Digital Commons Network^™

An Enhanced Deep Learning Framework For Crop Disease Detection Using Gan-Based Data Augmentation., Andrew Mc Guinn

ICT

Plant disease detection is a critical challenge in agriculture, where limited annotated datasets often hinder the training of robust deep learning models. This thesis explores the application of generative adversarial networks for data augmentation in the classification of six classes of sorghum diseases, addressing the scarcity of real training data. A lightweight convolutional neural network  (CNN) was employed as the classifier, trained under three experimental conditions: using only real images, a stratified mix of real and DCGAN-generated samples, and a stratified mix of real and WGAN-GP-generated samples.

The baseline CNN, trained on 279 real images, achieved an accuracy of 89.7%, …

Leaf Classification Using Convolutional Neural Networks And Vision Transformers, Louis Wilkie

ICT

This study investigates deep learning techniques for plant image classification using four publicly available datasets: Folio, Flavia, Swedish, and Sugarcane. Three modelling approaches—Convolutional Neural Networks (CNN), Vision Transformers (ViT), and hybrid CNN-ViT ensembles—are evaluated. Models leverage PyTorch implementations and pretrained weights from the Hugging Face hub, with hyperparameter tuning applied to optimize accuracy. Transfer learning with pretrained ViT models achieved the highest performance, attaining 100% accuracy on Folio, Flavia, and Swedish datasets, and 94.07% on Sugarcane. CNN and CNN-ViT models showed comparable results, but pretrained CNN-ViTs were more resource-efficient. The findings highlight the importance of evaluating multiple pretrained architectures to …

Exploring The Determinants Of Life Expectancy At Birth: Predicting And Forecasting Global Health Trends Using Statistical, Machine Learning, And Deep Learning Models., Emma Rath

ICT

Accurate life expectancy forecasting is essential for health policy planning, yet research comparing statistical, machine learning, and deep learning approaches under real-world constraints remains limited. This study evaluates ARIMA/ARIMAX, tree-based, and neural network models using Irish and global datasets, considering small samples, missing data, and COVID-19 shocks. ARIMAX with lagged socioeconomic variables outperformed LSTM and other ML/DL methods. Income-based stratification improved predictive accuracy and interpretability, with SHAP analysis highlighting GDP per capita for developed countries and school enrolment and trade indicators for developing contexts. Results provide practical guidance for policymakers and establish limits for model complexity under constrained health data.

Improving The Completeness Of Food Composition Databases Using Predictive Analysis., Carla Arenhart

ICT

This study investigates the use of machine learning regression models to impute missing micronutrient values in Food Composition Databases (FCDBs), focusing on the FAO/INFOODS dataset. A cascading prediction methodology leverages nutrient interdependencies to systematically estimate missing values. Four models—Random Forest (RF), Support Vector Machine (SVM), Gradient Boosting Machines (GBM), and Deep Neural Networks (DNN)—were evaluated using MAE, MSE, RMSE, and R². RF and GBM achieved the highest predictive accuracy for protein, phosphorus, calcium, and magnesium, demonstrating that ML-based predictive analytics can provide a more reliable alternative to traditional imputation methods. These findings support improved dietary assessments, nutritional research, and data-driven …

Towards Developing Glut2 Specific Probes For Biochemical And Biomedical Applications, Abdelrahman Ismail

Dissertations, Master's Theses and Master's Reports

Facilitated Glucose transporters (GLUTs) are membrane proteins responsible for the uptake of simple sugars and other structurally similar metabolites in all mammalian cells. One of the hallmarks of cancer is dysfunctional sugar metabolism, known as the Warburg effect, and one of the manifestations of that effect is abnormal GLUT expression. There are 14 known GLUT proteins with different expression patterns in different tissues, and tumors tend to express certain GLUTs that are not present in surrounding healthy tissue. Therefore, selective targeting of GLUTs presents a potentially powerful tool for diagnostics and treatment of cancer and other diseases. This has been …

Energy Transitions In Underserved Communities: Meaningful Engagement To Address Energy Burdens And Air Pollution, Nyasha Milanzi

Dissertations, Master's Theses and Master's Reports

This report presents two connected studies focused on environmental justice and community engagement in energy transitions. The first study examines engagement practices in EPA-funded projects using interviews and public abstracts. Applying the EngageINFEWS framework, it analyses stakeholder dynamics and offers recommendations for working with underserved communities. Key insights include addressing power imbalances, securing funding, building capacity, and fostering communication. The second study centres on the REJuST project, which supports energy justice in a rural northern community impacted by public health concerns linked to a bioenergy facility burning wood, paper, plastics, and tire-derived fuels. I use survey data to explore resident …

