Masters Theses

Permanent URI for this collection

Duke migrated to an electronic-only system for theses between 2006 and 2010. As such, theses completed between 2006 and 2010 may not be part of this system, and those completed before 2006 are not hosted here except for a small number that have been digitized.

Browse

Recent Submissions

Now showing 1 - 20 of 1389
  • ItemOpen Access
    Gamers' Relationships with Their Avatars & Fanfiction: An Exploration of Player-Avatar Relationships Through a Digital Project
    (2023) Zhou, Yuchen

    In this thesis, I discussed video game players’ relationships with their avatars, and how further differentiations can be made from the existing categories using fanfiction as an avenue. Past studies on player-avatar relationships and fanfiction surrounding the questions of identity and the process of identification have been examined. Among those studies, the player-avatar categories proposed by Jaime Banks and Nicholas David Bowman in their 2021 article served as the baseline for this project's development. Drawing upon Jon Robson and Aaron Meskin’s concept of “self-involving fictions,” or SIF, I proposed a new type of player-avatar relationship, “avatar-as-SIF.” This relationship emerges when players decide to embark on a journey with their avatar relationships beyond the original scope provided by the game, through fanfiction. This relationship manifests that which was previously digital into products that have impacts in the real world, for both the players and the audiences. This project then provides an actualization example of the “player-as-SIF” relationship through two videos composed of animated composited photographs between the avatars and the real world, featuring a narrative aligned with fanfiction.

  • ItemOpen Access
    Bridging Realities: Navigating Between Physical and Virtual Installation Art
    (2023) Zhang, Yiwen

    Many installation art projects have appeared from the 20th to the 21st century. There are diverse types of installation art. Due to the development and implementation of technology, installation art is divided into physical installation art that is entirely physical, without any digital application. On the other hand, virtual installation art implements virtual elements to physical installation art or is a complete virtual experience. This article discusses the relationship between these two art forms, from their definitions to the examples of different artists and works that have become famous in the field. It also explores how different digital platforms can serve as a medium for the completion of installation artists' work. A collaborative project is introduced at the end to deeply explore the relationship between physical installation art and virtual installation art. The project led to the conclusion that one can find elements and characteristics of physical installation art applications in a completely virtual installation artwork. However, with the development and usage of technology and digital tools, artists can flexibly create the installation art space they want to complete.

  • ItemOpen Access
    Modeling Generative Artificial Intelligence
    (2023) Xiong, Haochen

    The release of ChatGPT-4 has led to the prevalent use of a new term in the field of artificial intelligence (AI): generative AI. This paper aims to understand generative AI more thoroughly and place it within a broader framework of models and their relationship with knowledge. By closely examining AI’s historical development, this paper will first introduce the concept of emergence to distinguish generative AI from other forms of AI. Second, by theorizing generative AI as models, this paper will evaluate their significance in human knowledge production. Third, by classifying generative AI specifically as generative models, this paper will demonstrate their unique potential, especially for art creation.

  • ItemOpen Access
    See You Never: Exclusion in Electroencephalography and Neurotechnology
    (2023) Wilson, Victoria

    Electroencephalography (EEG), a neuroscience method which requires sustained access to the scalp and hair, has many clinical and research applications. It is an essential feature of the rapidly growing consumer neurotechnology market. Neuroethicists have criticized EEG for being unaccommodating to phenotypic differences in hair type - a flaw which contributes to the systematic exclusion of minority groups from research. This exclusion legitimizes concerns about the generalizability of EEG research and effectiveness of EEG-based technologies. The following report employs a review of the most current literature across neuroscience, ethics, and technology publication sources to demonstrate how exclusion EEG research creates gaps in theoretical knowledge that disproportionately impact minorities and have profound implications for medical and consumer products. This paper summarizes the many applications of EEG and examines the impact of exclusion on EEG-based research and technology development. It outlines the risks of maintaining exclusion and provides policy recommendations for how to mitigate those risks by prioritizing inclusion in research methods.

