Digital Commons Network^™

Deep Learning For Tomographic Image Reconstruction Guided By Generative Models And Image Science, Sayantan Bhadra

McKelvey School of Engineering Graduate Student Theses & Dissertations

Tomographic image reconstruction is generally an ill-posed inverse problem. Such inverse problems are typically regularized using prior knowledge of the sought-after object property. Recently, deep neural networks have been actively investigated for regularizing image reconstruction problems by learning a prior for the object properties from training images. Deep generative models such as generative adversarial networks (GANs) have demonstrated the ability to learn object distributions comprehensively and synthesize high-quality images. This dissertation explores novel generative model-constrained reconstruction methods that employ state-of-the-art GANs in the context of ill-posed tomographic imaging problems. The symbiotic relationship between image science and deep learning to enable …

Advancements In Magnetic Resonance Image Guided Radiotherapy, Austen Curcuru

McKelvey School of Engineering Graduate Student Theses & Dissertations

Magnetic resonance image guided radiation therapy (MRgRT) devices are a recently developed technology that integrate the excellent soft tissue contrast and real-time imaging capabilities of MRI with a medical linear accelerator (Linac). This provides an unprecedented ability to guide and adapt radiation therapy treatments based on real-time cine imaging. However, the merging of these technologies has come with unique challenges. MRI lacks the geometric fidelity of computed tomography (CT). Spatial inaccuracies in MRI can result from magnetic field (B0) or center frequency variations, gradient-induced eddy currents, and magnetic field gradient imperfections (e.g., nonlinearities, poor calibration, concomitant fields, and unsatisfactory electronic …

The Role Of Trpv4 In The Intervertebral Disc Response To Mechanical Loading, Garrett Easson

McKelvey School of Engineering Graduate Student Theses & Dissertations

Low back pain (LBP) is one of the most common and debilitating musculoskeletal disorders in the world, with a lifetime prevalence above 80%. The cause of LBP is multifactorial, with intervertebral disc (IVD) degeneration being one of the most prominent contributors to its development. IVD degeneration occurs due to aging, aberrant mechanical loading, and injury. The IVD responds to these stimuli and experiences inflammation, catabolism, neovascularization, and innervation. While the degenerative cascade is well characterized, how the IVD transduces mechanical loading and injury stimuli into a degenerative response is still poorly understood. Mechanosensitive ion channels are one mechanism that musculoskeletal …

To Ascertain Scientific Reliability In Pet Imaging For Alzheimer Disease, Charles Chen

McKelvey School of Engineering Graduate Student Theses & Dissertations

Alzheimer disease is one of the deadliest and most expensive diseases in the United States. Knowing whether a family member has Alzheimer disease may help households better plan for health-related expenses. However, fewer than half of primary care physicians report being comfortable diagnosing cognitive impairment due to Alzheimer disease. One solution is to supplement cognitive assessments with biomarkers to track abnormal levels of amyloid-β and tau, the pathological hallmarks of Alzheimer disease. Among these, positron emission tomography (PET) has emerged as an established standard that other biomarkers are compared to. However, as PET imaging moves from being used in research …

Cavity Optomechanics In Whispering-Gallery-Mode Microresonators And Its Applications, Zhoutian Fu

McKelvey School of Engineering Graduate Student Theses & Dissertations

Optical whispering-gallery-mode (WGM) microresonators with high quality factor and small mode volume are able to significantly enhance light-matter interaction, which is promising for the studies of nonlinear optics as well as for the sensing applications. Due to the enhanced intracavity optical intensity, the radiation pressure of the optical field can be strong enough to be coupled to the mechanical modes of the microresonator. This coupling between the optical mode and mechanical mode is referred to as the cavity optomechanical coupling, which has enabled many interesting effects including optomechanical heating and cooling, optomechanically induced transparency, phonon lasers and chaos.In this dissertation, …

Photochemical Formation Mechanisms And Applications Of Mn Oxide Nanomaterials, Zhenwei Gao

McKelvey School of Engineering Graduate Student Theses & Dissertations

Manganese (Mn) oxides are ubiquitous in aquatic and terrestrial environments, and the rich redox chemistry of manganese is of interest in engineered systems, such as batteries and water oxidation catalysts. Mn oxides have large surface areas and are highly active in adsorption and redox reactions. Our research group recently reported photochemically-assisted abiotic inorganic oxidation of Mn2+(aq) to δ-MnO2 nanosheets in the presence of nitrate and without a microorganism, demonstrating that abiotic Mn oxidation processes are not negligible. My doctoral work focuses on investigating the effects of various aquatic chemistry parameters, including reactive oxygen species (ROS), peroxyl radical (ROO•), reactive halogen …

