Digital Commons Network^™

Developing Imitation Learning Policies By Exploring Diffusion Framework For Quadcopters, Likith Swamireddy

Theses and Dissertations

Imitation learning offers a promising approach in developing near-optimal onboard policies for nonlinear systems, by learning from computationally expensive trajectory planners during offline training. Existing State-of-the-art methods commonly learn deterministic mappings from observations to actions by training fast neural networks to mimic the expert demonstrations, which can limit their ability to capture the action distribution, when the policy is trained on multiple possible expert commands to reach the same target, and this may lead to unsafe scenarios in the obstacle environments. To address these challenges, this study explores Diffusion Models (DMs) as a feedback controller for quadcopters that learns the …

In-Process Large Language Model Requesting Through Knowledge Graphs For Interfacing With Manufacturing Digital Twins, Reiki Tanaya Karuka, Vinh Nguyen

Michigan Tech Publications

Digital twins (DTs) have potential to improve manufacturing processes by providing in-process monitoring and reducing trial-and-error while improving process prediction and performance. However, identifying anomalies or defects in a DT environment remains a very time intensive process that requires multidisciplinary expertise and significant manual effort in analyzing the system. Large language models (LLMs) offer potential capabilities to interpret system-level information and support user-driven querying, but their effectiveness is limited by hallucinations when reasoning over large and dynamic manufacturing datasets inherent to process monitoring DTs. Knowledge graphs (KGs) have emerged as efficient tools in the industrial domain to efficiently represent data …

The Future Of Nature-Based Recreation In Warming Tropical Cities, Perrine Hamel, Emma E. Ramsay, Shawnda A. Morrison, Su Li Heng, Winston T. L. Chow, Beatrice H. Ho, Pearl Min Sze Tan, Moshe Mandelmilch, Lancy Sim, Jason Kai Wei Lee

Research Collection College of Integrative Studies

Nature-based recreation promotes health in tropical cities but is increasingly threatened by rising heat. This review presents recent evidence on the issue of humid heat stress and outdoor recreation in tropical cities and outlines key adaptation strategies – addressing hazard, exposure, and vulnerability – to enable ‘heat-smart’ nature-based activities. Despite existing solutions, critical research gaps remain, especially in integrating social, physiological, and technological insights to better address humid heat stress in tropical urban environments.

A Systematic Review Of Audio Deepfake Detection Techniques For Digital Investigation, Mahra Alnaqbi, Richard Adeyemi Ikuesan

All Works

Deepfake technology has been driven by advanced machine learning and revolutionized multimedia creation by synthesizing hyper-realistic content. It includes images, videos, and audio. While its creative applications in entertainment and accessibility are significant, the technology also poses critical risks, especially in fraud, disinformation, and identity theft. Audio deepfakes are a subset of this phenomenon that replicate human voices with enhanced precision, mimicking tone, accent, and subtle vocal nuances. This has raised concerns in security-sensitive domains like voice authentication and forensic investigations. This systematic literature review (SLR) adopts PRISMA guidelines to explore the state-of-the-art in audio deepfake detection. It examines existing …

Image And Metadata-Driven Personality Inference For Career Recommendation: A Social Media-Based Ai Framework For Adolescents, Heba Ismail, Maryam Alhefeiti, Ashraf Khalil

All Works

This study presents a novel AI-based framework that leverages Instagram image and metadata analysis to infer Big Five personality traits and deliver personalized career recommendations for high school students in the UAE. Addressing the limitations of traditional recommender systems that rely on self-reported questionnaires or text, the proposed approach uses multimodal visual features—including profile metrics, HSV color patterns, semantic image labels, and texture analysis—to enable a non-intrusive, scalable personalization method. A pilot study involving data from 30 student accounts served as a proof of concept. Correlation analysis identified profile and HSV features as the most predictive, and four machine learning …

Fig-Gan: Fundus Image Generation Via Deep Learning Based Generative Adversarial Network For Amd Disease Diagnosis, Kailasa Thrishul, Ahed Abugabah, Amina Salhi, Manel Ayadi, Mohamed M. Sithik, D. Jayaprakash, A. Ahilan