  • ItemOpen Access
    Bennett Place AR: Evaluating an AR Application at a Historic Site from a UX Design Perspective
    (2023) Shi, Ruojin

    This thesis explores the integration of User Experience (UX) Design in digital humanities, with a focus on Augmented Reality (AR) at Bennett Place. It draws on Brennan's public digital humanity concept, emphasizing the need for public-oriented approaches in digital humanities. The research employs UX design methods, adhering to a workflow comprising research, ideation, design, and user testing.In the research stage, Bennett Place's historical context and visitor personas are analyzed to inform design objectives and user expectations. The ideation stage addresses content design and AR technology selection, aiming for effective information delivery and inclusive user experiences. The design stage details the digital project's implementation. The final delivery of this thesis is an iOS AR app the final stage involves user testing to evaluate the application of AR in enhancing on-site visiting experience at Bennett Place. Although the testing results are not definitive, they provide valuable insights for future digital humanities projects, particularly in public engagement. This thesis demonstrates the effectiveness of AR in enhancing on-site visiting experience at Bennett Place and highlights the potential for incorporating UX methodologies in digital humanities, advocating for more user-focused, engaging, and informative experiences.

  • ItemOpen Access
    Innovation and Development of Biomedical Industry Clusters in Jiangsu Province: A Technological Perspective on Leading Enterprises
    (2023) Wang, Yi

    AbstractBackground: The vibrant state of the biopharmaceutical industry in Jiangsu Province, China, showcasing its strategic focus and significant growth. The province has implemented various policies to accelerate high-quality development, emphasizing innovation, intelligent and digital transformation. The industry, organized around biological drugs, chemical drugs, traditional Chinese medicine, and medical equipment, has become a national leader, with impressive revenue and output values. The study identifies research gaps related to the role of patents in the industry, highlighting the need for a micro-level analysis, better linkage between patents and economic outcomes, exploration of innovation capabilities, consideration of patent quality, and understanding temporal and geographical specificities. The research aims to address these gaps by assessing the impact of patent indicators at national, provincial, and enterprise levels, providing insights for policymakers and industry stakeholders to foster innovation and growth in the biomedical sector in Jiangsu Province. Methods: This research employs a comprehensive three-tiered analysis to explore the intricate interplay between innovative indicators and economic output in the biomedical industry. The study spans the national, provincial, and enterprise levels, encompassing five diverse countries, all 31 provinces in mainland China, and the top ten pharmaceutical companies globally, domestically, and in Jiangsu Province. In collaboration with the Jiangsu Intellectual Property Protection Center, the study compiles an extensive dataset. This dataset, meticulously curated from publicly available and non-sensitive patent data, includes variables such as patent applications, authorizations, PCT applications, and application growth rates. Economic data, including GDP and market values, is sourced from reputable institutions like the World Bank, the National Bureau of Statistics of China, and Torreya, a renowned biomedical industry innovation consulting company. Inclusive approach ensures a comprehensive representation for a cohort of 30 biomedical enterprises. The study extracts pertinent patent information into STATA for detailed analysis. The economic outcomes are measured through GDP at the national and provincial levels and market value at the enterprise level. The study meticulously examines the nuances of GDP computation and derives the 2022 national GDP data from the World Bank and provincial GDP data from the Bureau of Statistics of the People's Republic of China. Market value, or Market Capitalization, is defined and measured based on the total value of shares issued by listed companies, mainly in US dollars, with data sourced from Torreya, China Securities Regulatory Commission, and the Information Registration Center of the State Administration for Market Regulation. The biomedical patent indicators, comprising patent application numbers, PCT application numbers, patent authorized numbers, and patent in-force numbers, are systematically analyzed at national, provincial, and enterprise levels. Statistical methods, including one-way analysis of variance (ANOVA), correlation analysis, and linear regression analysis, are consistently applied across all levels to scrutinize the association between patent data indicators and economic output. The combined approach of correlation and regression analysis aims to derive precise patent innovation indicators and their impact on economic output, enhancing the clarity and intuitiveness of the research findings. Results: On the global stage, the findings reveal a robust growth in biomedical patents from 2000 to 2022. China emerges as a major player, exhibiting substantial patent applications (1804099 pcs), authorizations (35131 pcs), and PCT patents (1115204 pcs). Correlation analyses indicate positive relationships between a country's GDP and patent-related metrics, further validated by linear regression. These results emphasize the importance of economic strength in fostering biomedical innovation. Turning our attention to Chinese provinces, our investigation spans 31 regions, unraveling diverse patterns in patent outputs. Provinces with higher GDP consistently demonstrate elevated patent activity, showcasing a symbiotic relationship between economic prosperity and biomedical innovation. Correlation and regression analyses reaffirm these findings, emphasizing the quantitative impact of economic factors on biomedical research and development. At the micro-level, the study scrutinizes the top 10 pharmaceutical companies globally, in China, and within Jiangsu province. Striking disparities in patent application numbers, patents in force, and PCT applications highlight China's accelerating innovation capabilities. The correlation analysis establishes positive associations between market value and key patent indices (Coef. for number of patent application, patent in force, and PCT application, respectively 0.74, 0.71, 0.73,) emphasizing the pivotal role of patents in driving economic success for pharmaceutical companies. Linear regression analyses provide nuanced insights into the impact of patent indices on market value. The positive coefficients for patents in force (Coeff =9.77; 95% CI, 8.57 ~ 10.97) and PCT applications (Coeff=21.13; 95% CI, 18.68 ~ 23.58) signify their significant contribution to market values. The accompanying scatterplots visually reinforce these relationships, illustrating a positive linear correlation. Conclusions: Particularly, conducting a micro-level analysis, this research reveals individual pharmaceutical companies' contributions often overlooked in macro-level studies. The nuanced exploration of the relationship between patent indicators and economic outcomes emphasizes the positive correlation between regional economic levels and patent indicators, particularly in the pharmaceutical industry. At the enterprise level, the research establishes a connection between patent numbers and market competitiveness, with more patents associated with stronger market positions. Comparisons with other countries and Chinese provinces highlight strategic gaps and position Jiangsu as a leader in innovation capabilities. The study underscores the critical importance of patent quality over quantity, advocating for strategies that prioritize high-quality patents to drive sustained growth and competitiveness in the biomedical industry.

