Biostatistics Commons^™

A Comparative Evaluation Of Data Imbalance Handling Techniques In Machine Learning Models For One-Year Mortality Prediction In Liver Cirrhosis, Sumiya Hasan Trisha

UNF Graduate Theses and Dissertations

Liver cirrhosis is associated with substantial morbidity and mortality, making one-year mortality prediction a clinically relevant problem. Using a liver cirrhosis dataset as the motivating application, this thesis evaluates five machine learning classifiers—Logistic Regression, Random Forest, XGBoost, LightGBM, and CatBoost—under five class-imbalance handling strategies: Baseline learning, Random Oversampling, SMOTE-NC, ADASYN, and Cost-Sensitive Learning. Hyperparameter tuning was conducted using randomized search, and predictive performance was assessed over 200 iterations of Monte Carlo Cross-Validation using Accuracy, Precision, Recall, Fl-score, and ROC-AUC.

The results suggest that imbalance-handling strategies can materially affect predictive performance, particularly recall. Because the outcome of interest is death within …

Ridit-Based Adaptive Allocation In Two-Stage Clinical Trials With Binary Outcomes, Dewan Fahim

UNF Graduate Theses and Dissertations

This study explores better ways to assign patients to treatments in clinical trials with binary outcomes, such as success or failure. Adaptive methods are used to learn from early results and adjust treatment assignments during the trial, helping more patients receive better performing treatments while maintaining reliable conclusions. We focus on trials comparing multiple treatments using a two-stage design. In the first stage, several treatments are tested to identify the most promising one; in the second stage, that treatment is compared with a control. Unlike traditional equal assignment, we use adaptive allocation in the second stage to make better use …

A Comparative Study Of Classification Methods For Healthcare Analytics, Xueting Zhao

UNF Graduate Theses and Dissertations

This thesis presents a comparative study of logistic regression, Linear Discriminant Analy- sis (LDA), and Quadratic Discriminant Analysis (QDA) for binary classification in healthcare analytics, integrating theoretical derivation, simulation, and real-data application. A facto- rial simulation study crosses the covariance structure (equal vs. unequal), predictor correla- tion (ρ ∈ {0, 0.5, 0.9}), dimensionality (p ∈ {2, 5, 10}) and sample size (n ∈ {50, 100, 200}) across 54 scenarios with 1,000 Monte Carlo replicates each. Three main findings emerge. Logistic regression and LDA are nearly interchangeable when the assumption of equal-covariance holds. QDA achieves substantially better discrimi- nation when class-specific …

Body Mass Index Has No Impact On Complications And Mortality For Patients With Stage Iv Pancreatic Ductal Adenocarcinoma, Ren Bryant, Hannah Darnell, Megan Hall, Kelsey N. Karnik, Kristen J. Mcquerry, Ruben R. Plentz

Biostatistics Faculty Publications

Background: Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of United States (USA) cancer death. Overweight and obesity developing into a growing global medical and socio-economic problem, affecting approximately 42% of adults in the USA population. The aim of our analysis was to evaluate the influence of overweight and obesity on complications and clinical outcome in patients with stage IV PDAC.

Methods: We retrospectively reviewed electronic health records of patients diagnosed with stage IV PDAC (n=162) who followed with the University of Kentucky from January 2017–October 2024. Comparisons were based on the body mass index (BMI): low BMI …

Iso-Detr: A Novel Detection Transformer For Industrial Small Object Detection, Faisal Saeed, Anand Paul

School of Public Health Faculty Publications

Effectively detecting and assessing real-time structural and ecological parameters in contemporary manufacturing environments poses significant challenges, particularly in identifying minute objects within product images. The swift evolution of the industrial sector underscores the necessity for intelligent manufacturing environments to uphold stringent product quality standards. However, accelerating production processes at high speeds heightens the risk of defective product outcomes. This research addresses the challenges inherent in small object detection within industrial contexts, proposing an innovative detection transformer model tailored to modern manufacturing environments. The proposed model integrates a feature-enhanced multi-head self-attention block (FEMSA), merging cross-channel communication network and multiple multi-head self-attention …