Test Stand Development: An Analysis Of Excavation Forces Within A Simulated Lunar Environment, Faith D. Gaertner

Dissertations, Master's Theses and Master's Reports

Space exploration has rapidly developed in recent years as scientists look to the lunar surface for water ice in permanently shadowed regions. The excavation and decomposition of water outside of Earth’s atmosphere could provide in-situ resources for propellant manufacturing and sustaining life. However, without informed design parameters for lunar exploration and infrastructure development, time will be wasted on systems unfit for the harsh environment. Testing facilities that simulate the environmental conditions are required to better understand the forces these complex systems may face while extraterrestrial.

The force test stand is a unique mechanism that collects tensile and compressive force data …

Near-Infrared Hemicyanine And Bodipy Sensors For Monitoring Mitochondrial Viscosity In Live Cancer Cells, Drosophila Melanogaster Larvae, And Adpkd Kidney Tissue, Joseph Peters

Dissertations, Master's Theses and Master's Reports

Two near-infrared fluorescent sensors, sensor A and sensor B, were synthesized and characterized. Sensor A was a hemicyanine derived structure, and sensor B was a BODIPY derived sensor. The sensors were tested to measure their ability to monitor viscosity, and sensor A was used to monitor viscosity in live cells. Both sensors were found to be effective in monitoring the viscosity level through optical measurements with glycerol. Sensor A was found to be successful in monitoring mitochondrial viscosity in HeLa MBA-MD-453 and MDA-MB-231 cell lines. It was also successful in monitoring viscosity in Drosophila melanogaster larvae as well as …

Biomedical Applications Of Reactive Oxygen Species From Catechol Oxidation, Zhongtian Zhang

Dissertations, Master's Theses and Master's Reports

Catechol, a key functional moiety found in mussel adhesive proteins, undergoes oxidation to generate reactive oxygen species (ROS), which can be leveraged for biomedical and environmental applications. By modifying catechol-based polymers, ROS generation can be controlled and applied toward organic compound degradation, antimicrobial treatments, and polymer crosslinking. This dissertation focuses on the design and application of catechol-based materials that leverage ROS generation for biomedical and environmental applications.

Investigating Dimensions Of Social Vulnerability To Wildfire: A Case Study Of Western Montana, Ella Brodowski

Dissertations, Master's Theses and Master's Reports

Communities across the US are facing increased wildfire risks due in part to historic fire suppression efforts, a warming climate, and an influx of people residing in wildland urban interface (WUI) landscapes. Because of these increasing risks, there have been growing concerns about the social factors that can increase individual and community vulnerability to wildfires. This research adds to the discussion of social vulnerability by examining how people living in a geographic area with heightened wildfire risk think about the ways in which people can be more or less vulnerable to wildfires. Using semi-structured interviews, round table discussions, and field …

Relict Ponderosa Pine Mortality And Fuels Study In The Bob Marshall Wilderness, Montana, Usa, Laura K. Schimmel

Dissertations, Master's Theses and Master's Reports

Many forests have historically experienced mixed-severity fire regimes that were disrupted during the 20th century due to fire exclusion policy. A stand of relict Pinus ponderosa in the Bob Marshall Wilderness Complex that burned in 2003, 2011, and 2018 provides a unique case study for investigating the effects of returned fire to the landscape. Many of the trees have fire scars or Indigenous bark-peeling scars, preserving valuable ecological and anthropological history. We sought to determine whether scarred trees were more susceptible to fire damage, if larger trees had greater fire resilience than smaller trees, and if fuel loads have changed …

Life Cycle Assessment And Techno-Economic Analysis Of Mine Tailings As A Sustainable Raw Material For Mineral Carbonation And Recovery Of Critical Minerals, Vida Amankwaa

Dissertations, Master's Theses and Master's Reports

This study assesses the feasibility of using mine tailings for mineral carbonation and nickel recovery through a comprehensive Life Cycle Assessment (LCA) and Techno-Economic Analysis (TEA). The LCA quantifies greenhouse gas emissions, energy use, and resource consumption, while the TEA evaluates capital investment, operational costs, and process efficiency. The conventional beneficiation process at Eagle Mine serves as a baseline, highlighting its environmental footprint and operational costs. Results indicate that tailings reprocessing incurs higher costs ($34.76 per tonne of tailings processed) compared to conventional ore processing ($25.90 per tonne) due to increased energy consumption and material handling requirements. Additionally, several mineral …

