Master of Science in Computer Engineering, University of Dayton, 2024, Electrical and Computer Engineering

This study aims to incorporate some advantages of computed tomographic data into the chest X-ray deep lung segmentation paradigm. We do this by training a deep convolutional neural network on chest radiographs (a.k.a. X-rays) with manually drawn ground truth and an identical network on radiographs digitally reconstructed from computed tomographic data with ground truth generated for the given computed tomographic image using an automated morphological 3D lung segmentation algorithm. The resulting twin-network ensemble generates pairs of lung image segmentation labels for chest X-rays: 1) a “traditional” segmentation of the lungs encompassing the apparently low-density tissue and 2) a novel, “full” lung segmentation encompassing an expanded view of the lungs' position in a chest X-ray including those regions obscured by the heart, ribs, and viscera, in essence, a 2D projection of any portion of the 3D lung. These networks perform consistently, with mean Intersection-Over-Union scores of > 90% and > 95%, respectively, across five trials. By subjective analysis, the proposed lung segmentation approach shows satisfactory ability to generalize onto genuine check X-ray images. The proposed technique's high performance and robustness establish a precedent for applying computed tomographic data to automatic chest X-ray segmentation and present an opportunity to further refine existing computer-aided detection and diagnostic tools by considering the full lung.

PhD, University of Cincinnati, 2024, Engineering and Applied Science: Biomedical Engineering

Hyperpolarized 129Xe MRI is a monumental imaging advancement for studying pulmonary structure and function. Following an inhalation of hyperpolarized 129Xe gas, it enables regional quantification of gas diffusion inside the airways, alveolar sacs, interstitial blood air membrane, and the red blood cells. It is often described as the only 4-dimensional pulmonary function test, as it can generate functional images with high spatio-temporal resolution and no radiation. Consequently, it has gained considerable interest from both the lung and MRI communities, and has recently obtained approval by the U.S. Food and Drug Administration as a clinical tool. Despite its power, there are many areas in which hyperpolarized 129Xe MRI can be improved. One of its biggest technical challenges is that its magnetization suffers non-recoverable decay with each excitation. This constrains the sequences to settings that are rapid and efficient with their use of magnetization. Furthermore, its magnetization decay is highly susceptible to regional changes in flip angle, which is problematic when imaging the large thoracic-cavity field-of-view as inhomogeneities in B1 field strength can be severe. From a clinical standpoint, 129Xe MRI is relatively novel, so the true-definitions of healthy metrics are yet to be rigorously explored, especially with regard to demographics such as age. The goal of this dissertation is to meet these needs and deliver a series of technical means that improve hyperpolarized 129Xe MRI and its impact in the clinic. Firstly, we propose a keyhole reconstruction pipeline for 2D spiral MRI that supports the calculation of accurate flip angle maps with minimal additional data cost, and subsequently, analytical B1-inhomogeneity correction. Next, we propose a novel compressed sensing reconstruction framework that incorporates hyperpolarized decay directly into the forward model, subsequently ena (open full item for complete abstract)

Doctor of Philosophy, Case Western Reserve University, 2024, Biomedical Engineering

Ultrasound (US) is safe and low-cost relative to other imaging technologies, making it an increasingly popular diagnostic modality. US contrast is often limited by relatively low differences in acoustic impedance in the body, necessitating the use of contrast agents like microbubbles (MBs; ~1-10 µm diameter) which are intravenously injected and used to compensate for low contrast in conventional B-mode US imaging. The large size of MBs, however, limits their applications to the blood pool. In many diseases, including cancer, information beyond the blood pool is needed for diagnosis and staging. For instance, in many cancers (e.g. prostate, mammary, ovarian etc.), tumors are characterized by high vascular permeability and low lymphatic drainage, which increases the potential for enhanced permeability and retention (EPR) of macromolecules (~200 nm). When present, EPR leads to the tumor-localized accumulation of nano-agents. Nanobubbles (NBs) are new-age submicron bubble agents (100-500 nm diameter) capable of extravasation beyond the vascular network while providing enhanced US contrast similar to MBs. Recently, our group showed that active targeting of NBs to prostate specific membrane antigen (PSMA) rapidly and selectively enhances tumor accumulation and retention. These processes were visualized in real-time with clinical US. This project established NBs as a sensitive detection tool in the diagnosis of PSMA-positive prostate cancer due to localized NB accumulation, reduced diameter and higher number density (NBs per volume) of NBs compared to MBs of similar composition. Together, these points enable: 1) contrast visualization of small capillaries with higher fidelity, and 2) imaging of extravascular cellular targets reached via extravasation of NBs in leaky blood vessels while employing 3) a safe and widely accessible imaging modality. Together, contrast enhanced ultrasound (CEUS) using NBs (NB-CEUS) is a detection method with high biocompatibility and high safe (open full item for complete abstract)

