Abstract 67: Optimization of Lattice Structures for Lightweight Component Design Using nTop
Presenter(s): Lorenzo Gaglio
Institution: Kettering University
Additive manufacturing has enabled the practical use of lattice structures for lightweight engineering applications. By replacing solid material with periodic cellular architectures, it is possible to significantly reduce weight while maintaining sufficient mechanical performance. In this study, lattice-based design is applied to optimize a representative engineering component relevant to automotive applications. Using the computational design platform nTop, the interior of a solid reference part is replaced with parameterized lattice structures. Several lattice configurations are explored by varying unit cell type and relative density to investigate their influence on overall weight and structural stiffness. The performance of the lattice-filled designs is evaluated through comparative structural analysis, and the results are compared with the baseline solid model. The study focuses on identifying lattice configurations that provide meaningful weight reduction while preserving acceptable mechanical behavior. In addition, the potential of graded lattice structures, where lattice density varies according to expected load distribution, is briefly explored as an approach to further improve structural efficiency. This work demonstrates a rapid design workflow for lattice optimization using nTop and highlights the potential of cellular structures in lightweight component design for engineering applications.
Faculty Advisor(s): Seyed Jamaleddin Mostafavi Yazdi, Kettering University
Abstract 68: Behind the Seams
Presenter(s): Lydia Sturgeon
Institution: University of Michigan-Flint
My presentation will highlight the world of a costume designer working on threatre productions. I will describe the reality of designing for productions and how this differentiates from the assumptions and romanticized life that films portray.
Faculty Advisor(s): Shelby Newport, University of Michigan-Flint
Abstract 70: Antimicrobial Properties of Cannabidiol and Its Analogs Against Staphylococcus epidermidis
Presenter(s): Maria Alasbahi
Institution: University of Michigan-Flint
The rise of antibiotic resistance poses a major threat to modern medicine. Since the discovery of Penicillin in 1928, antibiotics have been the primary treatment for bacterial infections. However, increasing resistance among disease-causing bacteria is diminishing their effectiveness. Recent studies have demonstrated that Cannabidiol has antimicrobial properties and is effective in countering many bacteria. Hexocannabitriol, recently identified from Cannabis sativa, has shown promising pharmacological properties that may contribute to antibacterial activity. We evaluated the antimicrobial potential of hexocannabitriol and its analogs against Staphylococcus epidermidis, as well as a wide range of other bacteria, using minimum inhibitory concentration assays. We are also investigating the mechanism of action of the antimicrobial activity of the compounds. This work aims to contribute to the development of therapeutic strategies for combating antibiotic-resistant infections.
Faculty Advisor(s): Emily Lavering, University of Michigan-Flint
Abstract 71: Psychology Major Study
Presenter(s): Mars Baryo
Institution: University of Michigan-Flint
There is a significant amount of research on the reasoning behind choosing one’s major. One major theme found across present research includes measuring personality measures and relating them to major choice, but little is found specifically about choosing psychology as a major or how mental health conditions contribute. The objective of the current work and the second study being designed currently is to see what motivates psychology students to choose psychology as a major. We hypothesized that a) If you or someone close to you has experienced mental health diagnoses b) and/or if you or someone close to you has experienced bias for having mental health diagnoses you would be more likely to become a psychology major. We created a survey measuring the Big 5 personality traits, empathy, depression, anxiety, as well as open ended questions asking about major, major choice, and mental health diagnoses. We found that while not many people specifically wrote in the open ended questions about mental health diagnoses being prevalent for choosing psychology, our results shows that the presence of a mental health condition was significantly correlated. Therefor the person might not realize this is a reason for their major choice. While there was no significant findings directly relating mental health conditions of loved ones or experience of biases as being a reasons for choosing psychology, there was significance in the presence of a mental health condition and choosing psychology. It is important for us as researchers to understand the motivations behind joining psychology and how mental health contributes to these decisions to address biases presented to these individuals joining the field.
Faculty Advisor(s): Jennifer LaCosse, University of Michigan-Flint
Abstract 72: Increasing HPV Vaccination Rates through Provider and Patient Education
Presenter(s): Megan Inglese
Institution: University of Michigan-Flint
The Human papillomavirus (HPV) vaccination has demonstrated effectiveness against HPV-related cancers; however, vaccination rates remain below recommended guidelines. Evidence indicates that inconsistent education, language barriers between providers and patients, misinformation surrounding vaccines, and workflow inefficiencies all limit the delivery of the HPV vaccine. The project evaluated the feasibility of multifactorial quality improvement in provider and patient education to increase vaccination administration in the primary care setting. A pre-post feasibility design focused on the education of HPV vaccination guidelines, verification of vaccination status through Michigan Care Improvement Registry (MCIR), and native language automated text message reminders. Deidentified electronic medical record (EMR) data were analyzed to compare pre- and post-intervention measures. Differences were noted between pre- and post-intervention periods, but there were no statistically significant changes to the primary outcome. The absence of an EMR upgrade limited the implementation of automated text-message reminders. Despite the lack of the EMR upgrade, both educational and outreach components were implemented within the existing clinic workflow. The project highlights the need for readiness evaluation before initiating quality improvement projects. Although outcome improvements were limited, the additional enhancement of EMR functionality may have had a greater impact on HPV vaccination rates and interventions.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 73: Centering Neuroqueer Voices: A Community-Led Focus Group for Scale Development
Presenter(s): Megan Rouls
Institution: University of Michigan-Flint
Neuroqueer individuals, defined as people who are both neurodivergent and queer, often experience environments that are shaped by norms that do not reflect their communication styles, self expression, or real lived experiences. Although research has begun to describe neuroqueer identity, there are currently no tools that accurately measure societal perceptions of neuroqueer people. This project was Dr. LaCosse and I’s first phase in developing a scale that will assess attitudes toward neuroqueer individuals. The goal of this phase is to ensure that the scale is created with real lived experiences of the community it aims to present. A small number of undergraduate students here at UM-Flint that identify as neuroqueer participated in a one-hour focus group conducted either in person or on zoom. Participants discussed their experiences with identity expression, how they feel others perceive them, communication and emotional regulation expectations, and the challenges of navigating environments shaped by neurotypical and cisheteronormative norms. The discussion provided a space for participants to share their lived experiences and to describe what being neuroqueer means to them. An analysis of the discussion highlighted many recurring themes, and the emotional/social impact of being misunderstood. These insights will directly guide the next phase in our study, which will be to develop the first scale measuring perceptions of neuroqueer identity. By allowing participants to see the scale we have developed so far, their contributions will allow us to ensure that the scale is grounded in affirming community principles, while capturing the reallived experiences of the community it aims to measure.
