Digital Commons Network^™

Generative Data Augmentation: Using Dcgan To Expand Training Datasets For Chest X-Ray Pneumonia Detection, Ryan D. Maier

Master's Theses

Recent advancements in computer vision have demonstrated remarkable success in image classification tasks, particularly when provided with an ample supply of accurately labeled images for training. These techniques have also exhibited significant potential in revolutionizing computer-aided medical diagnosis by enabling the segmentation and classification of medical images, leveraging Convolutional Neural Networks (CNNs) and similar models. However, the integration of such technologies into clinical practice faces notable challenges. Chief among these is the obstacle of acquiring high-quality medical imaging data for training purposes. Patient privacy concerns often hinder researchers from accessing large datasets, while less common medical conditions pose additional hurdles …

Implementation And Efficiency Of Electric Motors In Lygus Bug Vacuums For Strawberries, Adam J. Rantz

Master's Theses

Further automation of the agricultural industry and an increase in organic fruit production are needed to address labor shortage and increased demand. The move to an autonomous bug vacuum is easier if the equipment is electric, and not powered by a tractor’s PTO (power take-off) shaft. The main objective of this project was to design a set of lygus bug vacuums to go on a robot that meets industry standards while using electric power instead of hydraulic. The frame of the vacuum was designed using the industry standard. While hydraulic systems have been a very effective power source for vacuums …

Engineering An Injectable Hydrogel With Self-Assembling 3d Vasculature, Kendyl Cohn

Master's Theses

This research developed methods for culturing self-assembling capillaries in an injectable gel as a potential method for vascularizing tissue-on-a-chip models to mimic physiological drug delivery. Additionally, a mathematical model was developed as a tool for understanding nutrient delivery and comparison of potential delivery systems. Organs-on-a-chip provide novel platforms for studying biology and physiology in 3D, allow exploration of tissue engineering on a manageable scale, and serve as models for drug screening and drug-delivery testing.

Methods were first developed for co-culture of endothelial cells and fibroblasts (3T3s or HDFs) in 2D, evaluating culture time, seeding density and ratio of HUVECs and …

Feasibility And Characterization Of Leak-Tight Single-Track Thin Walls Produced By Laser Powder Bed Fusion In 316l Stainless Steel, Peyton J. Archibald

Master's Theses

This thesis explores the optimization of process parameters for producing single-track thin walls using Laser Powder Bed Fusion (LPBF) additive manufacturing. Using two different coupon designs, the study assesses the feasibility of creating the thinnest possible leak-tight structures within LPBF and evaluating their mechanical properties, including burst pressure and modulus of elasticity under pressure loads. A series of experimental iterations were conducted, varying laser power and laser speed to identify optimal conditions. The findings indicate that a narrow range of process parameters can produce consistently leak-tight thin walls. The results contribute to understanding how to achieve high quality, reliable thin …

Anomaly Detection In Heterogeneous Iot Systems: Leveraging Symbolic Encoding Of Performance Metrics For Anomaly Classification, Maanav Patel

Master's Theses

Anomaly detection in Internet of Things (IoT) systems has become an increasingly popular field of research as the number of IoT devices proliferate year over year. Recent research often relies on machine learning algorithms to classify sensor readings directly. However, this approach leads to solutions being non-portable and unable to be applied to varying IoT platform infrastructure, as they are trained with sensor data specific to one configuration. Moreover, sensors generate varying amounts of non-standard data which complicates model training and limits generalization. This research focuses on addressing these problems in three ways a) the creation of an IoT Testbed …

Predicting Rheology Of Uv-Curable Nanoparticle Ink Components And Compositions For Inkjet Additive Manufacturing, Cameron D. Lutz