  • ItemOpen Access
    The Ethics of AI & Appropriation of Antiquity
    (2023) Childers, Caitlin Anessa

    In 2017 ‘Unite the Right’ rallied in Charlottesville. Then in 2018 Identity Evropa placed fliers on college campuses across America. In 2021 rioters stormed Capitol Hill. The uniting theme, apart from racist claims of white superiority, was the imagery of ancient Greece and Rome. The use of symbols, artwork, and quotations from classical antiquity. There were Roman legion flags in Charlottesville, Greek and Roman statues on the fliers, and Greek helmets at the Capitol. This public pattern of the alt-right misuse of this ancient imagery begs the question; what behavior does the American public now associate with antiquity?I’ve created an AI model for a museum exhibition to navigate this debate from the perspective of Roman historian, Cassius Dio. Dio’s obscurity, brash tone, and extensive bibliography make him an optimal model candidate. Creating this model involves considering the ethical implications of recreating a person without their permission or input, the misinterpretation of such figures by academia, and the role of data mining and its use. I close with a discussion on how the agency of such a model can impact public perception of a figure’s beliefs, and how this could be used to solidify opinion as fact in the public eye.

  • ItemEmbargo
    Site (Trans)Formation and Decolonial Praxis in Cuban Civic Art: Exploring Digital and Analog Approaches
    (2023) Fitzpatrick, Savannah

    Life in Cuba is largely defined by el Partido Comunista de Cuba’s (PCC) tradition of governance. Since the ratification of Decree 349 in 2018 – a law that punitively curtails freedom of expression – Cuba has witnessed an upsurge in publicly staged resistance. The emergence of several artist-led, non-partisan civic groups, united by their fight for human rights, exemplifies this. Two prominent examples are el Movimiento San Isidro (MSI) and 27N. This thesis investigates how the artistic interventions of MSI, 27N, and their members can be understood as decolonial praxis. To navigate and convey this argument and its associated logics, this thesis employs a two-part methodological approach: exploratory mapping in digital and analog forms, as well as critical feminist and queer phenomenological analysis that is woven with Doreen Massey’s relational spatial theory.