Master of Science (M.S.), University of Dayton, 2024, Aerospace Engineering

This research introduces an Origami-inspired dynamic spacecraft radiator, capable of adjusting heat rejection in response to orbital variations and extreme temperature fluctuations in lunar environments. The research centers around the square twist origami tessellation, an adaptable geometric structure with significant potential for revolutionizing radiative heat control in space. The investigative involves simulations of square twist origami tessellation panels using vector math and algebra. This study examines both a two-dimensional (2- D), infinitely thin tessellation, and a three-dimensional (3-D), rigidly-foldable tessellation, each characterized by an adjustable closure or actuation angle “φ”. Meticulously analyzed the heat loss characteristics of both the 2D and 3D radiators over a 180-degree range of actuation. Utilizing Monte Carlo Ray Tracing and the concept of “view factors”, the study quantifies radiative heat loss, exploring the interplay of emitted, interrupted, and escaped rays as the geometry adapts to various positions. This method allowed for an in-depth understanding of the changing radiative heat loss behavior as the tessellation actuates from fully closed to fully deployed. The findings reveal a significant divergence between the 2D and 3D square twist origami radiators. With an emissivity of 1, the 3D model demonstrated a slower decrease in the ratio of escaped to emitted rays (Ψ) as the closure/actuation angle increased, while the 2D model exhibited a more linear decline. This divergence underscores the superior radiative heat loss control capabilities of the 2D square twist origami geometry, offering a promising turndown ratio of 4.42, validating the model's efficiency and practicality for radiative heat loss control. Further exploration involved both non-rigidly and rigidly foldable radiator models. The non-rigidly foldable geometry, initially a theoretical concept, is realized through 3D modeling and physica (open full item for complete abstract)

Doctor of Education (Ed.D.), Bowling Green State University, 2023, Leadership Studies

The expectation of healthcare professionals is to provide quality, patient-centered care to all patients. Miscommunication between the healthcare team resulted in segmented care and medical errors. As disconnects were discovered, healthcare professionals began promoting a team-based approach to care. The team-based approach helped eliminate barriers that inhibited effective communication and quality care to patients, providing a more cohesive patient care experience. Implementing team-based, patient-centered care in professional practice requires training to be introduced at the educational level of healthcare programs. A teaching strategy called interprofessional education (IPE) was developed to help teach students from different healthcare professions to learn with, from, and about each other's professions. Over time, healthcare education program accreditors were able to integrate IPE recommendations into their learning standards. However, not all healthcare professions chose to include IPE in their educational accreditation standards which left the decision to use IPE in some healthcare programs up to the program director. One healthcare profession in particular, radiography, has not yet mandated IPE into its educational accreditation standards. This study explored if radiography program directors' self-reported attitudes, knowledge, and perceptions of IPE were associated with their self-reported level of use of IPE in their programs. The author created a survey to collect data from radiography program directors accredited by JRCERT (N = 262). Analysis of the data revealed a positive association between program directors' attitudes, knowledge, and perceptions of IPE and their decision to use IPE in their radiography programs. Investigating the relationship between program directors' attitudes, iv knowledge, and perceptions of IPE and their level of use of IPE contributed to an understanding of how educational leaders' make decisions that impact their progra (open full item for complete abstract)

Doctor of Healthcare Administration (D.H.A.), Franklin University, 2023, Health Programs