Generating Predictive Gene Expression Signatures For Alzheimer's Disease Using Postmortem Brain Tissue, Ashley Duche

Pharmaceutical Sciences (PhD) Dissertations

Background: Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder characterized by the accumulation of amyloid-beta (Aβ) plaques and tau protein aggregates. These pathological features develop in specific brain regions, but why some areas are more vulnerable to early AD-related changes remains unclear. To address this, predictive gene expression signatures were developed to explore the molecular mechanisms underlying regional susceptibility to AD pathology.

Methods: This was performed using postmortem brain (PMB) tissue from participants in the Religious Orders Study and Memory and Aging Project (ROSMAP), Mayo Clinic, and Mount Sinai Brain Bank (MSBB) to generate gene expression signatures from six brain …

Flexible Spatial Priors In Bayesian Neuroimaging: Gmrf, Nngp, And Deep Gmrf, Boyoung Hur

All Dissertations

Structural neuroimaging is essential for understanding neurological disorders such as Alzheimer’s disease, enabling accurate delineation of brain regions through image segmentation. Among various segmentation methods, multi-atlas-based approaches like label fusion have become leading techniques. In statistics, Bayesian hierarchical models for label fusion are increasingly favored for their ability to incorporate uncertainty and prior knowledge. Also, a key challenge in modeling neuroimaging data is spatial dependence among image voxels, making the choice of spatial prior critical—particularly in high-resolution settings where segmentation accuracy and computational efficiency are both essential.

This dissertation proposes fully Bayesian spatial hierarchical models that explore two flex- ible …

A Persistent Homology Framework For Scrna-Seq: Assessing Clustering Robustness And Quantifying Preprocessing And Integration Effects On Topological Features., Jonah Daneshmand

Electronic Theses and Dissertations

As single-cell RNA sequencing (scRNA-seq) data expands, robust methods for integrating diverse datasets are critical. This dissertation applies Persistent Homology (PH), a technique from Topological Data Analysis (TDA), to a collection of scRNA-seq datasets spanning eight tissue types to quantify how data integration affects topological features and biological interpretability. We assessed global topological structure using Betti curves, Euler characteristics, and persistence landscapes across raw, normalized, and integrated data representations. Our analysis revealed a performance inversion: while conventional methods excelled on unintegrated data, high-granularity topological methods, particularly those sensitive to global data structure, became superior after integration. This suggests a synergy …

Taxonomy Of Endophytic Fungi Associated With Vallisneria Neotropicalis, Md. Arafat Rashid

Graduate Theses and Dissertations (2019 - present)

This study presents the first comprehensive taxonomic and ecological investigation of endophytic fungi (EF) associated with the submerged aquatic macrophyte Vallisneria neotropicalis in the southern United States. Over a 12 month period, from April 2023 to March 2024, leaf samples were collected from two distinct sites in Mobile Bay, Alabama a disturbed, brackish Causeway location and a cleaner, less impacted site at Meaher State Park. Using culture dependent methods, a total of 257 fungal endophytes were isolated from 1,200 leaf segments. All isolates belonged to the phylum Ascomycota, distributed across 3 classes, 6 orders, 10 families, and 19 taxa. The …

Short-Term Preeclampsia Prediction: Cutoff Variations For Sflt-1/Plgf In U.S. Patients With Or Without Hypertensive Disorders, Yaxin Li, Kristen Cagino, Jim Yee, Caroline Andy, Dajana Borova, Ayush Shah, Isla Racine, Tracy Grossman, Zhen Zhao

Student Papers, Posters & Projects

BACKGROUND: Preeclampsia (PE) is a complex disorder with significant maternal and fetal risks. The soluble fms-like tyrosine kinase-1 (sFlt-1) and placental growth factor (PlGF) ratio shows promise as a diagnostic tool, but its adoption in the U.S. remains limited due to the lack of accessible testing platforms, U.S.-based studies, and evidence-based implementation guidelines.