Faculty Advisor(s): Jennifer LaCosse, University of Michigan-Flint
Abstract 74: Examining Mindfulness-Based Cognitive Therapy for Adults Who Experienced Sudden Loss During Adolescence
Presenter(s): Melinda Beavers
Institution: University of Michigan-Flint
This paper examines Mindfulness-Based Cognitive Therapy (MBCT) as a treatment for adults who have experienced sudden loss during adolescence. For some, bereavement extends into prolonged grief, leading to numerous health consequences. Sudden death can disrupt one’s ability to cope and function into adulthood, partly by hindering the development of slowly maturing brain regions. MBCT offers aid to these individuals by minimizing rumination and promoting emotional regulation. Most research prioritizes recent loss, often overlooking adults impacted by past adolescent trauma. This research gap leaves some adults unsupported, particularly when standard treatments fail to provide relief. Youth who experience sudden loss make up a small segment of this group, and the benefits of mindfulness-based interventions for these individuals in adulthood are underrepresented. Thematic analysis of the literature for MBCT and prolonged grief showed mixed findings, but suggests MBCT has broader potential and is effective at any stage of bereavement, especially benefiting secondary symptoms rather than preventing prolonged grief. However, few studies focus directly on MBCT for prolonged grief, highlighting the need for future research before it can be dismissed as a primary treatment.
Faculty Advisor(s): Amanda Taylor, University of Michigan-Flint
Abstract 75: How Hard is Math? Using Quantitative Metrics to Measure LLM Alignment to Human Intuitions of Difficulty
Presenter(s): Micah Helzerman
Institution: University of Michigan-Flint
Modern LLMs have been shown to have advanced reasoning skills, including the ability to solve Olympiad-level mathematics problems. However, there is limited research into the alignment of LLMs to human-annotated difficulty scores, which is important in educational contexts where teachers or students may ask LLMs for “easy” or “hard” questions. In this short paper, we explore various quantitative measurements from LLM generated solutions and test their predictive power for human annotated difficulty scores. While we find positive results using log probabilities and output lengths, including moderate correlations and some predicting power for difficulty scores, we find that new metrics are needed.
Faculty Advisor(s): Steve Wilson, University of Michigan-Flint
Abstract 76: Multi-Segment Inertial Sensor Analysis for Assessing Postural Stability and Fall Risk in Older Adults
Presenter(s): Milan Kadari
Institution: University of Michigan-Flint
Postural stability plays a critical role in maintaining safe movement and independence, especially for older adults. As people age, gradual declines in neuromuscular control can increase the likelihood of losing balance and experiencing falls. Falls remain one of the most common causes of injury and hospitalization among the elderly, highlighting the importance of reliable methods for evaluating balance. Traditional assessment techniques often depend on observational scoring systems or simplified motion measurements, which may not fully represent the complex mechanisms involved in maintaining stability, such as the interplay between sensory input, motor output, and cognitive processes that contribute to balance control. This study explores the use of inertial measurement unit (IMU) sensors to better understand body movement during balance tasks. Acceleration data were collected from five body locations, including the lumbar region, both wrists, and both feet, while participants performed several standing tasks with varying levels of difficulty. These tasks included standing normally, standing with eyes closed, standing with feet together, tandem standing, and single-leg standing. Placing sensors on multiple body segments allowed the study to capture how different parts of the body contribute to maintaining balance. To analyze the collected motion signals, both time-domain and frequency-domain techniques were applied. Measurements such as root mean square (RMS) acceleration and motion range were used to describe movement variability, while Fast Fourier Transform (FFT) and Power Spectral Density (PSD) helped identify how motion energy was distributed across different frequencies. The rate of change of acceleration, known as jerk, was also examined to detect rapid corrective movements during balance control. The findings suggest that higher RMS values, increased spectral power, and greater jerk are associated with more frequent balance corrections, indicating reduced stability. More challenging tasks, particularly single-leg standing, showed noticeably higher motion variability. Overall, this multi-segment IMU-based analysis provides a more detailed and objective approach for studying balance and may support improved fall-risk assessment in older adults.