Land Rematriation: Utilizing The National Register Of Historic Places And Developing Tribal Homeland Territories For The Restoration Of Historic Landscapes And Sovereignty, Robert E. Hazen

Dissertations, Master's Theses and Master's Reports

Land rematriation and cultural revitalization are crucial to addressing the ongoing impacts of settler colonialism on Indigenous communities, particularly for the Lac Vieux Desert Band of Lake Superior Chippewa. The National Register of Historic Places and federal recognition processes have traditionally marginalized Indigenous practices and land claims, yet, as I argue in this thesis, they hold potential for reappropriation in support of sovereignty and heritage. This thesis argues that integrating Indigenous methodologies, geospatial research, and archival analyses can forge a path toward effective land rematriation efforts. My findings reveal culturally significant sites connected to the historic village of Getegitigaaning, helping …

Life Cycle Assessment Of Renewable Hydrocarbon Fuels Produced By The New Cool Gas To Liquid (Cool Gtl) Process, Elochukwu Okoye

Dissertations, Master's Theses and Master's Reports

The Cool Gas-to-Liquid (Cool GTL) process converts biogas and captured CO₂ into renewable hydrocarbon fuels, contributing to GHG emissions reduction in transportation. This report explains a Life Cycle Assessment (LCA) study that was completed to evaluate the environmental impact of different feedstocks, hydrogen sources, and electricity inputs. The analysis considers biogas from food waste, manure, and landfill sources, as well as biogenic, fossil, and direct air capture (DAC) CO₂ feedstocks. Results show that biogas pathways provide the greatest GHG reductions, with the best case (food and manure biogas + solar hydrogen + solar electricity) achieving a net-negative GWP …

Life Cycle Assessment And Life Cycle Cost Assessment Of A Floating Treatment Wetland In A Wastewater Lagoon Located In Copper Harbor, Michigan, Erin Browne

Dissertations, Master's Theses and Master's Reports

A full-scale floating treatment wetland was implemented in a wastewater treatment lagoon in Copper Harbor, Michigan. Copper Harbor is a small, rural community located in a cold climate. Many studies have focused on the performance of floating treatment wetlands in mesocosm studies in warmer climates, but there is not a lot of research conducted on the feasibility and performance of full-scale floating treatment wetlands in wastewater lagoons, specifically in cold climates. The 2024 treatment season was the first season with the floating treatment wetland implemented in the wastewater lagoon. A preliminary assessment of the cost and environmental impacts of a …

Microwave Assisted Syntheses And Characterizations Of Near-Infrared Ratiometric Probes For Mitochondrial Ph Detection And Monitoring, Ronald J. Schwandt

Dissertations, Master's Theses and Master's Reports

Within the field of biochemistry, there has been a significant development of ratiometric fluorescent probes designed for mitochondrial pH detection; however, many of these probes have emissions within the visible spectrum and require long synthesis times with low-yield reactions. However, in recent years, microwave-assisted syntheses of organic molecules have led to significant increases in reaction yields and quicker reaction times. In this work, two common reactions in probe synthesis, Heck coupling, and Knoevenagel condensations are investigated using microwave-assisted synthesis. It was found the low-concentration Heck coupling reactions using palladium (II) acetate led to extremely efficient Heck coupling. Also, a novel …

Barbie Vlogger As Teen Influencer For Little Girls, Brilynn A. Janckila

Dissertations, Master's Theses and Master's Reports

Mattel’s YouTube Barbie Vlogs series (2015-2022) targets 6–11-year-old girls with short narratives and prosocial messages addressing their prevalent interests and everyday experiences. While the vlog is clearly an attempt by Mattel to appeal to a market sector that includes both little girls and their Millennial parents, this analysis examines Barbie Vlogger as a social media influencer in order to explore how little girls learn to become older girls through the performance of teenage girlhood as enacted in the Barbie Vlogs. What is Barbie doing in the vlogs to show little girls who a teen girl is and what she …

Microbial Inactivation And Spray Drying For The Recovery Of Single-Cell Protein From Chemically Deconstructed Polyethylene Terephthalate, Reeba Elizabeth Reji

Dissertations, Master's Theses and Master's Reports

Polyethylene terephthalate (PET) is one of the most widely used polymers in plastics packaging. However, the increasing demand for PET has created challenges in waste disposal. Previous research has shown that PET can be deconstructed using ammonium hydroxide (CDPET). Microorganisms capable of utilizing the product as a sole carbon source can be converted to single-cell protein (SCP). A custom microbial inactivation unit was developed to ensure the safety and stability of generated SCP. Our study evaluated the effect of exposure time (45–60 minutes) at 140 °C with the goal of maximizing microbial inactivation. Our findings suggest that microbial inactivation can …