  • ItemEmbargo
    Trust in Information Sources as a Predictor of Parents’ Knowledge, Attitudes, and Practices (KAP) Regarding PCV13 Vaccination for Children in China: A Mixed Methods Study
    (2023) Pan, Zhangyang

    Background: Pneumococcal vaccines could protect children from being infected by some of the deadly Streptococcus pneumonia bacteria. However, since the pneumococcal vaccines have not yet been listed in China’s National Immunization Program (NIP), their coverage remains low due to the non-mandate and self-paid features. Parents are usually the key decision makers regarding their children’s PCV13 vaccine taking. Their exposure to vaccine information from trusted sources might influence this decision-making process. This study aims to explore the association between trust in information sources and parents’ knowledge, attitudes, and practices regarding PCV13 vaccination for children in the Yangtze River Delta region, China.Methods: This study applied an explanatory sequential mixed-method design, combining a web-based quantitative survey (N = 1,871) and qualitative interviews (N =19). Adult parents who had at least one child aged less than 24 months (including 24 months) and lived in the Yangtze River Delta region were recruited. A Five-Point Likert scale measured the level of trust in different sources. Adjusted Ordinary Least Square (OLS) regression model was applied to estimate the association between participants’ level of trust in different information sources and the primary outcomes. Semi-structured individual interviews were conducted. A thematic analytic approach was adopted to analyze the transcripts. Results: CDC received the highest trust score (4.48/5). High levels of uncertainty were identified in parents’ PCV13 knowledge. Age, gender, education, and annual household income were related to varied trust levels in specific sources. After controlling the effect of covariates, trust in healthcare institution sources was significantly associated with better PCV13 knowledge, less vaccine hesitancy, and a higher likelihood of vaccine uptake. Trust in official media was negatively associated with vaccine uptake. Trust in online community sources was positively associated with vaccine uptake. The qualitative study also identified parents’ trust in healthcare-related sources. Parents would consult multiple sources for vaccine information. Insufficient information provision during the vaccination process was one of the key drivers of vaccine information seeking. Parents had varied informed levels about PCV13 when making vaccination decisions. Conclusion: Healthcare institution sources were highly trusted and crucial in enhancing parents’ vaccine knowledge and acceptance of PCV13. The influence of online community sources on vaccine uptake is worth more attention. Public health policymakers could utilize media platforms with high engagement among parents for future optional vaccine promotion. Public health workers could consider embedding vaccine information provision into the pediatric vaccination process to inform parents’ decision-making better.

  • ItemOpen Access
    The Duality of Max Frei: Problems of Identity in Contemporary Popular Russian Fiction
    (2023) Garman-Davis, Gillian Ann

    Max Frei, whose given name is Svetlana Martynchik, is an important cultural phenomenon in the Russian sphere because she represents a growing push for change. Her approach is to question her reader’s identity by presenting challenges to the current reader-character relationship, as well as through her unique approach to the author-character relationship. In doing so, she questions the current definition of character and uses her creative works to demonstrate what she believes is most important: giving her characters life. For this work, I have studied ten of Frei/Martynchik’s interviews, available on her website, as well as Russian and English fandom websites such as fantlab.ru and Goodreads. Included in this work are four of her short stories, translated by me and available in Russian in the orange Russian Foreign Tales series, Books One, Four, and Six, to demonstrate her approach to the problems of identity in contemporary Russian fiction.

  • ItemEmbargo
    Copper-Catalyzed 1,3-Aminocyclization of Cyclopropanes as A Rapid Entry to γ-Amino Heterocycles
    (2023) Nguyen, Andrew

    Heterocycles represent an important class of motifs found in many bioactive molecules, pharmaceuticals, and agrochemicals. The ability to rapidly construct a diverse set of these compounds remains an important endeavor in the field of synthetic chemistry. In this thesis, an intramolecular 1,3-difunctionalization of cyclopropanes is reported using a copper-NFSI catalyzed system. Direct oxidation of the substrate by a nitrogen centered radical activates the cyclopropane which then undergoes a ring-opening cascade to produce a variety of lactones, cyclic ethers, pyrrolidines, and oximes containing γ-amino functionalization. Further product derivatization can produce various protected amines and alkyl-sulfonamides.

  • ItemEmbargo
    Exploring Antimicrobial Resistance in Extended- Spectrum Β-Lactamase Producing Klebsiella pneumoniae Isolates from Chicken in Kunshan, China and Nairobi, Kenya.
    (2023) KUVE, SOPHIA ATINGO