All Works

Globally, age-related macular degeneration (AMD) remains a main cause of irreversible vision loss. Recently, deep learning models have primarily focused on classifying fundus images for early detection of AMD progression. However, existing models rarely address the generation of future progression-aware fundus images, particularly when complete real longitudinal follow-up scans are unavailable. This limitation makes it difficult to track retinal changes over time and highlights the need for generative models capable of producing realistic drusen-level structural variations. To address these issues, a novel deep learning-based FIG-GAN model is to generate synthetic future fundus images from baseline inputs. Multi-Attention U-Net (MAU-Net) is …

A Performance-Optimized V2v Task Offloading Framework For Real-Time Vehicular Communication, Tariq Qayyum, Asadullah Tariq, Ikbal Taleb, Mohamed Adel Serhani, Zouheir Trabelsi

All Works

As vehicular applications become increasingly complex, their computational demands often exceed the capabilities of individual vehicles. Vehicular Edge Computing (VEC) alleviates this limitation by enabling task delegation to nearby edge resources; however, high mobility, dynamic topology, and fluctuating vehicle density make real-time offloading decisions challenging. To address these issues, we propose a performance-optimized Vehicle-to-Vehicle (V2V) task offloading framework for dense and dynamic Vehicular Ad-hoc Networks (VANETs). The framework follows a two-stage design: (i) context-aware edge-node selection based on live topology capture via periodic beaconing, and (ii) cumulative score-based dynamic priority queuing at the selected edge node. The priority score jointly …

Smart Medical System Integrating Clinical Workflows For Robust Skin Cancer Detection Across Heterogeneous Pathologies, Ahed Abugabah, Prashant Kumar Shukla, Suchi Mishra, Abhishek Dwivedi

All Works

Skin cancer is among the most prevalent and life-threatening dermatological diseases worldwide, with melanoma responsible for a substantial proportion of skin cancer–related deaths due to delayed and unreliable diagnosis. Conventional clinical screening based on visual inspection and expert interpretation is inherently subjective and often affected by inter-observer variability, lesion heterogeneity, and imaging artifacts, highlighting the need for accurate and generalizable automated diagnostic systems. This study proposes a novel hybrid deep learning architecture for skin cancer classification that integrates an attention-guided autoencoder with a transformer-inspired global context modeling module, forming a unified and robust representation learning framework. The encoder–decoder structure is …

Temporally Rigorous And Traceable Predictive Maintenance Via Joint Labeler-Model Optimization, Maytha Al-Ali, Ahmad Alharbi

All Works

Predictive maintenance (PdM) is a critical enabler of intelligent asset management in Industry 4.0, yet many existing frameworks remain difficult to operationalize due to methodological fragmentation. Common limitations include sacrificing temporal realism and class granularity for computational expediency, decoupling labeling strategy design from model hyperparameter optimization, and insufficient support for reproducibility and deployment traceability; particularly in rare-failure regimes. To address these challenges, we propose a unified, end-to-end, and fully traceable PdM framework that jointly optimizes labeling and model parameters while enforcing strict temporal fidelity. The proposed pipeline co-optimizes the failure lookahead window () and LightGBM hyperparameters within a single Bayesian …

Computational Studies Of Triplet State Formation In Zinc Dipyrrin Complexes, Reuben Kwabla Adigbli

Electronic Theses and Dissertations

Zinc dipyrromethene complexes are promising earth-abundant photosensitizers due to their strong visible-light absorption and tunable excited states. To rationally design them for photocatalysis and photodynamic therapy, a molecular-level understanding of triplet-state formation is needed, but details of intersystem crossing remain unclear. To study the effect of π-extension, we synthesized an indole-substituted zinc dipyrrin complex: the dipyrromethane ligand was prepared from indole and mesitaldehyde, oxidized to the dipyrromethene, then coordinated with zinc. DFT and TD-DFT calculations determined ground and excited-state geometries and energies in different solvents, modeled absorption spectra, and visualized charge distributions. The results show how solvent polarity shifts energies …

An Assessment Of Accessibility And Affordability Of Inpatient Psychiatric Care In Tennessee, Kailee Havrda

Electronic Theses and Dissertations

Introduction: Access to mental health has seen changes over the past decade, but little research and attention have been given to access to inpatient psychiatric facilities. Inpatient psychiatric care is crucial for patients experiencing mental health emergencies who need immediate access to care. Gaps in spatial accessibility, affordability, and the access and assessment process were identified due to a lack of available information in the field.