Deep learning (DL) algorithms are prevalent in radiology as workflow assistants and as modality enhancements. Magnetic resonance imaging (MRI), computerized tomography (CT), diagnostic imaging (DI), ultrasound (US), and positron emission tomography (PET) are modalities that benefit from the DL algorithms and shorter exam times or greater image accuracy. Faster scan time is achieved by the signal to noise ratio (SNR). The distinction is that the technology can enhance images beyond the original resolution from the modality or shorten exam time and rebuild the image quality through SNR algorithms back to approximately the original standard of care (SOC) image. While artificial intelligence signal to noise ratio algorithms (AI-SNR) can enhance an image to a greater accuracy, shorter exam times are a measurable component of a return on investment (ROI) in calculating modality utilization. The algorithm-derived images may have visual variations that are not found on normally acquired original images. The research focused on DL-SNR images on a three-tesla magnetic resonance imaging (3.0T MRI) unit, a high resolution MRI deployment in the industry. The primary research question for this research study is: What are radiologists' perceptions in regard to image quality and increased utilization due to vendor provided DL-SNR on 3.0T MRI? This will be an exploratory qualitative research study using detailed interviews with fellowship-trained radiologists that are using AI-SNR in 3.0T MRI and shortened exam time protocols. Fifteen interviews were conducted. The interview transcriptions were coded using ATLAS.ti to identify common themes and sub-themes in the radiologists' perceptions of DL-SNR imaging. This paper assumes the reader has an adequate understanding of deep learning and radiology processes. The interviews included discussions on key elements on image quality, workflow, reimbursement, legal concerns, and radiologist workload. Issues were identified and potential solut (open full item for complete abstract)

Master of Science, The Ohio State University, 2023, Health and Rehabilitation Sciences

Age related bone loss leading to fragility fractures and osteoporosis has become a known public health issue.1 Currently, much of our understanding of age-related changes is primarily based on areal bone mineral density (aBMD) measured by DXA. While DXA has been shown to predict fracture risk to varying degrees, it is limited due to its 2D technology which is not able to measure aspects of whole bone geometry that contribute greatly to overall bone strength.2 The purpose of this study was to approach the radius, a common site of fracture, as a whole by assessing various skeletal traits and aspects of whole bone geometry to gain a better understanding age-related changes across the sexes and in bones of various size classifications. Quantitative computed tomography analyses were performed on n = 180 ex vivo post-mortem human subject radii from 94 males and 86 females. Skeletal traits such as Tt.Ar, Robustness, Ct.Ar, Ct.Th and vBMD were analyzed across three different sites of the diaphysis of the radius. aBMD and T-score values at the 33% site of the radius were measured for a sub sample (n = 129) which included 78 females and 51 males. Generally, this study found that skeletal traits do vary across the radius, however classifications of whole bone geometry do not tend to change across the radius. Females demonstrated significant decreases in measurements of Ct.Ar, Ct.Th, and vBMD with age (p< 0.001) but little to no relationship with Tt.Ar or robustness (p >0.47) while males demonstrated significant increases in the amount of bone (p < 0.012) but significant decreases in bone mineralization (p <0.019). When examining the relationship between slender and robust individuals within the sex categories, it was found that there were significant differences between slender and robust individuals within a sex category for various variables including Ct.Ar, Ct.Th and vBMD in both males and females (p < 0.044). Lastly, Pearson correlations demonstrated a positive re (open full item for complete abstract)

MS, University of Cincinnati, 2023, Engineering and Applied Science: Biomedical Engineering