Synthetic Biology Approaches For Genome-Wide Regulatory Response Characterization, Yichao Han

McKelvey School of Engineering Graduate Student Theses & Dissertations

Bacteria can survive and adapt to various environments by changing their gene expression. This process is controlled by both transcription factors and shared cellular resources. Elucidating the regulatory responses in bacteria at the genome scale is crucial for biomedical and biotechnology applications. To date, even for the best-characterized bacterium, Escherichia coli, the regulatory interactions of more than 60% of operons are still unclear. In this thesis, I described two synthetic biology methods that allow the perturbation of key component concentrations and the measurement of gene expression response systematically. I characterized two types of regulatory processes: transcriptional regulation and shared resource …

Numerical Simulation, Turbulence Modeling And Active Flow Control Of High Lift Aerodynamic Flows, Karsten Hendrickson

McKelvey School of Engineering Graduate Student Theses & Dissertations

The accurate predictions of 3D high lift flows have proven to be a significant challenge for Computational Fluid Dynamics (CFD) using the Reynolds-Averaged Navier-Stokes (RANS) equations with a turbulence model. While RANS methods typically require less computational resources than Scale Resolving Simulation (SRS) methods, they generally struggle to accurately predict flow separation and the complex interactions that occur between the fuselage, different elements of the wing, the engine nacelle, and the accompanying support hardware. As a result, significant effort has been put in improving the accurate prediction of these flows with the goal of developing better computational methodologies and turbulence …

Nanoparticle Nucleation, Trapping, & Effluent Charging In Low-Temperature Plasmas, Eric Husmann

McKelvey School of Engineering Graduate Student Theses & Dissertations

Nanoparticles in nonthermal plasmas (i.e., dusty plasmas) have been the subject of both great concern and great promise. For microelectronics fabrication, particles from nonthermal plasmas behave as a problematic system impurity; however, due to the interesting properties of the nanoparticles which can be synthesized in nonthermal plasmas, there has been much interest to develop nonthermal plasma-synthesized nanoparticles for a wide range of applications. Given this interest in nonthermal dusty plasma systems, understanding how systematic parameters affect dusty plasma systems is of great importance. As such, this work will focus on three particular phenomena: how systematic parameters affect nanoparticle formation, how …

A Quality Assurance Pipeline For Deep Learning Segmentation Models For Radiotherapy Applications, Xiyao Jin

McKelvey School of Engineering Graduate Student Theses & Dissertations

The applications of deep learning (DL) in medical imaging have been extensively developed and investigated to tackle a wide variety of problems, with many of them able to achieve state-of-the-art performances in different tasks. In the field of radiation therapy (RT), the workflow is a complex process consisting of multiple time and effort-consuming steps such as organ-at-risk (OAR) contouring that could impact treatment quality and patient outcomes. Due to the rapid growth in patient data and the advanced DL techniques, increased attention has been paid to the clinical application of machine learning (ML) and DL1,2 and commercial software has been …

Mechanisms Of Kcnq1-Calmodulin Channel Complex Activation To Dual Open States, Po Wei Kang

McKelvey School of Engineering Graduate Student Theses & Dissertations

Voltage-gated potassium channels (KV) facilitate cellular excitability essential for cardiac contraction and neuronal spiking. In response to changes in the membrane potential, KV channels conduct transmembrane K+ flux to terminate action potentials. Myriad drugs target KV channels to achieve therapeutic goals such as anti-arrhythmic and anti-convulsant effects. Owing to their physiological significance, KV channels are exquisitely regulated by a cohort of auxiliary proteins and signaling molecules. The KCNQ1 or KV7.1 voltage-gated potassium channel stands out as an archetype in this regard. Modulators such as the KCNE subunit family, membrane lipid PIP2, and calmodulin (CaM) enable KCNQ1 to conduct phenotypically distinct …

Ai-Driven Neuro-Oncology Imaging Analysis Of Intracranial Tumors, Satrajit Chakrabarty

McKelvey School of Engineering Graduate Student Theses & Dissertations

According to 2016 cancer statistics, brain tumors are the leading cause of cancer-related morbidity and mortality around the world. More than 100 types of brain tumors, distinguishable by unique histopathological features, have been identified that differ significantly in prognosis and treatment strategies. Currently, histopathology is the diagnostic standard for characterizing brain tumors, which carries risks and potential complications. So, MRI is frequently used as an alternative to or in conjunction with histopathology due to its non-invasive nature and high soft-tissue contrast. With the emergence of artificial intelligence (AI) based approaches, different machine learning (ML) and deep learning (DL) models have …