Methods: Secondary data from the Tennessee Department of Health were used to assess financial accessibility through payer source charges and patient encounters, as well as through mapping of inpatient psychiatric facilities and conducting drive …

Kindergarten Teachers’ Perceptions Of Instructional Approaches For Teaching Sequencing Concepts: Worksheet, Hands-On, And Beebot-Integrated Lessons, Savannah P. Partin

Electronic Theses and Dissertations

The purpose of this qualitative multiple-case study was to explore kindergarten teachers’ perceptions of three instructional approaches for teaching sequencing concepts: worksheet-based instruction, hands-on manipulative instruction, and BeeBot-integrated instruction. Sequencing is a foundational early literacy skill that supports reading comprehension, story retelling, and understanding the order of events. Although sequencing is commonly taught through traditional instructional methods, limited research has examined how kindergarten teachers perceive the use of educational robotics as a tool for literacy instruction. The study was guided by constructivist theory, sociocultural theory, and developmental systems theory. Participants included six kindergarten teachers from a public elementary school in …

Understanding The Out-Of-Distribution Generalization Of Chain-Of-Thought Reasoning In Llms, Nuojing Liang, Xiaotong Yuan, Pan Zhou

Research Collection School Of Computing and Information Systems

Chain-of-Thought (CoT) reasoning substantially improves Large Language Models (LLMs), yet its out-of-distribution (OOD) generalization mechanism remains theoretically underexplored. We study CoT as compositional OOD generalization: the target task lacks complete reasoning trajectories or direct input-output pairs, while training provides only in-distribution (ID) atomic step data. We show that CoT generalizes by repeatedly applying a shared single-step predictor to compose reusable ID transitions. Theoretically, the OOD task risk is controlled by the sum of ID subtask risks, scaled by a depth-independent hard-decoding coefficient Bℓ b = O(k) that is independent of the reasoning depth T. A Rademachercomplexity analysis further gives an …

Drag As Evidence: Motion-Grounded Latent Recomposition For Drag-Based Editing, Xinyu Pu, Hongsong Wang, Jie Gui, Pan Zhou

Research Collection School Of Computing and Information Systems

Modern image editors excel at semantic manipulation and visual synthesis, yet remain limited in precise spatial control, motivating the development of drag-based editing. However, existing drag-based methods often struggle to balance drag accuracy with natural, plausible, and intent-aligned generation. We propose MoReDrag, a motion-grounded drag-based editing method. Our key insight is to treat pixel-space warping as coarse motion evidence, and to inject this evidence into the generative sampling trajectory. Specifically, MoRe-Drag performs region-aware latent recomposition over refinement, inpainting, and anchor regions, coupled with stage-adaptive conditioning that progressively shifts from motion-grounded structure formation to semantic refinement. We further support an instruction-free …

Llm-Enabled Applications Require System-Level Threat Monitoring, Yedi Zhang, Haoyu Wang, Xianglin Yang, Jin Song Dong, Jun Sun

Research Collection School Of Computing and Information Systems

LLM-enabled applications are rapidly reshaping the software ecosystem by using large language models as core reasoning components for complex task execution. This paradigm shift, however, introduces fundamentally new reliability challenges and significantly expands the security attack surface, due to the non-deterministic, learning-driven, and difficult-to-verify nature of LLM behavior. In light of these emerging and unavoidable safety challenges, we argue that such risks should be treated as expected operational conditions rather than exceptional events, necessitating a dedicated incident-response perspective. Consequently, the primary barrier to trustworthy deployment is not further improving model capability but establishing system-level threat monitoring mechanisms that can detect …