Faculty Advisor(s): Linda Zhu, University of Michigan-Flint
Abstract 77: Improving Diabetic Eye Exam Rates in a Rural Health Clinic
Presenter(s): Natika Mitchell-Cowie
Institution: University of Michigan-Flint
Abstract Background The American Academy of Ophthalmology (AAO) recommendation is for patients with type 2 diabetes mellitus (T2MD) to have a diabetic eye exam (DEE) upon diagnosis, for type 1 diabetes mellitus (T1DM) to have a DEE within 5 years of diagnosis, and subsequent DEEs should be performed annually for both T1DM and T2DM. Local Problem Less than 36% of the patient population at this rural health clinic receive DEEs per AAO guidelines. Needs for standards of care are higher than the area can support. Purpose The aim of this study was to evaluate the impact of an educational session on patient acquisition of DEE and provider documentation of the exam after it was obtained. Methods An educational intervention occurred with staff and providers on the AAO guidelines for DEE frequency, retinopathy, and where to document the exam. Patients were provided with a handout on retinopathy and blank results sheet to take to their DEE for the optometrist’s office to fax back to the clinic with results. The study included all patients diagnosed with diabetes, excluding those patients with gestational diabetes or diabetics who were clinically blind. Results The study evaluated 100 pre-intervention patients and 90 post-intervention patients and when compared looking at numbers of DEE documentation and follow-up, there were no significant findings statistically. Conclusion Although no statistical significance was found in this study, the clinical significance of increased numbers of DEE documentation and follow-up were a positive sign that with additional time there would have been a statistically significant outcome.
Faculty Advisor(s): Roxanne Buterakos, University of Michigan-Flint
Abstract 78: Enhancing Testosterone Deficiency Screening in Primary Care Using the Adam Questionnaire
Presenter(s): Noah Wiggins
Institution: University of Michigan-Flint
Background: Testosterone deficiency (TD) is frequently underdiagnosed in primary care due to nonspecific symptoms, inconsistent screening practices, and reliance on laboratory testing without standardized symptom assessment. Delayed identification may adversely affect the quality of life and overall health in adult male patients. Purpose: This Doctor of Nursing Practice quality improvement project aims to enhance early identification of testosterone deficiency through implementation of the Androgen Deficiency in the Aging Male (ADAM) questionnaire as a standardized symptom-based screening tool in a rural primary care clinic. Methods: A quasi-experimental pre–post design is being used. The intervention includes provider education on evidence-based TD guidelines and integration of the ADAM questionnaire into routine clinical workflows for adult male patients presenting with symptoms suggestive of hypogonadism. A retrospective chart review will compare the number of serum testosterone tests ordered and the proportion of low testosterone results before and after implementation. Results: Data collection and analysis are ongoing. Conclusion: Pending results will evaluate whether the incorporation of the ADAM questionnaire improves guideline-based screening and supports the timely identification of testosterone deficiency in primary care.
Faculty Advisor(s): Kristina Yaklin, University of Michigan-Flint
Abstract 79: Developing Secure Networking Infrastructure at the HEART Clinic
Presenter(s): Noah Lutz, Drew Conner
Institution: University of Michigan-Flint
Reliable and secure networking infrastructure is essential for modern healthcare environments, where digital systems are core for all operations This project examines the development and implementation of networking systems in a clinical setting through work conducted for the University of Michigan-Flint HEART Clinic at the Insight Institute of Neurosurgery and Neuroscience. The work focuses on the practical challenges of establishing and maintaining network infrastructure in environments with significant security requirements and limited financial resources. This poster presentation will outline the planning and deployment process used to develop the clinic’s networking systems. It will present diagrams of the proposed network architecture, including device layout, demonstrations of the limitations at the facility, and connectivity between systems used to support clinic operations. The poster will also highlight key considerations in designing a network for a healthcare environment, such as maintaining secure access, supporting future expansions, and selecting cost-effective hardware solutions within budget constraints. In addition to the technical design, the presentation will discuss the implementation process, including documentation practices, installation planning, and strategies for ongoing maintenance and support. By illustrating both the design and deployment stages, the poster aims to demonstrate how networking infrastructure can be developed to meet real-world healthcare needs while balancing security, reliability, and financial limitations. Through this work, the project contributes practical insight into the deployment of healthcare IT infrastructure and highlights the value of hands-on research experiences that connect academic learning with real-world technical environments.
Faculty Advisor(s): Zahid Syed, University of Michigan-Flint
Abstract 80: Active listening and nonverbal engagement as strategies to enhance adolescent satisfaction in telehealth: A qualitative analysis
Presenter(s): Nour Almounajed, Melisa Gerard, Phani Davu
Institution: University of Michigan-Flint
The “Telemedicine and Doctor-Patient Communication Theoretical Framework” proposed by Edward Miller (2002) suggests that characteristics of both patients and healthcare providers shape communication during telehealth visits, which in turn influences patient health outcomes. Guided by this framework, the current ongoing study examines which provider characteristics and behaviors foster effective communication with adolescents during videobased telehealth visits. Recruitment flyers targeting parents and adolescents were distributed in several community locations, including pediatric clinics in Genesee County, Michigan. Adolescents participated in virtual interviews where they discussed their experiences with telehealth providers, including how providers demonstrate attentiveness and create an environment of trust and comfort when addressing sensitive topics. Interviews were recorded and transcribed by a transcription company. Two independent researchers reviewed the transcripts and identified recurring themes through qualitative analysis. The sample consisted of 81 adolescents ages 14 to 17: 16 (19.8%) were 14 years old, 20 (24.7%) were 15, 23 (28.4%) were 16, and 22 (27.2%) were 17. Participants were racially and ethnically diverse, with 45 (55.5%) identifying as non-Hispanic White, 20 (24.7%) as non-Hispanic Black or African American, 5 (6.2%) as White Latino, and the remaining participants representing other racial and ethnic backgrounds. Adolescents identified several provider behaviors that enhanced communication during video visits. This includes professional appearance, keeping cameras on, nonverbal cues such as nodding, smiling, maintaining eye contact, initiating casual conversation to build rapport, using active listening statements (“I understand” or “It sounds like you are saying..”), and using visual aids like graphs or images when explaining health information. By recognizing what makes telehealth communication effective, healthcare systems can design strategies that increase adolescent satisfaction and promote better health results.