Master's Theses

Inkjet additive manufacturing is the next step toward ubiquitous manufacturing by enabling multi-material printing that can exhibit various mechanical, electronic, and thermal properties. These characteristics are realized in the careful formulation of the inks and their functional materials, but there are many constraints that need to be satisfied to allow optimal jetting performance and build quality when used in an inkjet 3-D printer. Previous research has addressed the desirable rheology characteristics to enable stable drop formation and how the metallic nanoparticles affect the viscosity of inks. The contending goals of increasing nanoparticle-loading to improve material deposition rates while trying to …

A Federation Of Sentries: Secure And Efficient Trusted Hardware Element Communication, Blake A. Ward

Master's Theses

Previous work introduced TrustGuard, a design for a containment architecture that allows only the result of the correct execution of approved software to be outputted. A containment architecture prevents results from malicious hardware or software from being communicated externally. At the core of TrustGuard is a trusted, pluggable device that sits on the path between an untrusted processor and the outside world. This device, called the Sentry, is responsible for validating the correctness of all communication before it leaves the system. This thesis seeks to leverage the correctness guarantees that the Sentry provides to enable efficient secure communication between two …

Exploring The Feasibility Of The Resonance Corridor Method For Post Mission Disposal Of High-Leo Constellations, Payton G. Porter

Master's Theses

In the upcoming decade, the proliferation of high-LEO constellations is expected to exceed 20,000 objects, yet comprehensive Post Mission Disposal (PMD) strategies for these constellations are currently lacking. With the inherent challenges of efficiently deorbiting satellites from High-LEO orbits, there arises an urgent need to explore innovative approaches. Building upon insights garnered from the ReDSHIFT project and anticipating the proliferation of high-LEO constellations such as OneWeb, TeleSat, and GuoWang, this thesis delves into the potential viability of the Resonance Corridor Method for PMD. The investigation encompasses key metrics, including deorbit timelines and $\Delta v$ requirements to meet regulatory standards or …

Deep Learning Using Vision And Lidar For Global Robot Localization, Brett E. Gowling

Master's Theses

As the field of mobile robotics rapidly expands, precise understanding of a robot’s position and orientation becomes critical for autonomous navigation and efficient task performance. In this thesis, we present a snapshot-based global localization machine learning model for a mobile robot, the e-puck, in a simulated environment. Our model uses multimodal data to predict both position and orientation using the robot’s on-board cameras and LiDAR sensor. In an effort to minimize localization error, we explore different sensor configurations by varying the number of cameras and LiDAR layers used. Additionally, we investigate the performance benefits of different multimodal fusion strategies while …

Manufacturing Silicone In-House For The Creation Of Customized Neurovascular Blood Vessel Mimics, Jacob Wilbert Perisho

Master's Theses

The Tissue Engineering Lab at California Polytechnic State University San Luis Obispo focuses on creating tissue-engineered Blood Vessel Mimics (BVMs) designed for the preclinical testing of neurovascular devices. These BVMs are composed of silicone models, representing anatomically accurate neurovasculatures, that are sodded with vascular cell types and then cultivated in bioreactors (which maintain physiologic conditions). These silicone models are currently sourced externally from industry partners, so the primary goal of this thesis was to develop the means and methods for the Tissue Engineering Lab to manufacture silicone models in-house.

The first aim of this thesis was to develop and explore …

Development Of An In Vitro 3-Dimensional Co-Culture Human Colorectal Cancer Model In Microfluidic Devices, Abby Jens

Master's Theses

Colorectal cancer is the second most common cause of cancer-related deaths in the United States, with the relative 5-year survival rate for distant stage cancer being only 14%. The most common treatment for colorectal cancer is with chemotherapeutic drugs; however, the discovery of these drugs is costly, time-consuming, and often requires the use of animal models that do not yield results that translate to clinical trials. Due to these shortcomings, researchers seek to develop physiologically relevant in vitro tumor models that more accurately mimic the tumor microenvironment for cheaper and faster high-throughput drug screening. The aim of this research was …

Insights Into Cellular Evolution: Temporal Deep Learning Models And Analysis For Cell Image Classification, Xinran Zhao