PATIENTS/MATERIALS AND METHODS: We conducted a cohort study to evaluate the sFlt-1/PlGF ratio for predicting PE within two weeks among pregnant individuals ≥18 years, ≥20 weeks gestation. Serum samples were obtained from routine prenatal visits or triage evaluations. sFlt-1/PlGF ratios were measured using Roche Elecsys assays, and …

Improving Glycemic Control Among Indonesian Urban Adults: A Digital And Behavioral Extension Of The Information–Motivation–Behavioral Skills Model, Imelda Appulembang

Kesmas

Management of type 2 diabetes mellitus (T2DM) in Indonesia continues to face challenges due to behavioral, informational, and technological gaps among patients. This study analyzed the influence of self-regulatory competence and information, motivation, family support, and digital health literacy on glycemic control behavior. A cross-sectional survey was conducted from February to April 2025 among 587 adults aged >30 years with T2DM enrolled in the Chronic Disease Management Program at primary health care in six major cities: Jakarta, Surabaya, Yogyakarta, Medan, Makassar, and Banjarmasin. Data were collected through structured questionnaires and analyzed using partial least squares structural equation modeling. The findings …

Benchmarking Dna Foundation Models For Genomic And Genetic Tasks, Haonan Feng, Lang Wu, Bingxin Zhao, Chad Huff, Jianjun Zhang, Jia Wu, Lifeng Lin, Peng Wei, Chong Wu

School of Medicine Faculty Publications

The rapid evolution of DNA foundation models promises to revolutionize genomics, yet comprehensive evaluations are lacking. Here, we present a comprehensive, unbiased benchmark of five models (DNABERT-2, Nucleotide Transformer V2, HyenaDNA, Caduceus-Ph, and GROVER) across diverse genomic and genetic tasks including sequence classification, gene expression prediction, variant effect quantification, and topologically associating domain (TAD) region recognition, using zero-shot embeddings. Our analysis reveals that mean token embedding consistently and significantly improves sequence classification performance, outperforming other pooling strategies. Model performance varies among tasks and datasets; while general purpose DNA foundation models showed competitive performance in pathogenic variant identification, they were less …

Field Validation Of Multiple Species Distribution Models Shows Variation In Performance For Predicting Aedes Albopictus Distributions At The Invasion Edge, Anna V. Shattuck, Brandon D. Hollingsworth Ph.D., Jared Skrotzki, Scott R. Campbell, Christopher L. Romano, Courtney C. Murdock

Faculty Publications

Background

Climate and land use changes have resulted in range expansion of many species. In this shifting disease landscape, it is important to leverage tools that can predict the distributions of invading vectors to target surveillance and control efforts and identify at-risk populations. Species distribution models (SDMs) are used to predict ranges of invasive species; however, invasive species often violate assumptions of equilibrium and niche conservatism. Moreover, these studies are rarely validated using independent data.

Methods 

We use long-term surveillance data for Aedes albopictus, a highly invasive mosquito capable of transmitting several arboviruses, at its range edge to evaluate a …

Correlation With Car Density In Urban Environments And Its Influence On Chronic Obstructive Pulmonary Disease (Copd) Rates In The United States, Daniel Barreiro-Torres, Kedai Cheng

Annual Symposium on Biomathematics and Ecology Education and Research

No abstract provided.

Statistical Challenges And Simulation Results For Pilot Clinical Trials, Weiliang Cen

USF Tampa Graduate Theses and Dissertations

Background: The effect size estimated from a pilot trial is often an inaccurate reflection of the true effect size observed in a large trial, leading to either underestimation or overestimation. Published data suggest that effect sizes from large trials are typically smaller than those reported in their corresponding pilot trials. To address this discrepancy, conservative or discount adjustment methods are widely recommended to modify pilot trial effect sizes when calculating sample sizes, thereby maintaining adequate statistical power. This study aims to assess effect sizes from both pilot and large trials and to evaluate the performance of existing adjustment methods.