Modulation Of Penetration And Interphase Dynamics For Stable Lithium Batteries, Youngju Lee

McKelvey School of Engineering Graduate Student Theses & Dissertations

Li metal batteries are promising candidates for next-generation rechargeable batteries due to their high energy density and low redox potential. However, dendritic Li growth which can easily lead to short-circuit and cell failures have been a major hurdle that prevented the commercialization of metal anodes. The onset of dendrites occurs when the salt concentration gets depleted at the electrode surface, which can be predicted by the classical dilute solution theory and Sand’s equation. However, deviations from the theoretical predictions were often reported, and the dendrites were still observed in current densities orders of magnitude lower than the system-defined limiting current …

Colorectal And Ovarian Cancer Imaging Using Optical Coherence Tomography And Photoacoustic Tomography, Hongbo Luo

McKelvey School of Engineering Graduate Student Theses & Dissertations

Human colorectal cancer and ovarian cancer remain the leading causes of cancer mortality. However, if diagnosed early, both cancer types have high patient survival rate. Endoscopic screening and surveillance are particularly helpful in providing physicians with in situ information for diagnosis. Compared to traditional ultrasound endoscopy, optical endoscopy can provide higher imaging resolution with more detailed tissue morphology. Also, differences in tissue optical properties (e.g., scattering, absorption) can also by revealed with optical imaging methods. In this thesis, two optical coherence tomography (OCT) catheters are introduced for ex vivo sample studies and in vivo patient studies to differentiate malignant tumor …

Cancer Assessment With Optical Imaging: Algorithms And Machine Learning, Shuying Li

McKelvey School of Engineering Graduate Student Theses & Dissertations

Optical imaging uses non-ionizing radiation to provide information about the optical absorption and scattering of biological tissues, which are associated with tumor development and progression. This dissertation focuses on algorithmic and machine-learning-based approaches that facilitate the clinical adoption of optical imaging in cancer assessment.Breast cancer is the most diagnosed cancer among women in the United States. However, traditional diagnostic tools, such as X-ray mammography, ultrasound (US), and magnetic resonance imaging (MRI) have low specificity, resulting in numerous unnecessary benign biopsies. Diffuse optical tomography (DOT) has been explored as a potential solution to assist in breast cancer diagnosis and reduce the …

Explore Ultra-High Q Optical Whispering-Gallery-Mode Resonators: Innovative Sensing Capabilities And Broad Applications, Jie Liao

McKelvey School of Engineering Graduate Student Theses & Dissertations

Optical whispering-gallery-mode (WGM) microresonators can trap photons within a small volume for extended periods, resulting in strong interactions between light and matter. This makes WGM resonators an ideal platform for photonic sensors. In the past, WGM resonator sensors relied on analyzing and monitoring a single resonance. This thesis focuses on the study of optical WGM microresonators with ultra-high Q-factors and their diverse sensing applications exploiting multimode resonances. By leveraging the unique spectral properties of WGM resonators and the principles of information theory, the thesis aims to extract maximum information from the sensor’s signal to improve sensing performance and capabilities.The thesis …

Tfa Inference: Using Mathematical Modeling Of Gene Expression Data To Infer The Activity Of Transcription Factors, Cynthia Ma

McKelvey School of Engineering Graduate Student Theses & Dissertations

Transcription factors (TFs) are a set of proteins that play a key role in the information processing system that enables a cell to respond to changes in internal and external state. By binding near a gene in a cell’s DNA, a TF can influence that gene’s expression level, triggering the appropriate increase or decrease in production levels of proteins that are needed to handle stressors like a change in nutrient availability or damage to the cell’s internal structures. Transcription factor activity (TFA) is a measure of how much effect a TF has on its target genes in a given sample …

Interfacial Kinetics And Morphological Instabilities Of Alkali Metal Anodes, Bingyuan Ma

McKelvey School of Engineering Graduate Student Theses & Dissertations

Rechargeable alkali metal anodes hold the promise to significantly increase the energy density of current battery technologies. But they are plagued by dendritic growths and solid-electrolyte interphase (SEI) layers that undermine the battery safety and cycle life. The path towards high-performance alkali metal batteries requires multiscale understanding and optimization, from nanoscale solvation structure, microscale morphology and interphase chemistry, to macroscale metal-separator interaction. This dissertation focuses on the following aspects: (i) dynamic electrode|electrolyte interfacial stability optimization for stable anode-free alkali metal batteries, (ii) electrode|separator interfacial stability analysis of the stabilized metal anodes, (iii) Li plating in commercial Li-ion batteries at extreme …