    ABSTRACTThe World Health Organization (WHO) has declared antimicrobial resistance (AMR) a global health threat. New AMR resistance strains of bacteria have emerged which are associated with high morbidity and mortality rates. AMR has been regarded as a One-Health issue due to its growing concerns for livestock, poultry, and public health. Klebsiella pneumoniae (K.pneumoniae) is a prominent zoonotic bacterium and high prevalence of Extended Spectrum Beta Lactamases K. pneumoniae (ESBL-KP), carbapenem resistant K. pneumoniae (CR-KP) and colistin-resistant K. pneumoniae (COLR-KP) have been reported worldwide. Due to excessive and indiscriminate use of antibiotics in the poultry industry, its Multidrug Resistant (MDR) has increased in recent years. Despite the threat it possesses to the public, limited data exist about its prevalence, and resistance towards first line antimicrobial agents in China and Kenya. The main objective of this study was to investigate the prevalence and antibiotic resistance of K. pneumoniae isolated from chicken in China and Kenya. This cross-sectional study was approved by the Institutional Animal Care & Use Committee (IACUC), Duke Kunshan University, China, and the Institute of Primates Research (IPR) Kenya. A total of 385 cloacal swabs samples (193 from Kunshan and 192 from Nairobi) were collected from live chicken by using sterile cotton swabs during 2022-2023. K. pneumoniae isolates were identified and confirmed by using culture and Polymerase Chain Reaction (PCR) assays while sensitivity testing was done to determine the susceptibility of these isolates to selected antimicrobial agents. The presence of resistant genes was determined by PCR by using standard primers. The Klebsiella species isolation rate was 52.8% (102/193) and 48.4% (93/192) identified by culture but 46.1% (89/193) and 35.9% (69/192) confirmed as K. pneumoniae isolates in Kunshan and Nairobi respectively. There was no significant difference (p>0.05) in the prevalence rate of K. pneumoniae in Kunshan and Nairobi. All the isolates except one were resistant to at least one antimicrobial agent. Resistance towards antimicrobial agents was reported in more than half the samples with high resistance being noted towards Ampicillin (AMP 81.7%), Ciprofloxacin (CIP 60.8%) and Aztreonam (ATM 48.7%). Number of resistant isolates against AMP, CIP, and ATM in Kunshan was significantly higher (p<0.05) than Nairobi. There was significant difference (p<0.05) in the number of sensitive, resistance and intermediate isolates. Antimicrobial sensitivity of ATM, IMP was significantly higher (p<0.05) than antimicrobial sensitivity of AMP, CIP and CTX. MDR isolates of K. pneumoniae observed in this study, however, the pattern of MDR was variable between Kunshan and Nairobi. BlaOXA-1 was the predominant resistant gene, however blaCTX-M, blaNDM-1 were significant. The prevalence of blaOXA-1 among K. pneumoniae isolates was significantly higher (p<0.05) when compared with other resistant genes. A non-significant difference (P>0.05) was observed among ESB-KB, CR-KP and COLR-KP. The prevalence of resistant genes was significantly higher (p<0.05) in Nairobi isolates when compared with isolates from Kunshan. Higher prevalence of K. pneumoniae in Kunshan was attributed to high prevalence of K. pneumoniae, high consumption of poultry products and excess use of antimicrobial agents in poultry. Likewise, prevalence in Nairobi was associated with lack of consultation services and excessive use of antimicrobial agents in both human and companion animals. Findings highlighted the potential role of chicken as a reservoir of K. pneumoniae hence a potential threat to food safety and public health, subsequently to the World Health Organization (WHO) vision 2030 of sustainable health.

  • ItemEmbargo
    Health Insurance and Farm Labor Supply: Evidence from China's Urban and Rural Residents' Basic Medical Insurance Reform
    (2023) Liu, Jie