Debilitating lung diseases such as chronic obstructive pulmonary disease (COPD) and idiopathic pulmonary fibrosis (IPF) are highly problematic yet have etiologies that are poorly understood. This establishes a role for noninvasive diagnostic techniques that can characterize the structure and function of the lungs. To that end, developments in magnetic resonance imaging (MRI) have shown promise in providing high-resolution images using multiple types of contrast in the lungs without delivering ionizing radiation. MRI in the lungs has traditionally been viewed as a challenge due to rapid apparent transverse relaxation (characterized by the rate constant T2*) accelerated by magnetic susceptibility gradients at the alveolar air-tissue interfaces and a lack of tissue density. Fortunately, the short T2* constraint can be addressed by 1H ultrashort echo time (UTE) sequences which enable imaging at echo times on the order of tens of microseconds. Additionally, imaging exogenous agents such as inhaled hyperpolarized (HP) 129Xe gas overcomes low pulmonary tissue density, offering diffusion-weighted contrast useful for characterizing the lung microstructure. When paired with robust preclinical models such as transgenic models of pulmonary fibrosis, these techniques allow lung MRI to provide important insight into the emergence and etiologies of these chronic diseases. This thesis describes validations and improvements for two methods in preclinical lung MRI. First, we validated retrospectively gated 1H UTE T2* mapping as a method to track disease progression in a mouse model of progressive pulmonary fibrosis. Pulmonary T2* is a novel, quantitative contrast that does not depend on imaging parameters such as coil shading. Additionally, we developed an image processing and analysis methodology that automates the production of whole lung T2* maps from UTE MRI data. Second, we investigated the effects of card (open full item for complete abstract)

Master of Sciences, Case Western Reserve University, 2023, Applied Anatomy

The sympathetic nervous system, a subdivision of the autonomic nervous system, innervates glands, smooth and cardiac muscle of the body and drives the “fight or flight” response. The objective of this study is to use anatomical and radiological methods to definitively identify and investigate the sympathetic chain, specifically ganglia from stellate through T5. The overarching goal of this research is to help guide clinical treatments, including nerve block and ligation procedures, for various disorders of the sympathetic nervous system, including cardiac arrhythmias, hyperhidrosis and pain syndromes. This study uses anatomical methods, including cadaveric dissection, optical tracking, anatomic relationships and landmarks, to investigate the characteristics of the sympathetic chain. This study also uses radiologic methods, including conventional radiography and 3 Tesla (T) magnetic resonance imaging (MRI) with a high-resolution 3D constructive interference in steady state (CISS) sequence, to provide a clinically applicable comparison to gross anatomic observations and measurements.

Doctor of Philosophy, Case Western Reserve University, 2023, Biomedical Engineering

In many chronic inflammatory diseases, the vascular endothelium becomes pathologically permeable due to conditions like angiogenesis and production of growth factors and inflammatory cytokines (e.g., histamine, bradykinin, etc.). In cancer, this process can be exploited for delivery of nanoparticles to tumors via the enhanced permeability and retention (EPR) effect. However, nanoparticle-based therapeutics reliant on the EPR effect have led to inconsistent results in patients. This is due to many factors, with a significant one being heterogeneous tumor vascular architecture and morphology both between patients and within a single tumor. Transport of the nanoparticle to the tumor and into the parenchyma is complicated by uptake by the immune system, ineffective margination, and inefficient extravasation. Guidance is needed to inform clinicians on what therapies may be most effective for each patient. Effective guidance could reduce health-care costs and negative side effects of medication. An inexpensive, safe, non-invasive, and real-time imaging method that has high temporal and spatial resolution may be capable of categorizing the extent of vascular permeability in tumors and once validated, personalize therapeutic regimens for patients. Such a tool could be used not only for tumors, but for all diseases involving pathologically permeable vasculature. With this goal in mind, the objective of this thesis is to work toward development of a real-time method for evaluating vascular permeability over the entire tumor using novel nanobubble (NB)-based contrast-enhanced ultrasound (CEUS). This work builds upon dynamic CEUS protocols used clinically with microbubbles (MBs). NBs, which are 100-400 nm in diameter, are approximately 10x smaller than MBs and have been shown to extravasate into the tumor interstitium. To reach the final objective of this work, NB dynamics from intravenous injection to retention in the tumor must be studied. To this aim, in vitro studies con (open full item for complete abstract)

MS, University of Cincinnati, 2022, Engineering and Applied Science: Electrical Engineering