A Deep Learning Ensemble Framework For Multi-Subtype Renal Tumor Classification Using Contrast-Enhanced Ct, Hisham Abdeltawab, Ahmed Alksas, Mohamed Ghazal, Ashraf Khalil, Norah Saleh Alghamdi, Rasha T. Abouelkheir, Ahmed Elmahdy, Mohamed Abou El-Ghar, Sohail Contractor, Ayman El–Baz

All Works

Renal cell carcinoma (RCC) is considered the most aggressive and common form of renal cancer. Therefore, early detection is crucial to ensure appropriate and effective treatment planning. In our study, we propose a novel computer-aided diagnostic (CAD) approach which incorporates a deep learning ensemble to differentiate between five renal tumor subtypes, utilising the modality of contrast-enhanced computed tomography (CE-CT). The addressed renal lesions are malignant tumors (chromophobe RCC (chRCC), papillary RCC (pRCC), and clear cell RCC (ccRCC)) and benign tumors (renal oncocytoma (RO) and angiomyolipoma (AML)). Our study includes 280 patients who underwent renal biopsy, 112 patients were diagnosed with …

A Robust Approach For Olive Leaf Disease Detection In Uncontrolled Environments, Rima Grati, Khouloud Boukadi, Emna Ben Abdallah, Ahmed Seffah

All Works

Detecting diseases in olive leaves is crucial for maintaining tree health and ensuring stable olive production. Early signs of infection often appear on the leaves, making them a key indicator for timely disease detection and intervention. Traditionally, farmers rely on visual inspection or laboratory tests to diagnose plant diseases. However, recent advancements in deep learning (DL) have significantly improved the accuracy and efficiency of olive leaf disease diagnosis. Numerous studies in the literature have explored this task using CNN-based architectures and, more recently, Vision Transformers. While these models have shown promising performance on benchmark datasets, they are often trained and …

Ai-Driven Collision Risk Prediction And Debris Optimization In Low Earth Orbit, Emad Alkhazraji, Mohamed E. Yahia, R. E. Tolba, N. S. Alharthi

All Works

Debris congestion and collision in low Earth orbit have become a major point of concern due to the rapid expansion of activities in that region. In this work, an AI-driven framework is developed and deployed to predict, analyze, and optimize collision risk across the altitude inclination space. Local debris object population, mass density distribution, and conjunction derived risk are computed. Then, an AI surrogate model, viz., gradient-boosted decision trees, was trained to predict collision conjunction relative probabilities and risk. Thereafter, data analysis and explain ability techniques are utilized to uncover underlying risk drivers and reveal how altitude, inclination, mass loading, …

Representing The Brand: A Study To Examine The Best Practices For Structuring And Operating Communications Teams In Higher Education Institutions, Margaret L. Reisser

Masters Theses and Doctoral Dissertations

As higher education institutions increasingly compete for students, faculty, funding, and recognition, academic leaders are recognizing the essential role communications and public relations professionals play in maintaining and promoting the value of their institutional brands. Much has been written about the value of communications professionals in brand management in the corporate arena. However, it appears less attention has been paid to their role in higher education. This mixed-methods study was undertaken to examine the best practices for structuring and operating communications and public relations organizations for optimal success in promoting the higher education institutions they represent. Communications and public relations …

Examination Of Cleveland State Community College's Employee Professional Development Program: Leveraging Employee Learning And Relationship Building Amidst Significant Institutional Change, Ashley T. Raburn

Masters Theses and Doctoral Dissertations

Within the last decade, Cleveland State Community College has gone through many changes, including transitioning between five different administrations from 2021 to 2025. With this change in administration, there have been many different philosophies of how professional development should be handled at the college. The priority of professional development has shifted between very high and very low, and the ownership of active initiatives was unclear. Recently, college administrators have implemented a program intended to allow PD at the college to have more structure and visibility within the employee community. Literature in both Higher education and other industries discusses the importance …