Methods: …

Circulating Levels Of Insulin-Like Growth Factor I (Igf-I) And Risk Ofmultiple Myeloma: An Observational And Mendelian Randomisation Study, Yolanda Benavente, Sara Hermosa, Nikos Papadimitriou, Alyssa I. Clay-Gilmour Ph.D., Elizabeth E. Brown, Jonathan N. Hofmann, Nathaniel Rothman, Qing Lan, Sonja I. Berndt, Demetrius Albanes, Mark Purdue, Mitchell J. Machiela, Stephen J. Chanock, Parveen Bhatti, Wendy Cozen, Aaron Norman, Susan L. Slager, James R. Cerhan, Vincent Rajkumar, Shaji J. Kumar, Et. Al.

Faculty Publications

Evidence for an association between insulin-like growth factors (IGF) and multiple myeloma (MM) is inconsistent. We examined total IGF-I concentrations and risk of MM by combining baseline serological data among UK Biobank participants (n = 444 187; 732 incident MM) with a two-sample Mendelian randomisation (MR) analysis using identified genetic variants associated with circulating total IGF-I and IGF-binding protein 3 (IGFBP-3) in the InterLymph consortium (2434 MM and their 2567 controls). Finally, additional lymphoid neoplasm (LN) subtypes were included for comparison with the main hypothesis. Circulating IGF-I level was positively associated with MM risk Hazard ratio-HR-per one standard deviation-SD-increase …

Adverse Childhood Experiences, Positive Childhood Experiences, And Digital Media Use Among Children And Adolescents In The United States, Elizabeth L. Crouch, Emma Boswell, Monique J. Brown Ph.D., Kevin J. Bennett

Publications

No abstract provided.

Smarter Disease Detection From Electronic Health Record Data: An End-To-End Ai-Augmented Pipeline For Computable Phenotyping, Dylan Owens

Statistical Science Theses and Dissertations

Electronic Health Records (EHR) contain a wealth of structured and unstructured patient data that can be leveraged for computable phenotyping, the process of algorithmically identifying patient cohorts with specific diseases or conditions. Traditional rule-based phenotyping approaches, while interpretable, often struggle with scalability, portability across institutions, and effective use of unstructured clinical narratives. Recent advances in large language models (LLMs) present new opportunities for synthesizing complex free-text information into concise, clinically meaningful representations. However, integrating LLMs into phenotyping workflows requires careful design to maintain transparency, interpretability, and measurable uncertainty—features essential for clinical adoption and downstream applications such as decision support.

We …

Spatiotemporal Modeling Of Maternal Mortality In South Carolina 2018-2023, Leah Wood, Ray Bai, Emily Mann

Senior Theses

Maternal death serves as a public health indicator due to fact that it is considered preventable with the availability of modern biomedicine, however, it persists broadly throughout the United States. Current literature outlines national trends in maternal mortality with complicating, preexisting conditions, and structural upstream factors often cited as being the largest contributors to increased risk. This study utilizes publicly available, county-level data for maternal death in addition to demographic and descriptive data in order to estimate maternal mortality rates in each of South Carolina’s 46 counties from 2018 to 2023. In order to address sparsity in the outcome variable …

Glucodensity Functional Profiles Outperform Traditional Continuous Glucose Monitoring Metrics, Marcos Matabuena, Rahul Ghosal Ph.D., Javier Enrique Aguilar, Ayya Keshet, Robert Wagner, Carmen Fernández Merino, Juan Sánchez Castro, Vadim Zipunnikov, Jukka-Pekka Onnela, Francisco Gude

Faculty Publications

Continuous glucose monitoring (CGM) data have revolutionized the management of type 1 diabetes, particularly when integrated with insulin pumps to mitigate clinical events such as hypoglycemia. Recently, there has been growing interest in utilizing CGM devices in clinical studies involving healthy and diabetic populations. However, efficiently exploiting the high temporal resolution of CGM profiles remains a significant challenge. Numerous indices—such as time–in–range metrics and glucose variability measures–have been proposed, but evidence suggests these metrics overlook critical aspects of dynamic glucose homeostasis. As an alternative method, this paper explores the clinical value of glucodensity metrics in capturing glucose dynamics—specifically the speed …