Faculty Advisor(s): Gergana Kodjebacheva, University of Michigan-Flint
Abstract 81: Promoting Hepatitis B Prevention: An Educational Initiative in Urban Outpatient Settings
Presenter(s): Paige Parker, Alexis Bozek
Institution: University of Michigan-Flint
Hepatitis B virus (HBV) remains a significant global and national public health concern, with substantial morbidity and mortality despite the availability of a safe and effective vaccine. In 2023, the Centers for Disease Control and Prevention (CDC) updated hepatitis B vaccination guidelines to recommend vaccination for all adults aged 19–59 years and for adults aged ≥60 years with or without risk factors, highlighting the need for improved provider and patient education. This project addresses gaps in HBV prevention among adults aged 33 years and older in an urban outpatient internal medicine setting, where vaccination practices are inconsistent and often provider-dependent. Using a quasi-experimental, nonequivalent control group design, the study evaluates the impact of a standardized educational intervention for a nurse practitioner and eligible patients. The provider intervention consists of an educational PowerPoint on HBV risks and updated CDC guidelines, while the patient intervention includes verbal education and CDC-developed written materials. Outcomes measured over a threemonth period include documentation of HBV education, antibody titer ordering, and vaccine administration using an electronic medical record smart-phrase. The primary aims are to increase provider guideline adherence and improve patient knowledge to support informed decision-making. This initiative seeks to reduce missed vaccination opportunities, promote standardized preventive care, and ultimately contribute to decreased HBV transmission and improved long-term health outcomes.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 82: The Role of TRPV4 Calcium Channel Activity under Oxidative Stress in Articular Chondrocytes
Presenter(s): Paulina Rivera, Benjamin Bush
Institution: University of Michigan-Flint
Osteoarthritis (OA) is a prevalent degenerative joint disease primarily affecting adults over the age of 50. OA impacts cartilage maintenance by altering the functional integrity of chondrocytes. Key to this altered function are Transient Receptor Potential Vanilloid 4 (TRPV4) channels that mediate calcium influx in chondrocytes during mechanical stimulation. Previous studies have shown that when this influx is limited, chondrocytes express an anabolic phenotype. However, when chondrocytes are exposed to an OA environment, TRPV4 calcium influx is much higher and results in chondrocytes expressing a catabolic phenotype. Here, I seek to determine the cause of this gain-of-function of TRPV4 channels and if this increase is related to aging. As one ages, oxidative stress accumulates from age-related inflammation, leading to the generation of reactive oxygen species (ROS). However, our knowledge of ROS’s influence on mechanosensory channels such as TRPV4 remains limited. I hypothesize that increased ROS will increase calcium influx through the TRPV4 channel. To test this hypothesis, I will expose ATDC5 chondrocytes to hydrogen peroxide to induce oxidative stress at various time points. I will verify increased ROS levels using the fluorescent assay, CellROX. I will then measure changes in calcium influx during oxidative stress using the FLUO-4 fluorescent calcium chelator. Finally, I will evaluate the role of TRPV4 in calcium influx by applying GSK205, a TRPV4 inhibitor, and measure the change compared with and without the inhibitor. I predict that oxidative stress will increase the activity of the TRPV4 calcium channel in chondrocytes. The imbalance of calcium influx will then signal further accumulation of ROS in articular chondrocytes. Future experiments will investigate the relationship between disruption in calcium homeostasis, whether it begins with mitochondrial ROS production, or if increased TRPV4 activity activates mtROS acting as a feed-forward mechanism that amplifies oxidative stress in chondrocytes.
Faculty Advisor(s): Randall Duncan, University of Michigan-Flint
Abstract 83: Analyzing the Transcription Network of Caulobacter crescentus
Presenter(s): Priscilla Mac-Kittah
Institution: University of Michigan-Flint
To survive, bacteria need to sense and respond to fluctuating environments. Transcription factors play a central role in this process by detecting chemical, metabolic, or stress‑related signals and altering gene expression in ways that drive changes in cell physiology. The model Alphaproteobacteria Caulobacter crescentus is an excellent system to study how transcriptional regulators impact environmental acclimation. Here, we examined the previously uncharacterized XRE-family TF CCNA_03612 and the LRP-family TF CCNA_03281. Publicly available fitness data suggest that disruption of CCNA_03612 or CCNA_03281 increases cellular fitness in numerous stress conditions, indicating that these TFs may normally contribute to regulatory pathways that influence cellular survival under stress. Given their similar fitness profiles, we asked whether CCNA_03612 and CCNA_03281 control related transcriptional targets. Using ChIP‑seq, we defined the global binding profiles of both regulators. We detected 821 binding sites for CCNA_03612 and 91 binding sites for CCNA_03281. Despite their parallel stress phenotypes, the two TFs shared only one overlapping binding site, the promoter region of CCNA_03612. This finding suggests that CCNA_03612 is a direct target of CCNA_03281 and that CCNA_03612 likely autoregulates its own expression. To place CCNA_03612 within a broader regulatory context, we compared its binding profile to that of CdxA, a previously characterized XRE-family TF. CCNA_03612 and CdxA showed co-localized binding at genes required for pH homeostasis (carbonic anhydrase) and motility (CheYIII), indicating shared regulatory logic. Overall, this work expands our understanding of transcription factor targets in C. crescentus and identifies CCNA_03612 as a key regulator in the bacterium’s response to environmental stress.