Liver cancer is a significant public health burden; as of 2020, it is the second leading cause of cancer-related mortality worldwide. Hepatic resection is considered the gold standard for the treatment of liver malignancies. However, this procedure is only possible in a minority of patients, necessitating treatment modalities with comparatively worse performance, such as thermal ablation. Thermal ablation generally results in poorer clinical outcomes relative to resection, with a higher rate of recurrence and the potential for complications related to damage to healthy tissue near the ablation zone. Medical imaging techniques can improve thermal ablation procedures via assistance in preoperative planning, probe placement and postoperative evaluation, but clinicians lack a method to monitor and control thermal ablation while the procedure is ongoing. Echo decorrelation imaging is a pulse-echo ultrasound imaging technique that measures stochastic variations in echo signals arising from thermal treatment. The method has been shown to accurately predict thermal lesioning in in vivo and ex vivo studies of thermal ablation using conventional 2D ultrasound imaging. This thesis aims to apply the echo decorrelation methodology to volumetric ultrasound data to control RFA procedures in real-time. Feedback control is implemented as a bang-bang type controller that automatically stops thermal treatment if the spatial mean of the cumulative decorrelation map exceeds a set threshold. 3D echo decorrelation-based control was evaluated through a series of feedback- controlled and uncontrolled ablation trials on ex vivo bovine liver tissue using a clinical RFA system. The RFA system was set to target a 15 mm radius spherical region of tissue while decorrelation maps were computed from captured volumetric ultrasound data; if the control criterion was met, the procedure was automatically stopped using a custom- designed microcontroller circuit. Trials were divided into two groups (open full item for complete abstract)

Doctor of Philosophy, Case Western Reserve University, 0, Molecular Medicine

Novel therapeutic regimens which spur the endogenous immune system to kill cancer cells, such as stereotactic radiosurgery (SRS) and immune checkpoint inhibition (ICI), are heterogeneously effective. Understanding causal factors of response is vital to guide risk assessment and treatment decisions. In this thesis, I evaluate the ability of three methods to prognosticate survival for brain metastases patients following SRS and ICI treatment. These include the clinically utilized response assessment in neuro-oncology for brain metastases (RANO-BM) protocol, as well as investigational computational methods such as radiomic feature analysis and convolutional neural network (CNN) image analysis. I find that easing the 10mm RANO-BM diameter threshold for measurable disease allows new lesions to be discovered as proof of progression in ICI-treated metastases. Further, I find that the trajectory of RANO-BM diameter can be more instructive for risk prediction than the ratio-change and that neither volume nor number of metastases, nor RANO-BM diameter can significantly predict survival until a year after treatment. Reproducing common radiomic methodology flaws observed in the published literature, I demonstrate that inconsistent partitioning, or the improper division of radiomic feature data into Training, Validation, Test, and External test sets, can provide a 1.4x performance boost to reported accuracy (AUROC) for predictive models. Additionally, I highlight how spurious correlations with biological variables can overstate the importance of radiomic features. Leveraging the conclusions from my radiomic reproduction study, I assess the ability of radiomic features and convolutional neural networks (CNNs) to predict overall survival in the largest ICI-treated brain metastases cohort assembled to date, comprising 175 patients from three institutions in two countries. I find that neither radiomic features nor any architecture of the survival AI model MetsSurv is capable of p (open full item for complete abstract)

Master of Science, The Ohio State University, 2022, Health and Rehabilitation Sciences

Cardiovascular disease (CVD) is a group of disorders affecting the heart and blood vessels. This insidious pathology is the leading cause of death in the United States, accounting for 868,662 deaths in 2017. The prevalence of this disease is expected to increase, with 45.1% of the population expected to have some form of CVD by 2035. Aside from the growing health concern, CVD is also the costliest chronic disease in the country, projected to hit 1.1 trillion dollars in total cost by 2035. This information underscores the importance of advancing CVD detection and primary prevention. Current CVD risk assessment usually relies on clinical prediction models that estimate a patient's risk of having a CVD event in the future. The most recent recommendation for assessing risk of CVD in asymptomatic populations, is the use of the pooled cohort equations (PCE) atherosclerotic cardiovascular disease (ASCVD) risk estimator. Many patients who are evaluated using these clinical prediction models end up needing a more refined risk assessment to develop the most appropriate care plan. For this purpose, coronary artery calcium (CAC) scoring with computed tomography is the most widely used. While these tools are well validated, they are not without limitations. The need for new, non-invasive, accessible, relatively inexpensive, and low-risk CVD assessment tools is vital to further improve detection and prevention. This project explored the use of abdominal aortic sonography for use as a CVD risk assessment tool. Participants provided their imaging and health data in order to both evaluate the feasibility of using sonography to assess atherosclerotic plaque burden in the abdominal aorta, and explore associations between the imaging data and traditional CVD risk factors. After developing an imaging protocol and novel grading system, abdominal aortic sonography was proven to be a reliable, and practical method of measuring atherosclerotic burden in the inferior porti (open full item for complete abstract)