Heuristic Weight Initialization For Transfer Learning In Classification Problems, Musulmon Lolaev, Anand Paul, Jeonghong Kim

School of Public Health Faculty Publications

Transfer learning is the predominant method for adapting pre-trained models on another task to new domains while preserving their internal architectures and augmenting them with requisite layers in Deep Neural Network models. Training intricate pre-trained models on a sizable dataset requires significant resources to fine-tune hyperparameters carefully. Most existing initialization methods mainly focus on gradient flow-related problems, such as gradient vanishing or exploding, or other existing approaches that require extra models that do not consider our setting, which is more practical. To address these problems, we suggest employing gradient-free heuristic methods to initialize the weights of the final new-added fully …

Effect Of Vitamin D Supplementation During Pregnancy And Lactation On The Development Of Infants Born To Tanzanian Women Living With Hiv: A Secondary Analysis Of A Randomised Controlled Trial, Temiwunmi Shobanke, Alfa Muhihi, Nandita Perumal Phd, Nzovu Ulenga, Fadhlun M. Alwy Al-Beity, Christopher P. Duggan, Wafaie W. Fawzi, Karim P. Manji, Christopher R. Sudfeld

Faculty Publications

Background Infants born to pregnant women living with HIV (WLHIV) are at greater risk for morbidity and mortality and may also have poorer developmental outcomes as compared with infants who are not exposed to HIV. Nutrition interventions in pregnancy may affect developmental outcomes.

Objectives This study evaluated the effect of maternal vitamin D supplementation on infant development outcomes.

Design We conducted a secondary analysis of a randomised, triple-blind, placebo-controlled trial of maternal vitamin D supplementation from June 2015 to October 2019.

Setting Antenatal care clinics in Dar es Salaam, Tanzania.

Participants Pregnant WLHIV and their offspring.

Interventions Daily 3000 IU …

Bayesian Survival Analysis For High-Dimensional Compositional Data, Zhenying Ding

All ETDs from UAB

Survival analysis integrating microbiome and clinical data offers powerful insights into human health. The microbiome influences immunity, inflammation, and cancer outcomes. Combining microbial profiles with clinical factors like tumor stage, treatment, and demographics enhances understanding of how host-microbe interactions affect survival. To analyze these complex datasets, advanced statistical methods, including Bayesian models, penalized Cox proportional hazards models, and machine learning techniques are employed. These approaches can uncover novel biomarkers and therapeutic targets, leading to personalized treatment strategies that optimize patient survival. However, challenges arise due to the compositionality, high-dimensionality, and phylogenetic relation between taxa in microbiome data. In this dissertation, …

Bayesian Mediation Analyses For Zero-Inflated Data, Jinhong Cui

All ETDs from UAB

Mediation analysis is a fundamental tool in causal inference, enabling researchers to disentangle the pathways through which an exposure affects an outcome via intermediate variables (mediators). It has been widely applied in social sciences, psychology, and healthcare-related studies. Traditionally, mediation analysis has been conducted within the linear structural equation modeling (LSEM) framework. However, recent advances have shifted attention toward the counterfactual framework due to its flexibility in defining causal effects and its capacity to accommodate non-linear relationships, parametric and non-parametric models. Building on this foundation, methodological developments in mediation analysis have expanded to address more complex data structures, such as …

Access To Information On Toddler Family Development Program And Family Participation In Child Growth And Development, Dita Dhammayanti, Demsa Simbolon, Lissa Ervina, Yusran Fauzi

Kesmas

The comprehension of the Toddler Family Development (TFD) Program among families in Indonesia remains limited, likely due to insufficient access to information and low participation rates. Limited participation can negatively affect a family’s ability to support optimal child growth and development. This study examined the relationship between access to information on the TFD Program and family participation in child growth and development. Using secondary data from the 2019 Program Performance and Accountability Survey in Indonesia, the cross-sectional analysis included 21,497 respondents. The results revealed an association between access to information on the TFD Program and family participation in child growth …