Moment And Heuristic Approaches To Control And Stabilization Of Dynamic Populations, Xin Ning

McKelvey School of Engineering Graduate Student Theses & Dissertations

Large-scale dynamic population systems are prevalent in diverse domains at different scales from cell to society. Robust control of such ensemble systems, consisting of a vast collection of isolated or interconnected, nearly identical dynamical units is compelling to enable many cutting-edge applications, such as nuclear magnetic resonance spectroscopy and imaging in quantum control, pinning control in complex networks, and neural stimulation in brain medicine. Dynamic network systems, which consist of coupled dynamical systems, are also frequently investigated for emerging problems in finance, social science, public health, and more. In this thesis, moment-based and heuristic approaches are proposed for systems-theoretic analysis …

Development Of Image-Segmentation And Objective Evaluation Methods For Quantitative Nuclear-Medicine Imaging, Ziping Liu

McKelvey School of Engineering Graduate Student Theses & Dissertations

Nuclear-medicine imaging (PET/SPECT) provides a mechanism to study in vivo physiological processes in the human body and, thus, plays an important role in cancer diagnosis, prognosis, and treatment. Recently, quantitative imaging (QI), i.e., the procedure to extract numerical or statistical features from medical images, is showing substantial promise in multiple clinical applications. However, clinical advancement of quantitative PET/SPECT requires methods for improved quantification from images and techniques to objectively evaluate these methods with patient data. The goal of this dissertation is to develop methodologies to fulfill these needs.Reliable quantification of features (metabolic tumor volume, regional tracer uptake) from PET/SPECT images …

The Development Of Metabolic Models And Machine Learning Methods For Biofuel-Producing Bacteria, Garrett Roell

McKelvey School of Engineering Graduate Student Theses & Dissertations

Humans have harnessed the power of microbiology for the creation of fermented foods for over 10,000 years. Synthetic biology now allows chemicals and materials that historically were derived from fossil fuels to be produced via fermentation. These technologies will lessen our dependence on nonrenewable resources and reduce the rate CO2 is released into the atmosphere. Non-model microbial species with unique metabolic traits are promising solutions for bioproduction from renewable feedstocks including lignin, agricultural waste, and CO2. Systems biology characterizations, metabolic models, and data-driven approaches are essential for deciphering and optimizing cell metabolism for these applications.Rhodococcus opacus, R. opacus, is an …

Injective Mapping Under Constraints, Xingyi Du

McKelvey School of Engineering Graduate Student Theses & Dissertations

Injective mapping of meshes is a fundamental yet long-standing problem in computer graphics. The problem is particularly challenging in the presence of constraints, as an injective initial map is often not available. In this dissertation, we propose methods for computing injective maps that satisfy constraints without the need for an injective initial map. Our key contribution is a family of novel energies that are smoothly defined for arbitrary maps (injective or non-injective) and that promote injectivity when minimized. We first introduce the Total Lifted Content (TLC) energy for mapping 2D and 3D meshes into target domains with constrained boundaries. Next, …

An Efficient Task-Parallel Platform For Interactive Applications, Kyle Singer

McKelvey School of Engineering Graduate Student Theses & Dissertations

The performance growth of individual processor cores has been slowed by physical limitations that are difficult to overcome; as such, the trend has been for processor manufacturers to instead increase the number of processing cores in a single chip. As a result, programmers must write parallel code to gain performance. While prior work on task-parallel platforms has simplified the process of writing throughput-oriented scientific applications, they have paid little attention to the types of workloads seen in modern interactive applications such as web services.

Interactive applications are programs which interact with users or other programs using input/output (I/O) operations and …

Chronic Prestress Regulation Of Cardiomyocyte Electrophysiology, Daniel Simmons

McKelvey School of Engineering Graduate Student Theses & Dissertations

Mechanical loading plays critical roles in regulating heart and cardiomyocyte physiology, with volume overload playing an important role in healthy cardiac development from the fetus to adulthood. Inhibition of mechanical loading during fetal development leads to weak and dysfunctional heart chambers, hypertension, organ failure, and even fetal death in severe cases. Increased loading on the heart has also been clinically shown to increase arrhythmia risk in athletes, including those with no genetic predisposition.Clinical observations studying the effects of cardiac genetic disorders may reveal the mechanisms in which mechanical loading influences cardiomyocyte development. Genetic mutations affecting plakophilin-2 (PKP2), a protein in …

Herbicide-Amine Salt Volatilization From The Solution And Dried Residue, Stephen M. Sharkey