Evaluating Challenges And Solutions For Buckthorn Classification In Shadowed Environments, Moses Tangwam

Dissertations, Master's Theses and Master's Reports

Invasive species, such as buckthorns, pose significant ecological threats by displacing native vegetation and reducing biodiversity. This study examines the impact of shadows on the classification accuracy of buckthorns using drone-based multispectral imagery collected in a forested area near Michigan Technological University. Shadows impacted approximately 70% of the imagery, notably distorting reflectance in key spectral bands such as near-infrared (NIR) and red edge. The study evaluated vegetation indices like NDVI and the performance of machine learning models, specifically the Random Forest classifier, under these shadowed conditions. Conventional shadow correction techniques, including histogram normalization, Shadow Index (SI), and Inverse Distance Weighting …

"An Ancient Vanished Race:" Industrial Mythistory And The Alternative Archaeology Of Lake Superior Copper, Donovan C. Murphy

Dissertations, Master's Theses and Master's Reports

This thesis explores how the perspectives, biases, and values of Euro-Americans regarding the copper mining industry have shaped a mythology which centers Lake Superior copper in an imagined pre-Columbian global trade network. Notably, many iterations of this mythology are rooted in a rejection of Indigenous agency in the creation of these sites, ascribing them instead to a ‘vanished race’ which had surely once occupied the region. The core of this thesis is formed around a set of semi-structured interviews with representatives of regional heritage institutions and avocationalists which were conducted with the goal of identifying how and why these stories …

Energy Equity In India’S Renewable Energy Transition: A Policy Design Agenda, Aritra Chakrabarty

Dissertations, Master's Theses and Master's Reports

India’s transition to renewable energy has prioritized large-scale, grid-based electrification, often overlooking energy equity, particularly in rural communities. Many households would benefit more from decentralized solutions than from grid extension, yet current policies neither guarantee equitable access nor address pre-existing social and economic inequalities in energy distribution. This dissertation examines the sensitivity of India’s renewable energy policies to energy equity, with a specific focus on gender equity in rural last-mile connectivity. The study is guided by the fundamental question: How can implementation of solar energy program alleviate the lives of rural communities? Using a mixed-method approach that combines policy …

Generalizing Medical Image Segmentation Task With Efficient Deep Learning Models, Abel A. Reyes-Angulo

Dissertations, Master's Theses and Master's Reports

Medical Image Segmentation is a critical task in the field of medical imaging, playing a crucial role in diagnostics, treatment planning, and disease monitoring. The emergence of Deep Learning (DL) has ushered in a new era in Artificial Intelligence (AI), propelling remarkable advancements in key domains like language translation, object recognition, and recommendation systems. This evolution has been accompanied by continuous enhancements in computational efficiency and improvements in predictive accuracy. The introduction of sophisticated algorithms, such as convolutional neural networks (CNNs) and transformers, exemplifies these advancements. DL algorithms have demonstrated exceptional efficacy in medical image segmentation tasks, showcasing the potential …

Patient Handling Interventions And Their Role In Reducing Musculoskeletal Disorders Among Healthcare Workers, Ashley L. Hawke

Dissertations, Master's Theses and Master's Reports

Musculoskeletal disorders (MSDs) are prominent among healthcare professionals, especially those who perform manual patient handling tasks. Assistive devices (e.g., slider sheet, gait belt, mechanical lift) have been used for patient handling. However, the effectiveness of these devices in reducing MSD risk factors is varied. Additionally, the relationship between healthcare workers’ acceptability of patient handling devices and organizational policies that facilitate or hinder widespread implementation is not well understood. The goal of my dissertation was to conduct a comprehensive analysis of existing patient handling devices and their role in mitigating MSD risk among healthcare workers. For Study 1, I conducted a …

Triboelectrostatic Separation Of Lunar Regolith, Hannah Peterson

Dissertations, Master's Theses and Master's Reports

Lunar regolith as a source of raw materials for a long-term human presence on the moon would need to be separated into single mineral fractions. To do this, triboelectric properties can be utilized to separate minerals without needing to add a working fluid, such as air and water, or expensive, consumable chemicals. A lightweight machine can be created to do this, one that also requires minimal electricity. A series of three prototypes for this machine were created and tested to determine if this concept would work, and what properties the machine would need. A mixture of magnetite and silica were …

Dam It All: A Multi-Scalar Investigation Into The Ecological Role And Function Of Beaver Dams, Kenneth Larsen

Dissertations, Master's Theses and Master's Reports

The central theme of this dissertation is ecological scaling, which examines how small, localized features, such as beaver ponds, contribute to large-scale watershed processes and how different modeling approaches at various scales reveal distinct insights. Through a combination of probabilistic modeling and high-resolution empirical research, this work demonstrates that small ponded systems, particularly those created by beavers (Castor canadensis), serve as critical biogeochemical control points, whose contributions to sediment and nutrient dynamics have been historically underestimated.