Faculty Advisor(s): Maeve McLaughlin, University of Michigan-Flint
Abstract 84: Enhancing Orthopedic Patient Outcomes Through Anxiety Screening and Provider Education.
Presenter(s): Rashahla Yazdani
Institution: University of Michigan-Flint
Anxiety is a common yet frequently underrecognized condition among orthopedic patients and can negatively influence pain perception, treatment adherence, and postoperative recovery. Despite growing evidence linking psychological factors to surgical outcomes, routine anxiety screening is rarely integrated into orthopedic outpatient workflows. The purpose of this Doctor of Nursing Practice (DNP) quality improvement project was to evaluate the feasibility and impact of implementing a brief anxiety screening process combined with structured provider education in an orthopedic outpatient specialty practice. A quasi-experimental pre–post quality improvement design was used. The intervention included provider education on the relationship between anxiety, pain, and recovery; training on the use of the Generalized Anxiety Disorder-7 (GAD-7) screening tool; and implementation of a provider resource toolkit designed to support screening, communication strategies, and referral pathways. Interdisciplinary providers and clinical staff participated in the educational intervention. Pre- and post-intervention surveys assessed provider awareness, confidence, and perceived feasibility of incorporating anxiety screening into routine care. Following implementation, providers reported improvements in awareness of anxiety’s impact on orthopedic outcomes (83%), confidence addressing anxiety with patients (75%), and identification of anxiety-related care barriers (67%). Improvements were also observed in patient-centered communication (83%) and perceived feasibility of integrating psychosocial screening within clinic workflow (75%). These findings support the feasibility of integrating brief anxiety screening and targeted provider education into specialty outpatient practice, promoting a more holistic, patientcentered approach to orthopedic care.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 85: Closing the Communication Gap: A Quality Improvement Initiative Using IPASS in the Trauma ICU
Presenter(s): Rose Klann, Tina Pestun
Institution: University of Michigan-Flint
Ineffective handoff communication remains a significant contributor to preventable patient harm, particularly in high-acuity intensive care unit (ICU) settings where frequent transitions of care occur. This quality improvement project aimed to implement a standardized I-PASS (Illness Severity, Patient Summary, Action List, Situational Awareness and Contingency Planning, and Synthesis by Receiver) handoff system in a Level I trauma center ICU to improve continuity of care, communication quality, and provider satisfaction. The intervention included structured provider education, integration of an I-PASS handoff template into the electronic medical record, and ongoing reinforcement during clinical practice. Providers completed preand post-implementation surveys assessing handoff satisfaction, perceived accuracy, and completeness of information exchange. Compliance with I-PASS utilization was monitored throughout the implementation period. Data collection occurred from October 7, 2025 through January 7, 2026. Preliminary findings suggest improved provider confidence and satisfaction with handoff communication following implementation. This project demonstrates the feasibility of implementing a standardized handoff framework within a complex ICU environment and highlights the role of structured communication in enhancing patient safety and interdisciplinary collaboration. Findings support broader adoption of I-PASS as a sustainable strategy to improve handoff reliability and continuity of care in acute care settings. Keywords: I-PASS, patient handoff, quality improvement, patient safety, intensive care unit
Faculty Advisor(s): Kristina Yaklin, University of Michigan-Flint
Abstract 86: Characterization of the antioxidant effects of cannabinoids on human cells
Presenter(s): Rose Leganr, Lilian Moore, Molly Maier, Lauren Showler
Institution: University of Michigan-Flint
Interest in alternative treatments, and Cannabis sativa specifically, is high in the general public in the US. Sales of cannabidiol (CBD) products has been popular in the United States for years. There is some promise for the use cannabis derived products for pain management and other diseases and medical use of cannabis is legal in a number of states, with some states also legalizing recreational use. While CBD and tetrahydrocanninol (THC) are often stated to be responsible for the medicinal effects of cannabis, several other cannabinoid molecules exist in the plant material that may also have medicinal effect. One of these compounds is hexocannabitriol. In this study, we plan to characterize the biological effects of not only hexocannabitriol, but also many of the synthetic intermediates made during the synthesis of hexocannabitriol. We will determine the ability of these compounds to activate the cellular antioxidant response element via Nrf2 in human embyonic kidney (HEK293) cells.
Faculty Advisor(s): Jessica Kelts, University of Michigan-Flint
Abstract 87: Structural determinants of sirt2 function: implications for ran regulation in neurodegeneration
Presenter(s): Samantha Jarbou
Institution: University of Michigan-Flint
Sirtuin 2 (SIRT2) is a NAD⁺-dependent deacetylase that controls nuclear transport and cytoskeletal dynamics. Dysregulated acetylation by SIRT2 has been connected to neurodegenerative diseases, including Alzheimer’s and Parkinson’s. One key substrate is the small GTPase RAN, which SIRT2 deacetylates at lysine 71, a modification that can affect nucleocytoplasmic transport, a pathway often disturbed in neurodegeneration. Aimed to compare the activity of wild-type (WT) SIRT2 with a deletion mutant lacking residues 292–303 (Δ292–303) to figure out whether this region is required for RAN deacetylation and neurodegeneration-relevant function. SIRT2 variants were expressed in E. coli BL21 (DE3) following IPTG induction and genetic code expansion for TFAcK incorporation. Proteins were purified by immobilized metal affinity chromatography (IMAC) and size-exclusion chromatography (SEC), with purity confirmed by SDS-PAGE. Deacetylase activity was measured by incubating purified SIRT2 with acetylated RAN over a 0–120 min time course, followed by anti-acetyl-lysine Western blot and densitometric analysis. All results were confirmed across independent replicates with WT and mutant controls. The SDS-PAGE confirmed successful expression and purification of both SIRT2 variants. SEC improved purity and yielded monomeric fractions. Western blot analysis showed timedependent RAN deacetylation by WT SIRT2, whereas Δ292–303 showed no measurable activity. Quantification confirmed complete loss of activity in the mutant, highlighting the functional importance of this structural region. Results identify residues 292–303 of the SIRT2 as essential for the deacetylation of RAN. Given SIRT2’s role in neuronal protein regulation, these findings provide insight into how specific structural elements contribute to enzymatic activity and can guide the design of selective SIRT2 inhibitors for neurodegenerative disease therapeutics.