McKelvey School of Engineering Graduate Student Theses & Dissertations

Due to the increasing reports of glyphosate resistant weeds, genetically modified (GM) glyphosate tolerant crops have been supplemented with crops resistant to alternative herbicides such as dicamba (3,6-dichloro-2-methoxybenzoic acid) and 2,4-D (2,4-dichlorophenoxyacetic acid). Since the introduction of dicamba and 2,4-D tolerant crops in the U.S., there has been a substantial increase in reports of off-target dicamba and 2,4-D movement and damage (i.e., drift) to non-tolerant vegetation and crops. The off-target damage has been documented by U.S. news sources and increased complaints to State Departments of Agriculture. These complaints have been especially high for dicamba, with the U.S. Environmental Protection Agency …

Models And Algorithms For Real-Time Systems, Abhishek Singh

McKelvey School of Engineering Graduate Student Theses & Dissertations

We apply the framework of parameterized algorithms to the FP (fixed-priority) and EDF (earliest- deadline-first) scheduling of recurrent (periodic and sporadic) task systems in a preemptive uniprocessor environment, which are fundamental models in real-time scheduling theory. We identify parameters for these models that allow the models to be analyzed efficiently; reciprocally, we pick out parameters for the models that should not lead to efficient analysis, in the absence of major algorithmic breakthroughs.Fixed-point iteration algorithms like RTA (response time analysis) and QPA (quick processor-demand analysis) are arguably the most popular ways of solving some of the fundamental problems mentioned above. We …

Throughput Optimizations For Irregular Dataflow Streaming Applications On Wide-Simd Architectures, Stephen William Timcheck

McKelvey School of Engineering Graduate Student Theses & Dissertations

Streaming dataflow applications exist in numerous fields of study including, but not limited to, bio-sequence analysis, astrophysics, network packet analysis, and data integration. The inputs for these applications are independent of one another, making the problems prime candidates for parallel acceleration using wide-SIMD vector processors, such as graphics cards. However, many of these streaming applications exhibit irregular dataflow, where the number of outputs per input from any computation within the application is data dependent and unknown a priori. To ameliorate the throughput issues caused by irregular dataflow on wide-SIMD systems, we utilize a framework such as MERCATOR, which queues data …

Structural Understanding Of Glass-Forming Metallic Alloys, Porter Weeks

McKelvey School of Engineering Graduate Student Theses & Dissertations

Despite intense interest in the discovery and design of metallic glasses, the efficient a priori identification of novel glass-formers without the need for time-consuming experimental characterization has remained an unattained goal due to the inability to fully understand why certain metallic alloys are better glass-formers than others. Using geometric alignment and density-based clustering algorithms to quantitatively describe the short-range atomic structure in simulated metallic liquids reveals that each liquid is comprised of a surprisingly small number of geometrically similar atomic clusters (6-8 characteristic motifs) with the variance in the population distribution (relative sizes of the associated groups) providing an a …

Nalcn-Encoded Sodium Leak Currents In The Regulation Of Daily Rhythms In The Excitability Of Neurons In The Suprachiasmatic Nucleus, Nien-Du Yang

McKelvey School of Engineering Graduate Student Theses & Dissertations

Circadian rhythms have evolved to enable organisms to adapt to daily (24 hour) changes in the environment. In mammals, including humans, the suprachiasmatic nucleus (SCN), located in the hypothalamus, is the master circadian pacemaker that governs daily rhythms in physiology and behavior. It is known that neurons in the SCN generate circadian changes in spontaneous repetitive firing rates (higher during the day than at night) that regulate and synchronize daily physiological and behavioral rhythms, although the critical ionic conductances driving these rhythms remain elusive. Considerable evidence suggests that the daily rhythms in the repetitive firing rates of SCN neurons are …

Ultrasound Guided Diffuse Optical Tomography: Better Image Quality Enables Better Classification, Menghao Zhang

McKelvey School of Engineering Graduate Student Theses & Dissertations

According to the American Cancer Society, breast cancer incidence rates have risen in the past four decades, and breast cancer has become the second most diagnosed cancer in the US women. Ultrasound (US)-guided diffuse optical tomography (DOT), a promising non-invasive functional imaging technique for diagnosing breast cancer and monitoring breast cancer treatment response, has been applied in clinical studies over the past two decades. By utilizing multiple wavelengths in the near-infrared range (NIR), USDOT can provide quantitative estimations of functional information related to tumor angiogenesis, including oxygenated and deoxygenated hemoglobin, and total hemoglobin concentration. However, due to the extensive scattering …