Writing Assessment In Tennessee Community Colleges, Joshua Tucker

Masters Theses and Doctoral Dissertations

This mixed-methods study examined how composition faculty at Tennessee community colleges used margin comments to support student writing, whether those comments reflected principled or novice feedback language, and how instructors understood and described their own commenting practices. Marginal feedback remains the most widely used yet least theoretically examined form of response in first-year writing instruction, and the legislative reforms reshaping Tennessee's community college system, including the corequisite model, TNPromise, and TNReconnect have intensified questions about the pedagogical practices used to support an increasingly diverse student population. Eight full-time English instructors at Tennessee community colleges participated in the study, submitting 771 …

More Than Just Play: The Impact Of Campus Recreation On Retention In Modern Higher Education, Corbin D. Hedges

Masters Theses and Doctoral Dissertations

Student retention remains a central concern in higher education, particularly as the demographic composition of college populations shifts toward a growing number of contemporary students. While extensive research has examined retention strategies for traditional students, there is a notable gap in understanding how institutional resources, such as Campus Recreation, support contemporary student success. This study explores the impact of Campus Recreation participation on student retention, with a specific focus on contemporary students at the University of Tennessee at Chattanooga. Using a convergent mixed-methods design, this research investigates participation trends, perceived benefits, and barriers to engagement in Campus Recreation programs. Quantitative …

Arithmetic And Necessity: A Quasi-Concrete Modal Structuralist Account, Benjamin Faltesek

Mathematics, Philosophy and Physics (D.Sc.) Dissertations

This dissertation addresses the necessity of arithmetic by distinguishing two tasks that are often run together: the derivation task, which seeks to prove that mathematical truths are necessary, and the characterization task, which asks what kind of necessity arithmetic has and what grounds it. Setting the former aside for now, I focus on the latter, arguing that an adequate explanation of arithmetical necessity requires two sources: logical necessity and the necessary possibility that the right kind of structure exists. The second source makes mathematical modality distinctively mathematical, and the dissertation’s positive contribution emerges chiefly in that area. I survey and …

From Solitons To Chaos: Nonlinear Dynamics Of The Time-Fractional Modified Regularized Long-Wave Equation, Samsun Nahar Ananna, Chunhui Lu, Abdulla Al Mamun, Md Al Amin Meia, Nur Hasan Mahmud Shahen

School of Mathematical & Statistical Sciences Faculty Publications

This study investigates the nonlinear wave structures and complex dynamical behavior of the time-fractional modified regularized long-wave (mRLW) equation through an integrated analytical, numerical, and dynamical framework, with particular emphasis on connecting localized wave formation with perturbation-induced nonlinear dynamics. Within the adopted fractional framework, a Hirota-type bilinear approach is employed to construct one-wave, two-wave, three-wave, lump, one-lump–one-soliton, and one-lump–one-stripe structures, whose numerical consistency is subsequently examined using the Modified Variational Iteration (MVI) method through graphical comparisons and L 2 and L ∞ error norms. The periodically perturbed mRLW system is further investigated using phase portraits, Lyapunov exponents, bifurcation diagrams, Poincaré …

Mtorc2-Nav1.2 Signaling Drives Early Hyperexcitability In Alzheimer’S Disease Mouse Model, Nolan M Dvorak, Jeffrey L Noebels

Faculty, Staff and Students Publications

Hyperexcitability is a biomarker of early-stage Alzheimer’s Disease (AD) and hastens cognitive decline later in its course. Mechanistic target of rapamycin (mTOR) signaling contributes to the slope of this trajectory, as evidenced by early increased brain expression and the rescue of hyperexcitability by genetic deletion of mTOR complex 2 (mTORC2); however, a molecular mechanism directly linking mTOR signaling to membrane hyperexcitability in early-stage AD remains elusive. Here, we show that hyperactive mTOR signaling stimulates the voltage-gated Na+ channel 1.2 (Nav1.2), a previously identified downstream phosphorylation target of mTORC2 and a key regulator of membrane electrogenesis. Augmented Nav1.2 channel function induced …