Faculty Advisor(s): Kevin Tang, University of Michigan-Flint
Abstract 88: From Enforcement to Engagement: A Community-Based Problem-Oriented Public Safety Evaluation
Presenter(s): Samantha Jarbou
Institution: University of Michigan-Flint
Problem-Oriented Public Safety (POPS) emphasizes data-driven, community-centered strategies to address persistent public safety concerns. This project examines whether the implementation of a POPS initiative by the Department of Public Safety (DPS) in the River Village community influenced resident perceptions of police presence and neighborhood safety. Using the SARA model (Scanning, Analysis, Response, Assessment), DPS engaged residents through three community meetings to identify key challenges, including limited youth programming, high perceived crime, inconsistent curfew enforcement, and delayed police response times due to contractual patrol limitations. In response, DPS implemented weekly youth-focused programming beginning October 1st, including games, outdoor activities, seasonal events, and community meals, while maintaining consistent patrols and building relationships with apartment management. Participation ranged from 5 to 30 children per week. To assess impact, a six-question perception-of-safety survey was administered to residents in River Village townhomes (intervention group) and towers (control group). Results demonstrated a statistically significant improvement in perceptions of police and neighborhood safety among townhome residents following the intervention, compared to the control group. These findings suggest that sustained relationship-building and youth engagement can positively influence community perceptions of law enforcement. Improving trust and perceived legitimacy is critical, as it increases residents’ willingness to seek assistance and collaborate with police to enhance overall community safety.
Faculty Advisor(s): Melissa Sreckovic, University of Michigan-Flint
Abstract 89: Transforming Food Deserts into Food Forests
Presenter(s): Sarah Cagle
Institution: University of Michigan-Flint
My presentation will outline a plan for transforming food deserts into food forests. I will provide information about food deserts, share information on some of the policies that attempt to address this problem, and which of these polcies have been successful or unsuccessful. Then, I will detail my journey to start a food forest in Flint.
Faculty Advisor(s): Derwin Munroe, University of Michigan-Flint
Abstract 90: Transforming Nurse Well-Being: Evaluating a 90-Day Mindfulness Intervention to Reduce Burnout in Inpatient Settings
Presenter(s): Sarah Forsyth
Institution: University of Michigan-Flint
Background: Nurse burnout is a persistent threat to workforce stability, patient safety, and care quality, with emotional exhaustion identified as a primary driver. Mindfulness based interventions show promise but often lack feasibility in high-acuity inpatient settings. Purpose: This quality improvement project evaluated the feasibility and potential impact of a 90-day app-based mindfulness intervention on nurse burnout in an inpatient hospital setting. Methods: A quasi experimental pre–post design was used with full and part time inpatient registered nurses on two medical surgical units at a 560 bed community hospital. Participants were encouraged to engage in a daily 10 minute mindfulness practice using the Headspace app for 90 days. Burnout was measured using the Maslach Burnout Inventory–Human Services Survey (MBI-HSS). Paired analyses were conducted for participants completing both baseline and follow-up assessments. Results: Twelve of eighteen enrolled nurses completed paired assessments. Mean emotional exhaustion decreased from 25.7 to 24.0, depersonalization decreased from 10.0 to 9.1, and personal accomplishment increased from 33.5 to 37.0. While changes did not reach statistical significance, consistent directional improvement across all three subscales was observed. Attrition reflected known engagement challenges in digital wellness interventions among inpatient nurses. Conclusion: Findings suggest that brief, app based mindfulness interventions are feasible and may contribute to reductions in nurse burnout, particularly emotional exhaustion. Organizational support and workflow integration are critical for sustained engagement and scalability.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 91: Strengthening Psychiatric Consult Pathways for Hospitalized Older Adults Through Depression Screening Education
Presenter(s): Shelley Hughes
Institution: University of Michigan-Flint
Depression is prevalent among adults aged 65 and older, especially those hospitalized, and is linked to increased illness and reduced quality of life (Moodley & Maisto, 2022). Despite the United States Preventive Services Task Force (USPSTF) recommendations for annual screening, depression often goes undiagnosed in this group. In a mid-Michigan hospital, mental health screening emphasizes suicide risk using the C-SSRS, potentially missing broader depressive symptoms. Nurses’ limited familiarity with the PHQ-9, a comprehensive depression screening tool, may result in missed opportunities for psychiatric consultation before discharge. This Doctor of Nursing Practice (DNP) quality improvement project aimed to increase nurses’ understanding of the PHQ-9 to improve early recognition of depression and support timely psychiatric consultation for older inpatients. Nurses received structured education on the PHQ9’s purpose, scoring, interpretation, and its role in supplementing the C-SSRS screening. Preand post-education data were reviewed to assess changes in nurses’ understanding and psychiatric consult initiation. After the intervention, nurses showed improved knowledge of the PHQ-9 and initiated more psychiatric consultations, supporting earlier identification and evaluation of depression in hospitalized older adults. Learning Objectives 1. Medical–surgical nurses will be able to describe the purpose, scoring, and clinical interpretation of the PHQ-9 as an evidence-based depression screening tool for adults aged 65 and older, and how it is different from the C-SSRS to help identify broader depressive symptoms. 2. Medical-surgical nurses will be able to identify how applying the PHQ-9 during an intake process can help initiate timely psychiatric consultations for hospitalized older adults, supporting earlier recognition of depressive symptoms and facilitating evaluation prior to discharge.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 92: Positive Caregiver–Child Interactions and Child Aggressive Behavior in Lowand Middle-Income Countries
Presenter(s): Tasheikya Hunter
Institution: University of Michigan-Flint
This study examines the association between positive caregiver-child interactions and child aggressive behavior in low- and middle-income countries. Positive interactions between caregivers and children include using non-violent discipline such as explaining why a behavior is inappropriate and choosing non-violent guidance methods and engaging in developmental activities like reading together, telling stories, playing games, and singing songs. Child aggressive behavior is defined as actions such as kicking, biting, or hitting other children or adults. Through a comprehensive literature review, this study explores whether positive caregiver-child interactions are negatively associated with aggressive behaviors among children in low- and middle-income countries.