Master of Science, The Ohio State University, 2022, Nuclear Engineering

In this thesis, an algorithm addresses a simple method to help the fusion of different radiation imaging modalities into one image. A computational program was utilized to achieve this goal by retaining the structural and compositional information in the final fused image. It appears to be a promising method to gather as many details as possible from different imaging modalities, such as neutron and X-ray images. The essential information on each imaging modality can be comprehensive and vital for the final assessment. This thesis focuses on the algorithm used to achieve the primary goal of complete data from each involved imaging modality. Fusing two images can have significant benefits in industrial and medical imaging, such as non-destructive testing applications and helping to diagnose certain diseases. A similar procedure was performed by another research group earlier to achieve similar kinds of objectives [1][2][3][4]. It is vital to have this tool available to be helpful for the correct assessment of images in several image fusion applications. It will solve the issue of incomplete information in an image by merely one imaging method. This algorithm is based on discrete wavelet transform, which shows superiority over other fusion methods as it does not lose quality during the process and keeps the spatial resolution of the images [5]. Various materials have similar atomic numbers and differing thermal cross-sections, or vice versa. This could cause confusion in the image understanding; therefore, this will be solved by the fusion of neutron and X-ray, which will provide a more convenient and comprehensive image to be analyzed by limiting the shortcomings of both imaging methods. This algorithm can also be applied to other radiation imaging modules. After obtaining the fused image, quality assessment techniques were performed and compared to previous work results to validate our algorithm.

Doctor of Philosophy, The Ohio State University, 2022, Health and Rehabilitation Sciences

Cardiovascular disease (CVD) proceeds to be the leading cause of death in the United States. Policy and intervention strategies focus on maintaining cardiovascular health in adulthood, but little focus is on childhood cardiovascular health and maintaining healthy behaviors to prevent CVD. While childhood obesity continues to rise above epidemic proportions, little research exists on evaluating, preventing, and detecting cardiovascular disease in children. Even less research is available on prepubescent children, when lifestyles, behaviors and habits are forming. Additionally, comprehensive cardiovascular health status among children is limited due to the complexity of the development of CVD over time. Due to the lack of consistency among screening protocols, few research studies have utilized comprehensive risk assessments. In order to better address the complexity of cardiovascular health in children, a more comprehensive and holistic approach is needed to reduce the risk of CVD and provide primordial care to this age group. Furthermore, innovative approaches in risk assessment are warranted to best individualize risk assessment and improve overall cardiovascular risk profiles. Providing early screening assessments for vulnerable populations, such as children, could provide empirical evidence for the promotion of earlier prevention measures. In addition to probable screening during pediatrician well visits, non-invasive techniques to fully gauge CVD risk should be considered. The use of sonography to quantitatively assess carotid intima media thickness (CIMT) and abdominal adiposity, could be an ideal non-invasive, non-ionizing, portable and relatively inexpensive method to add onto current risk assessment measures. The review of the current literature provided substantial evidence for continued improvement of CVD risk assessment in children. Although the existing literature has substantial gaps, it is evident that risk profiles could be improved through compre (open full item for complete abstract)

Doctor of Philosophy, Case Western Reserve University, 2022, Biomedical Engineering