    Background: In 2017, China launched the Urban and Rural Residents' Basic Medical Insurance program, with the primary goal of improving insurance benefits, particularly for those living in rural areas, and reducing rural-urban disparities. This study examined the impact of China’s Urban and Rural Residents' Basic Medical Insurance on the retirement decisions and labor market participation of individuals in rural areas aged 45 years and older. I hypothesized that the health insurance reform may increase the likelihood of retirement and reduce weekly working hours by alleviating their concerns about potential catastrophic medical expenditures in the future and motivating them to allocate more time towards managing their health conditions.Methods: To test this hypothesis, I used the nationally representative data from the China Health and Retirement Longitudinal Study, which comprises data from the waves conducted in 2011, 2013, 2015, and 2018. The final sample consisted of 6,536 rural participants aged 45 years and older who had complete data for outcome and control variables. The outcome variables examined in this study were retirement status, a binary variable that takes 1 if the participant is retired, and working hours per week, a continuous variable. This study controlled for individual-level variables including age, gender, marital status, education level, income, number of living children, number of living sons, and self-reported health, and city-level variables including GDP per capita and unemployment rate. To address potential endogeneity, this study employed the instrumental variable method, using whether the program was implemented before a given year as an instrument. Specifically, the Probit model with the IV as an independent variable was first implemented to estimate the probabilities of enrolling in the URRBMI. Subsequently, the fitted enrollment probability was included as an instrument for the endogenous variable along with exogenous covariates in a Two-stage Least-squares Regression. Results: In the final sample, 26.24% of the participants were in a retired status. The mean weekly working hours stood at 32.03 hours. The IV regression results indicated that the URRBMI increased the likelihood of retirement by 22.5% and led to a reduction of 13.51 hours in the working hours after adjusting for individual-level control variables and city-level control variables. The decrease in working hours was primarily driven by a decrease in labor supply among those who continue to work. Through heterogeneous analysis, I found that this behavioral shift could be primarily attributed to the reform's incentivization of individuals to allocate a greater portion of their time to healthcare management and its role in mitigating the probability of catastrophic health expenditure. Furthermore, I found suggestive evidence that after this reform individuals exhibited a 54% increased likelihood of engaging in self-employment and a 66% decreased likelihood of working off-farm. Conclusions: This study provided evidence that among those aged 45 and above in rural settings, the implementation of the URRBMI motivated them to retire earlier and reduced their working hours per week. Additionally, the URRBMI enhanced the likelihood of agricultural employment and self-employment, which is instrumental in curbing rural depopulation and fostering entrepreneurship within society.

  • ItemEmbargo
    "It's his cheerfulness that gives me hope:" A Qualitative Analysis of Access to Pediatric Cancer Care in Northern Tanzania
    (2023) Metcalf, Madeline

    Background:Pediatric cancer is a significant and growing burden in low- and middle-income countries such as Tanzania. The objective of this project was to describe the barriers and facilitators to accessing pediatric cancer care in Northern Tanzania using the Three Delays Model

    Methods:This was a cross-sectional qualitative study conducted between June and September 2023 at Kilimanjaro Christian Medical Center (KCMC). Parents and caregivers of children obtaining pediatric cancer care at KCMC were approached for participation in in-depth interviews (IDIs) and complimentary demographic surveys. This study is part of an ongoing collaboration between researchers from Duke Global Health Institute at Duke University and KCMC.

    Results:Participants suggested significant financial barriers to accessing pediatric cancer care along the entire care continuum. Early delays were impacted by waiting for symptoms to resolve and the use of traditional medicine. Additional delays resulted from health infrastructure at mid-level health facilities, misdiagnoses, and delayed referral to KCMC for definitive treatment. Participants did not describe clinical delays after arrival to KCMC and rather offered perspective on their child’s cancer diagnosis, their concerns while obtaining care, and their hopes for the future.

    Conclusions: The most significant clinical delays occurred within Delay 1 (the decision to seek care) and Delay 2 (reaching a medical facility). Advancements in clinical infrastructure and training at intermediary medical centers should be considered to support the prompt diagnosis and treatment for pediatric cancer. Lastly, financial and social schemes which limit out-of-pocket expenses should be considered.

  • ItemEmbargo
    Predicting All-cause Mortality among Chinese Community-Dwelling Elderly
    (2020) Jin, Xurui

    Background and aim: This study aimed at building the prediction model of all-cause mortality among Chinese dwelling elderly with different methods including regression models and machine learning models and to compare the performance of machine learning models with regression models on predicting mortality. Additionally, this study also aimed at ranking the predictors of mortality within different models and comparing the predictive value of different groups of predictors using the model with the best performance.Method: I used data from the Healthy Ageing and Biomarkers Cohort Study, a sub-study of the Chinese Longitudinal Healthy Longevity Survey (CLHLS). The baseline survey was conducted in 2008 and participants were followed every 2-3 years till 2018. The analysis sample included 2,448 participants. I used totally 117 predictors to build the prediction model, including 65 questionnaires, 39 biomarkers, and 15 genetics predictors. Four models were built (XG-Boost, random survival forest [RSF], Cox regression with all variables, and Cox-backward). I used C-index and integrated Brier score to evaluate the performance of those four models. Results: The XG-Boost model and RSF model shows slightly better predictive performance than Cox models and Cox-backward models based on the C-index and integrated Brier score. Age. The activity of daily living and Mini-Mental State Examination score were identified as the top 3 predictors in the XG-Boost and RSF models. Biomarker and questionnaire predictors have a similar predictive value, while genetic predictors have no addictive predictive value when combined with questionnaire or biomarker predictors. Conclusion: In this work, it is shown that machine learning techniques can be a useful tool for both prediction and its performance sightly outperformed the regression model in predicting survival.