Evaluating The Pulse Of Exercise Programs: Developing And Validating The Intervention Usability Scale For Exercise (Iuse) To Enhance Implementation And Adherence, Anne Inger Mørtvedt, Eva Ageberg, Steven Elmer, Kevin Trewartha, Erich Petushek

Michigan Tech Publications

Background: Adherence to exercise interventions is often suboptimal, despite numerous studies documenting barriers and facilitators. Usability may be a critical yet underexplored determinant of adherence. This study aimed to develop and assess the psychometric properties of the Intervention Usability Scale for Exercise (IUSE). Methods: Item generation and content validation involved cognitive interviews and feedback from eight exercise intervention stakeholders and ten target users from the general public. Subsequently, 526 target users from University, Qualtrics and Prolific participant panels assessed exercise programs through an online survey. Dimensionality was assessed using Principal Component Analysis (PCA), Exploratory Factor Analysis (EFA), and bifactor models. …

Stimuli-Responsive Proton Transfer: Photoacid Design, Co2 Sorbent Regeneration, And Polymer Applications, Melyse Ava Laud

Theses and Dissertations

Control over proton transfer enables the regulation of many pH-dependent chemical and material processes. Stimuli-responsive functional materials provide a means of achieving this control. Functional materials such as Metastable-state photoacids (mPAHs) reversibly release large amounts of protons when stimulated by light from a weakly acidic site within their structure. However, the dark acidities of previously reported mPAHs limit their use under basic conditions. To address this limitation, a benzimidazole-type mPAH was developed with an improved ground-state pKa of 11.9 and a metastable-state pKa of 9.8, enabling reversible proton transfer under mildly basic conditions. Additional mPAHs were examined to determine how …

Historical Piccolo Concerti And Works By Women Composers, Repertoire Catalogue, And A New Concerto For Piccolo And Orchestra, Mikaela Storm

Doctoral Scholarly Projects & Capstones

The following document is an overview of historical piccolo concerti with an emphasis on works by women, a repertoire catalogue of the piccolo concerti, and an analysis/performance guide for my own Concerto for Piccolo and Orchestra: The Dreamer. The primary goal of this document and subsequent score is to provide a resource for future composers, music theorists, conductors, and performers. The piccolo has limited repertoire in comparison to many of the standard instruments in our modern Western musical canon, and the number of piccolo concertos (particularly by women) is even smaller. The concerto that I have written, subtitled “The Dreamer,” …

The Hidden Heatwave: Thermal Hazards In The New York City Subway System, Abdou Rachid Bah, Maimuna Muntaha, Lenny L. Metlitsky, Cesar Pascal Vasquez, Tyler Ayala, Isaac Morel Lopez, Shaquan A. Larose, Kingston Ditsch, Kip Nielsen, Shaunak Sharma, Reginald U. Avila-Ortega Blake, Hamidreza I. Norouzi

Publications and Research

Underground transit systems (Subway) are susceptible to frequent heat waves, which pose serious health and safety risks to commuters and transit workers. Yet, the patterns and extent of temperature changes within the subway system remain insufficiently understood. To address this gap, we conducted a study of thermal variability across 112 subway stations in New York City. Temperature and relative humidity were measured at both platform and street levels. Results show that during the summer, 70% of subway stations had platform heat index higher than street level heat index. We also developed a machine learning forecasting model to predict platform-level temperatures. …

Study Of Fast Switching Metastable-State Photoacids And Controlled Drug Release From Polycaprolactone (Pcl) Vascular Scaffold, Rana Salman Abbood Abbood

Theses and Dissertations

Metastable-state photoacid (mPAH) has become a common tool for controlling and driving chemical processes with light. mPAHs with fast reverse reactions are desirable for precise temporal control or generating quick pulses of proton concentration. In this work, different approaches towards fast reversing mPAHs are studied. Experimental and computational results showed that stabilizing the charge–transfer intermediate is an effective way to increase the rate. A novel mPAH with a reverse reaction ≈500 times faster than the most widely used mPAH in methanol has been developed. Another water-soluble mPAH exhibited a reverse reaction with a rate constant of 7.8 s−1, the fastest …