Exploring Stunting In South Kalimantan Province Using R Programming-Based Data Visualization, Muhammad Hudaya, Wahyudin Nor, Mellani Yuliastina, Muhammad Nordiansyah

Kesmas

South Kalimantan Province continues to face the challenge of relatively high stunting prevalence despite being endowed with abundant coal resources that could serve as a source of funding for public health. Therefore, this study aimed to examine differences in stunting prevalence among cities, mining districts, and non-mining districts in South Kalimantan to raise stakeholder awareness of disparities across these regional types. This study was conducted between April and December 2024, using secondary data obtained from the Indonesian Ministry of Health and the South Kalimantan Provincial Government. R programming was used to process the data, generate visualizations, and perform analysis of …

Understanding User Needs In Health Crisis Risk Monitoring Information System Development: A Lesson From Tasikmalaya District, Indonesia, Arief Tarmansyah Iman, Hari Kusnanto, Ariani Arista Putri Pertiwi

Kesmas

A health crisis risk monitoring information system needs to be developed, especially during the pre-disaster phase; therefore, understanding the needs of prospective users is crucial. This study aimed to investigate the needs of potential users regarding the development of this system. This study employed a qualitative, exploratory approach to gather user needs from stakeholders through interviews (n = 7) and one focus group discussion (n = 12). The data were audio-recorded, transcribed verbatim, and then thematically analyzed using qualitative content analysis. The need for information was related to disaster preparedness and its preferred format. The system should be targeted, multiplatform, …

Low Economic Level And The Risk Of Overweight Among Indonesian Junior And Senior High School Students, Ariyanto Nugroho, Purwo Setiyo Nugroho, Amarin Yudhana, Sri Sunarti

Kesmas

Overweight and obesity among Indonesian adolescents have emerged as a pressing public health issue, reflecting global trends. This study examines the relationship between economic status and overweight prevalence among junior and senior high school students in Indonesia, using secondary data from the Global School-based Health Survey (GSHS). This study analyzed data from 9,977 students aged 11–18 years through a cross-sectional design and binary logistic regression, adjusting for dietary habits, physical activity, and sedentary behavior. Overall, 14.7% of students were overweight; the prevalence was notably higher among low-income students (27.4%) compared to high-income groups (14.2%). Students from lower economic backgrounds were …

Spatial Patterns In Electronic Health Record Data-Based Predictive Modeling: A Case Study Of Prenatal Care And Preeclampsia, Abigail Lewis

McKelvey School of Engineering Graduate Student Theses & Dissertations

The electronic health record (EHR) documents interactions between patients and healthcare systems and is widely used for secondary research on health outcomes due to broad adoption. While EHR data support efficient clinical informatics research, they are subject to biases stemming from healthcare seeking behavior, health system attributes, and individual or environmental characteristics. Place-based characteristics, in particular, strongly influence when and how patients access care and shape health outcomes. These biases in EHR data can affect predictive modeling pipelines which leverage them, contributing to variation in model performance across groups. In healthcare, where predictive models are increasingly used to guide care, …

Rna’S Symphony: Harmonizing Splice Junctions And Exon Counts For A Novel Approach To Differential Splicing Analysis, Jelard Aquino

UNLV Theses, Dissertations, Professional Papers, and Capstones

Alternative Splicing (AS) plays a critical role in transcriptome complexity and cell-type-specific gene regulation, yet its analysis remains methodologically fragmented, especially in the context of noisy and sparse single-cell RNA sequencing (scRNA-seq) data. This dissertation addresses key computational challenges in AS detection by evaluating existing tools, developing integrative frameworks, and proposing new strategies for improving analysis accuracy in both bulk and single-cell contexts. In chapter 1, I present a comprehensive literature review of computational tools designed for detecting and quantifying AS from bulk and scRNA-seq data. This review outlines major methodological paradigms, including exon-based and splice junction-based approaches, and evaluates …