  • ItemOpen Access
    Utility and Evaluation of MeTree on Family Health History Collection in Sri Lanka
    (2019) Hu, Ruoyu

    Abstract

    Introduction: Information about family history of illness is increasingly important to ensure each patient receive optimal promotion advice, preventive health services and appropriate treatment. MeTree is a web-based tool used to collect family health history directly from participants through a website. Despite growing evidence regarding the importance and efficacy of using family history, and the need for healthcare providers to have family history triage tools for personalized healthcare delivery, tools like MeTree have not been broadly applied in clinical practice. Little is known about its utility in settings with different disease profiles, health care systems and traditions and different cultural and socioeconomic contexts.

    Methods: This study enrolled 304 medical students from the University of Ruhuna in Sri Lanka. Participants constructed family pedigree and entered family health history in MeTree with the help of the researcher and a local research assistant. Once participants had created a full family pedigree, they were asked to complete a paper-based questionnaire asking about their experience with MeTree and what perceived to be the benefits of MeTree.

    Results: The proportion of males and females were relatively similar, 52.6% and 47.4%, respectively. Family health histories were entered for 3352 relatives in total. All diseases were divided into 21 groups. Diabetes was the most common disease group reported and accounted for 24.58% of all diseases reported. The following two most frequent diseases were hypertension (14.51%) and cardiovascular diseases

    (12.06%). Relatives had much higher numbers of all disease groups compared to index participants. Prevalence for each disease was different between index participants and the relatives. Hypertension was present in at least one or more family members in 65.13% of all pedigrees. On average, 29.92% of relatives in each family have diabetes. The mean time to complete entering information into MeTree was 36.3 minutes. The overall feelings and satisfaction level towards MeTree were favorable. Over 90% of participants indicated that MeTree could be generalized in the context of Sri Lanka. Results from t-test at the significance level of 5% didn’t indicate any significant preference for completion time used by males (M = 35.51, SD = 11.47) over completion time used by females (M = 37.04, SD =11.04), t = 0.65, p = 0.5177. A significant effect of grade on completion time wasn’t observed at the p< 0.05 level in the ANOVA procedure either, for the three conditions F (3, 83) = 1.80, p = 0.1539. According to the correlation procedure, completion time and age were not correlated (r = 0.2129, p = 0.13182). Among all questions in the questionnaire, significant difference was only observed between males and females for reported awareness (p=0.0184) and knowledge (p<0.0001) change after being introduced to MeTree.

    Conclusions: Most medical students found it easy to use MeTree and considered it a useful experience. The majority of the students thought it possible to generalize MeTree in the context of Sri Lanka, while barriers still needs to be overcome to have a web-based tool like MeTree put into real practice in Sri Lanka.

  • ItemEmbargo
    Analyzing Patient and Provider Perceived Delays to Snakebite Envenoming Care in Amazonas State, Brazil: A qualitative assessment using the Three Delay Model
    (2023) Mackey, Chandra Duanae

    Background: This thesis describes the perceived delays faced by patients and providers in Amazonas, Brazil while seeking or providing snakebite envenoming care in the region. Additionally, we compare the delays described by indigenous and non-indigenous patients while seeking care. Methods: This study analyzed previously recorded data from snakebite envenoming (SBE) patients and indigenous health care providers with intent to discuss their experiences. In-depth interviews and focus group transcripts were analyzed using the Three Delay Model. This model groups data into three separate delays; deciding to seek care, arriving to care, and receiving adequate care. Results: From this analysis we found that patients described many different themes for their decisions to seek care including choosing alternative methods, cultural restrictions, refusing traditional medicines and many more. The reasons behind the decisions to seek care or not were different between indigenous and non-indigenous patients. While travelling to care both study groups described the need to use multiple means of travel to arrive to distribution hospitals and the unavailability of transportation and emergency services which caused delays. Finally, once they arrived to health care facility delays were again presented by the need for multiple facilities to receive adequate care. Conclusion: The findings of this study support public health researchers push for the decentralization of antivenom. Both survivors of SBE and health care providers in the region have expressed the need for treatment to be available in their region. Due to the population of the region, any interventions whether education or political will need to consider the culture of the indigenous people to ensure positive uptake.