Most of the current AI methods are based on a single modality or based of a single biomarker. However, information across different modalities and scales may hold complementary information and integrating them together may enhance the performance of AI models. In this dissertation we introduced new AI biomarkers and fusion strategies to combine biomarkers across multiple modalities and scales for disease risk stratification. First, we used population atlas-based models to find regions of possible organ deformations between patients with mild and severe disease. In addition, we used a multi- scale approach with deep learning (DL) to exploit contextual information from peri- tumoral (PT) regions (region immediately surround the visible tumor) for disease risk stratification. Furthermore, we developed fusion strategies to integrate DL on imaging with clinical parameters into novel nomograms. Besides, we investigated if computer extracted features from different modalities such as radiology and histopathology have added benefit for patient outcome prediction over their individual counterparts. Finally, we stress tested DL approaches to assess their robustness to slight variations in input parameters using test-retest repeatability analysis. We evaluated these approaches on two use cases, namely; prostate cancer (PCa) risk stratification and COVID-19 prognosis. In the context of PCa, due to the lack of accurate risk stratification tools many patients are over diagnosed and over treated. Therefore, we developed PCa risk stratification models and demonstrated the added benefit of; 1) DL signatures from PT regions on MRI and 2) the DL integrated imaging and clinical nomogram. Additionally, we demonstrated that combined signatures from prostate MRI and whole slide images and can better predict PCa outcome as compared to the individual counterparts. Furthermore, test-retest repeatability analysis revealed that DL was repeatable for simple classification tasks but was (open full item for complete abstract)

Doctor of Philosophy, Case Western Reserve University, 2022, Biomedical Engineering

MRI is a non-invasive medical imaging modality with many clinical applications in major body regions. To generate an MRI image, a RF pulse is first generated through the RF transmit hardware, and the resulting MR signal is subsequently acquired through RF receive hardware. While there are many major advances in RF receive chain, RF transmit chain development has been largely stagnant. Outside of conventionalRF pulses for exciting magnetization, the RF transmit signals could have clinical applications to stimulate RF technology development. This dissertation focuses on exploring novel applications and technological development of the RF transmit chain to influence MRI acquisition. The following projects compose this exploration, which aims to stimulate new developments in RF technologies or improve current implementations. First, the RF transmit chain is used as a constant RF signal to create a motion navigator called Pilot tone (PT) navigator. This navigator is integrated with a fast simultaneous quantification method calledMagnetic Resonance Fingerprinting (MRF) to create 2D and 3D free-breathing T1 relaxation times, T2 relaxation times, and proton density (M0) maps. This study is a first step to address difficulties associated with quantitative abdominal imaging due to motion and long scan times from conventional quantitative methods. Second, the PT navigator is integrated with a quadratic RF (qRF) based MRF sequence for a 2D free-breathing simultaneous quantification of T1 relaxation times, T2 relaxation times, T2* relaxation times, and fat fraction (FF). This study provides a more comprehensive 2D free-breathing abdominal technique. Third, an old RF engineering concept called outphasing, is introduced to integrate with current hardware to improve the dynamic range of small amplitude pulses. This control technique serves as the cornerstone of other projects with future digital compatibility and high controllability, which will allow gene (open full item for complete abstract)

PHD, Kent State University, 2022, College of Arts and Sciences / School of Biomedical Sciences

The human ilium is significantly shorter and broader than are those of all other primates. In addition, it exhibits an anterior inferior iliac spine that emerges via a secondary center of ossification. It is also unique to hominins. Here we track the ontogeny of the ilium in human and subadult primate ossa coxae. We find that its ontogeny is exclusive among primates from anlagen to adulthood and that the fusion of the anterior inferior iliac spine is a capstone event of a unique growth process that repositions the anterior gluteal muscles for control of pelvic drop during upright walking. This novel growth process is therefore a hominin synapomorphy that can be used to assess the presence of bipedal locomotion in extinct taxa. We recently reported that a unique physis modulates broadening of the hominin ilium and shortening of its isthmus. We report here the discovery of a large, constant vascular foramen which lies close to the novel growth plate and serves as a central structure in the hominin ilium's vascular network. No likely homologues appear in Old World Monkeys but are sometimes present in African great ape pelves. However, the human foramen (the Anterior Iliac Foramen) is significantly larger than the same individual's nutrient foramen, and when corrected for body size, the human anterior iliac foramen is substantially larger than are those of apes. Those of Pan and Gorilla do not differ significantly from one another when so corrected, establishing that a small foramen is primitive and that its enlarged state is derived in hominins. This likely reflects amplification of the blood supply to the novel hominin physis during growth. Its presence in hominin fossil ilia can therefore provide evidence of iliac ontogenetic specialization for bipedality. The unique presence of this synapomorphy provides robust evidence that non-saltatory bipedality is a singular adaptation restricted to hominins, and that it has occurred only once in known primates.