Master's Theses

Understanding the temporal evolution of cells poses a significant challenge in developmental biology. This study embarks on a comparative analysis of various machine-learning techniques to classify cell colony images across different timestamps, thereby aiming to capture dynamic transitions of cellular states. By performing Transfer Learning with state-of-the-art classification networks, we achieve high accuracy in categorizing single-timestamp images. Furthermore, this research introduces the integration of temporal models, notably LSTM (Long Short Term Memory Network), R-Transformer (Recurrent Neural Network enhanced Transformer) and ViViT (Video Vision Transformer), to undertake this classification task to verify the effectiveness of incorporating temporal features into the classification …

Comparison Of Scanning Electron Microscopy And Confocal Laser Microscopy For Tissue Surface Roughness Characterization, Tarnvir Dhaliwal

Master's Theses

It was found that the measurements captured by confocal microscopy and scanning electron microscopy had a statistically significant difference for bovine tissue. There was not a statistically significant for porcine and poultry tissue.

The intent of the study is to perform a comparative study to examine efficacy of two distinct technologies for a singular purpose: tissue surface roughness characterization. The two technologies compared are a confocal reflectance microscope and a scanning electron microscope. The comparison was made by comparing two surface roughness parameters [Ra and Rq] within ImageJ.

The study examined three different animal species [porcine, bovine, and poultry] to …

A Comparison Of Western And Eastern Soft Systems Approaches, John L. Anaya, John L. Anaya, John L. Anaya

Master's Theses

Soft System Approaches have been developed worldwide to help problem-solvers and decision-makers develop solutions to complex problems, such as aerospace systems. Soft System Approaches were designed to help lower the disorder of developing a complex system by increasing understanding of a situation. Four Soft System Approaches were investigated, two from the West and two from the East. Within the context of the paper, the West refers to thought patterns associated with thinkers and scientists in Europe and North America, and the East refers to those from and around China. The two from the West are Peter Checkland’s Soft Systems Methodology …

A Study Of Random Partitions Vs. Patient-Based Partitions In Breast Cancer Tumor Detection Using Convolutional Neural Networks, Joshua N. Ramos

Master's Theses

Breast cancer is one of the deadliest cancers for women. In the US, 1 in 8 women will be diagnosed with breast cancer within their lifetimes. Detection and diagnosis play an important role in saving lives. To this end, many classifiers with varying structures have been designed to classify breast cancer histopathological images. However, randomly partitioning data, like many previous works have done, can lead to artificially inflated accuracies and classifiers that do not generalize. Data leakage occurs when researchers assume that every image in a dataset is independent of each other, which is often not the case for medical …

A Sindy Hardware Accelerator For Efficient System Identification On Edge Devices, Michael Sean Gallagher

Master's Theses

The SINDy (Sparse Identification of Non-linear Dynamics) algorithm is a method of turning a set of data representing non-linear dynamics into a much smaller set of equations comprised of non-linear functions summed together. This provides a human readable system model the represents the dynamic system analyzed. The SINDy algorithm is important for a variety of applications, including high precision industrial and robotic applications. A Hardware Accelerator was designed to decrease the time spent doing calculations. This thesis proposes an efficient hardware accelerator approach for a broad range of applications that use SINDy and similar system identification algorithms. The accelerator is …

Finite Element Analysis Of The Bearing Component Of Total Ankle Replacement Implants During The Stance Phase Of Gait, Timothy S. Jain

Master's Theses

Total ankle replacement (TAR) implants are an effective option to restore the range of motion of the ankle joint for arthritic patients. An effective tool for analyzing these implants’ mechanical performance and longevity in-silico is finite element analysis (FEA). ABAQUS FEA was used to statically analyze the von Mises stress and contact pressure on the articulating surface of the bearing component in two newly installed fixed-bearing total ankle replacement implants (the Wright Medical INBONE II and the Exactech Vantage). This bearing component rotates on the talar component to induce primary ankle joint motion of plantarflexion and dorsiflexion. The stress response …