In Chapter 1, I contextualize this research within the broader fields of freshwater ecology, watershed modeling, and ecological restoration. I introduce the …

Cyanine-Based Dual-Emission Sensor For Ratiometric Nad(P)H Imaging In Cells, Renal Tissues, And Entire Organisms, Peter Agyemang

Dissertations, Master's Theses and Master's Reports

Monitoring the dynamic fluctuations of NADH and NADPH, key coenzymes in Intracellular redox homeostasis and bioenergetics, are essential for understanding both normal physiology and disease pathology. However, precise, real-time imaging of NAD(P)H levels in complex biological systems remains a major challenge due to interference from endogenous fluorescence and the limitations of intensity-based sensors. We present a newly engineered dual-emission fluorescent sensor, sensor A, which overcomes these obstacles by offering ratiometric detection with high specificity and sensitivity. The sensor features a 3-quinolinium moiety conjugated to a cyanine dye scaffold via an amine linker, enabling selective reactivity toward NAD(P)H. Upon reduction, …

Design And Development Of Fluorescent Probes For Sensing And Bioimaging Of Nad(P)H, Intracellular Viscosity, And So2, Dilka Liyana Arachchige

Dissertations, Master's Theses and Master's Reports

Understanding the dynamic biochemical and biophysical changes within living systems is vital for advancing disease diagnosis and therapeutic development. This work presents the design and development of innovative fluorescent probes, based on cyanine and coumarin scaffolds, for the highly sensitive and selective detection of key metabolic and biophysical markers, NAD(P)H, mitochondrial viscosity, and sulfur dioxide (SO₂) in live cells and fruit fly larvae. A series of cyanine-based probes were engineered to undergo strong fluorescence enhancement upon reduction by NAD(P)H, enabling real-time visualization of metabolic fluctuations in cancer cells and Drosophila larvae. These probes exhibit excellent sensitivity, mitochondrial targeting capability, and …

Method For Assessing Gasoline Powered Internal Combustion Engine Exhaust Accumulation When Idling In Enclosed Space By Automatic Shut Off, Eamon Mcclintock

Dissertations, Master's Theses and Master's Reports

About 120 fatalities/year are caused by Carbon Monoxide (CO) poisoning outside of the vehicle which can take place in the garage or attached house. A controlled test environment was created to mimic a garage with representative ventilation, volume, and instrumentation. Tests were performed with vehicles of varying engine displacement size using parameters tuned to replicate four scenarios of vehicle operation including vehicle start state, HVAC operation, and ambient temperature. A house model was built to model CO accumulation using realistic air exchange rates to simulate air transfer between the garage and attached house. Depletion of Oxygen (O2) in the garage …

Stepper Motor Load Estimation Using A Neural Network, Soham S. Kulkarni

Dissertations, Master's Theses and Master's Reports

Bipolar stepper motors are common in scenarios that require low-cost precision including 3d printers. The ability to estimate stepper motor load provides an opportunity for health monitoring. For example, excessive load estimates may indicate a future problem in the driven system such as bearing wear. Stepper motor drivers often have built-in stall detection strategies that exploit the driver's precise knowledge of the high-frequency voltage pulses being sent to the motor. The goal of this work was to develop a retrofit approach to stepper motor load estimation using externally measured voltage, current, and speed. A stepper motor dynamometer was created to …

Study Of Two-Phase Void Fraction In A Rectangular Channel Using Capacitance Sensor, Peter Ngoc Nguyen

Dissertations, Master's Theses and Master's Reports

Two-phase liquid-gas flow has many applications in the nuclear industry, including active heat exchangers, reactor core designs, secondary steam generators, and two-phase flow loop balance of the plant. The void fraction is a crucial parameter to determine the pressure drop, flow regimes and liquid levels inside the nuclear reactor core. For the past forty years, void fraction has been studied by many researchers by different measurement techniques. Each method has positive and negative attributes. In this study, after extensive literature reviews, the author has chosen a capacitance type void fraction sensor over other void fraction measuring techniques. The principal concept …