Master of Science in Biomedical Engineering, Cleveland State University, 2021, Washkewicz College of Engineering

The usage of radiomics for extracting high-dimensional features from radiographic imaging to quantify subtle changes in tissue structure and heterogeneity has shown great potential for disease diagnosis and prognosis. However, radiomic features are known to be impacted by acquisition-related changes (e.g., dose and reconstruction variations in CT scans) as well as technical variations between cohorts (i.e., batch effects due to varying dosage and tube currents). Using features which are not resilient to such imaging variations can result in poor performance of the downstream radiomics classifier models. In this study, we present a framework to systematically identify radiomic features that are resilient to both batch effects and acquisition differences, as well as evaluate the impact of such variations on radiomic model performance. We demonstrate the utility of our approach in the context of distinguishing active Crohn's disease (CD) from healthy controls using a uniquely accrued cohort of 164 CTE scans accrued from a single institution, which included (a) batch effects due to variations in effective dosage and tube current, as well as (b) scans simultaneously acquired at multiple doses and reconstructions (3 variations per patient). Our framework involves systematically evaluating the impact of acquisition variations (based on feature robustness to explicit dose/acquisition changes) and batch effects (based on feature stability to implicit dosage/current variations). Resilient radiomic features identified after accounting for both types of variations yielded the best random forest classifier performance across both discovery (AUC=0.819 ± 0.043) and validation (AUC=0.787) cohorts when using full-dose images; also found to be significantly more generalizable than features that were not optimized for such variations (AUC=0.419 in validation). This subset of radiomic features that were both robust and stable (resilient) also maintained their performance when evaluate (open full item for complete abstract)

PHD, Kent State University, 2021, College of Public Health

Title: Transjugular Intrahepatic Portosystemic Shunt (TIPS), Duration of Procedural Time and Correlation with Early Morbidity and Mortality Author: Juan Carlos Barrera Gutierrez, PhD., Candidate in Epidemiology ABSTRACT Purpose: 1. To examine the relationship between patient factors (demographics or clinical) and duration of TIPS procedure. 2. To determine if duration of TIPS procedure is associated with early morbidity, including acute kidney injury, liver dysfunction, and intraoperative or postoperative bleeding. 3. Determine if duration of TIPS procedure is related to early mortality (or operative mortality). Materials and Methods: Data for this retrospective study was abstracted using the REDcap system from patients' electronic records. Inclusion criteria were patients over 18 years of age who underwent an initial TIPS procedure (for any indication) between January 2005 and August 2020. Exclusion criteria were those TIPS performed out of the institution and failed TIPS procedures. Regression analysis was used to identify the predictors of procedural time, the predictors of morbidity and mortality. Results: The mean age at TIPS procedure was 57 years. Most patients (70%) were male, and Non-Hispanic whites made up the largest group (80.5%). The main predictor of procedural time was baseline bilirubin (beta coefficient: 4.9 minutes, p = 0.005) in those patients that did not require extra access. Additionally etiology of cirrhosis and staff experience were predictors of procedural time. The main predictor of acute kidney injury (AKI) immediately after TIPS were gender (p = 0.02) and body mass index (BMI) (p = 0.04). MELD score (p = 0.02), age (p = 0.003), gender (p = 0.001), and time (p = 0.01) became the best predictors of liver function post TIPS in those patients with ascites and who did not require extra access. The mortality rate in the first month after TIPS was 14.5%, and the predictors associate to this were MELD score (OR: 1.17, p=0.004 (open full item for complete abstract)