  • ItemOpen Access
    VR Touch Museum
    (2018) Zhao, Yuchen

    In recent years, digital technology has become ubiquitous in the museum. They have changed the ways museums document, preserve and present cultural heritage. Now, we are exploring if there are some ways that could provide more historical context to a displayed object and make an exhibition more immersive. Therefore, we did a project called “The Virtual Reality Touch Museum” and used an experiment to test if such museum performs better on “Presence” and learning achievements. As the results show, our VR Touch Museum was outstanding in “presence” but more research is necessary to verify how effective it is for learning.

  • ItemEmbargo
    Elevated Serum Chymase as a Risk Factor for Severe Dengue
    (2016) Farouk, Farouk Shihab

    Dengue Virus (DENV) is one of the major viral diseases that has a high burden in Southeast Asia and the Americas. Despite advances in supportive care for mild Dengue Fever (DF) and the more severe Dengue Hemorrhagic Fever (DHF), little is known about how to identify the progression to DHF before it occurs. The purpose of our retrospective study is to assess whether there is an association between the concentration of a serum protein, chymase, with DHF (and the symptoms, pre-existing conditions, and comorbidities associated with it).

    Data from 291 dengue-confirmed patients were collected from a surveillance study conducted in Sri Lanka. We selected a series of demographic, symptom, and pre-existing condition variables to see if there was an association between these, elevated chymase levels, and a DHF diagnosis. Our results confirmed that the correlation between a DHF diagnosis and increased chymase levels was statistically significant (p = <0.001). Furthermore, the negative correlation between a DF diagnosis and increased chymase levels was significant (p = <0.001). Obesity was also significantly predictive of increased chymase levels (p = <0.001). Elevated chymase levels correlated with (and predicted) a DHF diagnosis between ages 9 and 44, with those under age 9 having the strongest correlation.

    Overall, an increased chymase level was associated with a DHF diagnosis. The identification of chymase as a biomarker for severe dengue may lead to improved diagnostic and surveillance systems that can identify and treat patients at risk for developing DHF.

  • ItemEmbargo
    Digital Hydraulics Simulation in Mathematica on Sudden Expansion Flows
    (2023) Frechette, August

    In this work, we offer readers the ability to numerically simulate flow through a sudden expansion themselves. We choose to study the sudden expansion due to its prevalence in engineered and natural water distribution networks (i.e., pipes and rivers, respectively). The simulation is written in the Wolfram Language, also known as Mathematica. The symbolic nature of this programming language enables readers to implement physical theory directly, resulting in a highly readable numerical flow solver; a stark contrast with commonplace commercial flow solvers, which operate like “black box” technologies, and low-level programming languages, which require an advanced level of syntax knowledge and programming proficiency. Upon completion of this laboratory exercise, users should be able to: (i) describe the main principles underpinning the numerical simulation of non-linear models, (ii) apply numerical models to investigate the accuracy of simplified analytical models, (iii) demonstrate a beginner-level understanding of Mathematica and, more broadly, symbolic coding environments, (ii) and most generally, (iv) understand the proper context for physical and numerical experimentation. The novelty of this work is attributed to the fact that no such simulation tool is detailed and provided in the literature for readers to utilize and alter at their discretion.

    This work was developed and undertaken in collaboration with my co-authors, Dr. Anil Ganti (A.G.), and Dr. Zbigniew Kabala (Z.J.K), my master’s advisor. Author contributions are as follows: conceptualization, Z.J.K.; methodology, A.H.F, A.G. and Z.J.K.; software, A.H.F and A.G.; validation, A.H.F, A.G. and Z.J.K.; formal analysis, A.H.F; investigation, A.H.F, A.G. and Z.J.K.; resources, Z.J.K; data curation, A.H.F, A.G. and Z.J.K.; writing—original draft preparation, A.H.F and Z.J.K.; writing—review and editing, A.H.F, A.G. and Z.J.K.; visualization, A.H.F.; supervision, Z.J.K.; project administration, A.H.F and Z.J.K.

    Partial funding for this project has been received from Duke University Undergraduate Program Enhancement Fund (UPEF) grant 399-000226.