Faculty Advisor(s): Julie Ma, University of Michigan-Flint
Abstract 93: Improving Depression Screening in Primary Care: A Quality Improvement Project
Presenter(s): Thi Tran, Sheema Khan
Institution: University of Michigan-Flint
Depression is a common yet frequently underrecognized condition in adult primary care, contributing to reduced quality of life and increased healthcare utilization. This quality improvement project evaluated a targeted educational intervention aimed at supporting early identification and management of depression among adults aged 19 and older. The intervention combined staff education on depressive symptom recognition with integration of the Patient Health Questionnaire 9 into routine clinical workflow. Pre and post intervention data included patient record audits, and documentation of screening practices. Results demonstrated modest improvements in staff confidence and awareness related to depression, with slight increases in consistent use of standardized screening tools. Although changes were not statistically significant, qualitative feedback suggested enhanced attention to depressive symptoms and more structured approaches to follow up care. These findings highlight the challenges of implementing evidence based practices in real world primary care settings and underscore the need for sustained training, ongoing support, and system level strategies. This project provides a foundation for incremental quality improvement efforts and supports future initiatives to strengthen depression screening and management across adult populations.
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 94: The Effects of Marijuana on Mental Health in Youth
Presenter(s): Tiffany McClure
Institution: University of Michigan-Flint
This study evaluated whether educating providers at a community mental health facility on the harmful effects of marijuana use in youth would increase the number of youth who subsequently received education on the topic. An educational meeting was conducted to review current research regarding the harmful effects of marijuana use in adolescents and young adults. Provider documentation was reviewed for the three months prior to the educational intervention and for three months following the intervention to assess its impact. An educational handout was also developed to guide provider–patient discussions and could be distributed directly to patients as needed. The study focused specifically on patients aged 13–26; however, the educational content may also be applicable to individuals in later adulthood. The poster I plan to present highlights current research on marijuana use during brain development, including its mental health implications as well as associated emotional and physical changes. It also summarizes the study that was conducted and presents the resulting outcomes.
Faculty Advisor(s): Jill Brown, University of Michigan-Flint
Abstract 95: Performance and Privacy in Vehicle-to-Infrastructure Authentication: a ZKPOTP Approach
Presenter(s): Tom Rose
Institution: University of Michigan-Flint
As Vehicle-to-Infrastructure(V2I) communication becomes more prevalent, securing and maintaining the privacy of authentication in Intelligent Transportation Systems(ITS) is an important challenge. Traditional authentication methods such as Public Key Infrastructure(PKI) can expose sensitive information and may not scale effectively in high traffic conditions. The framework being investigated in this study is intended to overcome the challenges present in existing authentication methods by combining One-Time Passwords(OTPs) and Zero-Knowledge Proofs(ZKPs) to allow vehicles to authenticate with Roadside Units(RSUs) without the need to reveal underlying credentials. In order to test this approach, an integrated Simulation Environment(ISE) is in development that combines a traffic simulator (SUMO) with a network simulation layer, achieved either through implementation of OMNeT++ or approximated using a modeled communication system. ZoKrates is utilized to manage and verify the zk-SNARK proofs, and the entire system is wrapped in a Python-based orchestration layer running on a Linux VM. This ISE allows us to measure the scalability, resistance to replay attacks, and authentication latency of the proposed framework in various traffic conditions, densities, and scenarios. An array of configurations, combining zk-SNARK techniques, road networks, traffic conditions, and network limitations, are available for testing in order to examine the effects of varying conditions on authentication performance. The results will provide insight into the feasibility of the proposed framework, as well as indicate avenues for future research.
Faculty Advisor(s): Stephen Turner, University of Michigan-Flint
Abstract 96: Providing Education to Reduce Vaping in Middle School Aged Youth
Presenter(s): Tresa Alaniz
Institution: University of Michigan-Flint
Providing education using “You and me together vape free” to reduce vaping in Middle School aged youth.
Faculty Advisor(s): Kristina Yaklin, University of Michigan-Flint
Abstract 97: State-of-Health Monitoring of Lithium-Ion Batteries Using a Laser Doppler Vibrometer
Presenter(s): Varun Rangaraj
Institution: Kettering University
Accurate measurement of battery state of charge (SOC) and state of health (SOH) is necessary for optimizing battery performance, ensuring safe operation, and measuring battery lifetime. Battery management system (BMS) measurements and calculations are common nondestructive ways to determine SOC and SOH. Due to a lack of measurement standards and algorithm adaption to aging batteries, alternative methods of measuring SOC and SOH are being investigated. This study examines how mechanical properties of a lithium-ion battery at different SOC and SOH change by measuring modal frequencies with a scanning laser doppler vibrometer (SLDV) as the battery is subjected to subsonic vibration. The work shows that SLDV measurements and modal analysis can be used to determine SOC and SOH based on known results for specific states of health and charge. Varying SOC was obtained by battery discharging and charging, while varying SOH was obtained by subjecting batteries to electrical, thermal, and mechanical damage.
Faculty Advisor(s): Foroogh Rouhollahi, Kettering University
Abstract 98: The ABCs of Diabetes: Measuring Quality Indicators and Implementing Evidence-Based Interventions in Endocrinology Clinical Practice.
Presenter(s): Veronica Kelty, Marilyn Stancroff
Institution: University of Michigan-Flint
BACKGROUND: Diabetes remains a leading contributor to morbidity, mortality, and healthcare utilization in the United States. Adherence to evidence-based quality indicators is critical for optimizing glycemic control and enhancing long-term patient outcomes. Unfortunately, gaps persist in consistent monitoring and documentation of recommended diabetes quality measures in clinical practice. Systemic evaluation and provider-focused interventions may improve adherence to these standards of care. PURPOSE: This quality improvement project aims to evaluate and enhance the implementation of diabetes quality indicators within an endocrinology clinical setting. METHODS: Data is being collected through real-time EMR review of all patients with diabetes, as well as retrospective chart audits of their visits within the previous year. The project assesses whether established quality indicators are documented and addressed during patient encounters. To promote adherence, providers received targeted education on diabetes quality indicators and were given a visual clinical aid to prompt discussions during patient encounters. Ongoing analysis was done to compare real time visits and pre-intervention data to determine improvements in adherence to quality measures and identify areas for continued optimization of diabetes management in endocrinology practice. DATA VARIABLES: Hemoglobin A1c, blood pressure, LDL, monofilament foot exam documentation, statin therapy, ACE inhibitor or Angiotensin II receptor blocker therapy, tobacco and alcohol use, referral to diabetic education classes (for new diagnoses or medication changes), podiatry and diabetic eye exam documentation, and recent hospitalizations.
Faculty Advisor(s): Roxanne Buterakos, University of Michigan-Flint
Abstract 99: Applying the United States Preventive Services Task Force Fall Prevention Guideline
Presenter(s): Wafa Nasser
Institution: University of Michigan-Flint
Applying the United States Preventive Services Task Force Fall Prevention Guideline IN Primary Care Clinic Falls are a leading cause of morbidity, mortality, and loss of independence among adults aged 65 years and older. In the United States, approximately 14 million older adults nearly one in four reports falling each year, with many experiencing injuries that require medical treatment or result in activity limitation. Despite strong evidence supporting fall prevention strategies, fall risk assessment and prevention plans are often inconsistently documented and implemented in primary care settings. The United States Preventive Services Task Force (USPSTF, 2024) recommends exercise interventions for community dwelling adults aged 65 years and older at increased risk for falls and advises clinicians to individualize multifactorial interventions based on patient specific risk factors and preferences. This DNP quality improvement project aims to improve provider adherence to USPSTF fall prevention guidelines by enhancing documentation and implementation of fall prevention plans in a community primary care clinic. Nurse practitioners, primary care providers, and medical assistants participated in a brief, in-person educational intervention consisting of a PowerPoint presentation, case-based discussions, and evidence-based reference materials. A pre- and postintervention chart review conducted over a three-month period to compare rates of documented fall risk assessments, fall prevention plans, and patient agreement with recommended interventions. Project outcomes will be analyzed to evaluate changes in clinical practice and disseminated through a DNP poster presentation. Learning Objectives • Participants will be able to describe USPSTF-recommended fall prevention interventions for adults aged 65 years and older. •Participants will be able to implement and document evidence-based fall prevention plans in the electronic health record
Faculty Advisor(s): Marilyn Filter, University of Michigan-Flint
Abstract 100: The Role of Furin in Cancer Progression under the Hypoxic Tumor Environment
Presenter(s): Xu Tang
Institution: University of Michigan-Flint
This presentation will discuss the role of furin in cancer progression under the hypoxic tumor environment based on the two academic articles.
Faculty Advisor(s): You-Shin Chen, University of Michigan-Flint
Abstract 101: Metal-Chitosan Composites Synthesis and Characterization as Biodegradable Alternative Materials
Presenter(s): Kate Garrison
Institution: Kettering University
Synthetic plastics and polyelectrolyte membranes are essential for daily and industrial applications, but contribute significantly to environmental pollution due to their nonbiodegradability and reliance on petroleum-based resources. Through the synthesis of biodegradable thin films, this project aims to develop metal-chitosan composites that offer sustainable alternatives with the ability to incorporate specific properties. Crustacean waste shells supply abundant chitosan when treated with proper waste management and reutilization processes, and not only mitigate solid waste environmental challenges, but also offer excellent alternatives to fossil-fuel-based plastics, without compromising the environment. In this project, the metal-chitosan composites were synthesized with various metal types (Cu2+, Fe2+, Zn2+, Au nanoparticles, etc.) and loadings with chitosan as the biopolymer base, and were evaluated using a range of characterization techniques, such as FTIR-ATR, Environmental Scanning Electron Microscopy (ESEM), and Differential Scanning Calorimetry (DSC) to assess their properties. The results of the composites’ bonding types, thermal features, etc., demonstrate that these composites afford very promising outcomes.
Faculty Advisor(s): Sihe Zhang, Kettering University