Theses and Dissertations from UMD
Permanent URI for this communityhttp://hdl.handle.net/1903/2
New submissions to the thesis/dissertation collections are added automatically as they are received from the Graduate School. Currently, the Graduate School deposits all theses and dissertations from a given semester after the official graduation date. This means that there may be up to a 4 month delay in the appearance of a give thesis/dissertation in DRUM
More information is available at Theses and Dissertations at University of Maryland Libraries.
Browse
3 results
Search Results
Item MATHEMATICAL MODELS AND NOVEL BIOMARKERS TOWARD OPTIMIZATION OF BURN INJURY RESUSCITATION(2022) Arabidarrehdor, Ghazal; Hahn, Jin-Oh; Mechanical Engineering; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Extensive burn injury is not only devastating but also a significant challenge for healthcare providers. Following a chain of inflammatory responses post-burn, significant amounts of plasma shift from the vascular compartment into the tissues, simultaneously posing the risks of hypovolemic shock and edema. Standard burn resuscitation protocols aim to replace the lost blood volume while not exacerbating the edema through hourly-titrated intravenous fluid infusion. Due to the significant variability in treatment efficacy, there is a substantial ongoing effort to optimize and individualize the burn resuscitation protocols. In this work, we aim to contribute to this effort by (i) developing a platform for the virtual evaluation of burn resuscitation protocols and (ii) identifying biomarkers to guide fluid resuscitation effectively. The first part of this work presents a mathematical model of burn injury and resuscitation, which can be used for the development and non-clinical testing of burn resuscitation protocols and algorithms, as well as to garner knowledge and intuition into this complex pathophysiology. Our mathematical model consists of a multi-compartmental model of blood volume kinetics, a hybrid mechanistic-phenomenological model of kidney function, and novel lumped-parameter models of burn-induced perturbations in volume kinetics and renal function. We examined our mathematical model’s prediction accuracy and reliability using a rich dataset from 16 sheep with extensive burn injuries and clinical data from 233 real-world burn patients. The second part of this work presents the expansion of the mathematical model to incorporate the cardiovascular and renin-angiotensin-aldosterone systems, as well as detailed descriptions of the kidney’s mechanisms, particularly regarding its blood volume and blood pressure regulation roles. This expansion was motivated by the importance of cardiovascular monitoring in the critical care of burn injury patients. We trained and validated the expanded mathematical model for three species: nine sheep subjects and 15 swine subjects with rich cardiovascular and volume kinetics data, and 233 human subjects with demographic and urinary output (UO) data. To the best of our knowledge, our mathematical model may be the first of its kind which is extensively validated for use as a digital twin to replicate realistic burn patients and replace standard large animal pre-clinical testing of burn resuscitation protocols. The third part of this work presents the identification of biomarkers capable of guiding, optimizing, and individualizing burn resuscitation. The UO, the most common endpoint used to titrate burn resuscitation fluid doses, has many limitations as a single variable. Hence, this work aimed to find convenient and reliable biomarkers from arterial blood pressure (ABP) waveform to complement UO in guiding burn resuscitation. Pulse pressure variation (PPV), systolic pressure variation (SPV), and stroke volume variation (SVV) are dynamic indices derived from ABP that have shown promise in hemorrhage resuscitation but are not investigated for different resuscitation paradigms for burn injury. We observed the longitudinal behavior of PPV, SPV, and SVV for 21 porcine subjects with 40% burn injury, which were each either under-resuscitated, adequately resuscitated, or deliberately over-resuscitated. We investigated the features' potential in tracking reference cardiac output (CO) and stroke volume (SV) via linear regression and correlation analysis. PPV, SPV, and SVV showed plausible and statistically different trends for different paradigms. While they performed just as well as UO in tracking CO and SV, their inherent advantage of being available in real-time and their disagreement with UO in determining the subject status suggest that they may potentially complement UO in the hemodynamic assessment of burn patients.Item OBESITY, METABOLIC SYNDROME, AND CARDIOVASCULAR OUTCOMES IN PEDIATRIC KIDNEY TRANSPLANT RECIPIENTS(2017) Sgambat, Kristen; Lei, David; Nutrition; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)Cardiovascular (CV) disease is a leading cause of morbidity amongst children after kidney transplant. The contribution of abdominal obesity and metabolic syndrome (MS) to CV risk is not well defined in this population. A prospective controlled longitudinal cohort study was conducted to investigate contributions of obesity and MS to CV morbidity in a multiracial pediatric kidney transplant population. Aims of the study were to 1) identify prevalence of CV and metabolic abnormalities 2) evaluate effects of obesity and MS on adverse CV outcomes, defined by left ventricular hypertrophy (LVH), impaired myocardial strain and increased carotid intima-media thickness (CIMT) and 3) identify the anthropometric measure of obesity, Body Mass Index (BMI), Waist-to-Height ratio (WHr), or Waist Circumference (WC), that best predicts CV risk. Transplant recipients had standard echocardiographic measures of left ventricular size and function, strain by speckle tracking echocardiography, and CIMT measured at 1, 18, and 30 months post-transplant. 35 pre-transplant echocardiograms were analyzed retrospectively. Multivariate longitudinal regression was used to determine associations of obesity and MS with CV outcomes. Results indicated obesity and MS are prevalent among pediatric kidney transplant recipients. WHr is a more sensitive indicator of obesity-associated adverse CV outcomes compared with BMI or WC, due in part to the prevalence of short stature in this population. Obesity, MS, and hypertension are associated with post-transplant LVH. Significant predictors of impaired longitudinal strain include obesity, hypertension, and a combination of MS with elevated LDL-C cholesterol, whereas higher estimated glomerular filtration rate confers a protective effect. African American pediatric kidney transplant recipients have increased CIMT, which is negatively impacted by MS, whereas the CIMT of non-African American children appears unaffected after transplant. In conclusion, obesity and MS adversely affect CV outcomes in pediatric kidney transplant recipients, highlighting the importance of efforts to maintain healthy weight, blood pressure, and lipid profile after transplant. Further studies are needed to investigate the etiology and consequences of increased CIMT in African American transplant recipients. Imaging techniques such as speckle tracking echocardiography and CIMT may provide a means of detecting subclinical myocardial dysfunction and provide opportunity for early intervention in this population.Item Multiple Testing Procedures for the Analysis of Microarray Data(2013) Nuriely, Ayala; Smith, Paul J; Mathematics; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)We reviewed literature about various multiple testing techniques, especially addressing microarray analyses and small sample sizes, and reanalyzed data from Yuen et al. (Physiological Genomics, 2006) which compared the effect of HgCl2 and Ischemia/Reperfusion injuries on rat kidney tissues. Our analysis uses only 22 rats with small numbers of rats in each treatment group, and 9,501 genes under study. We used empirical Bayes (EB) and permutation testing (implemented in Bioconductor) in an effort to identify differentially expressed genes. EB identified a large number of genes as differentially expressed, including both previously identified and newly identified genes. The newly identified genes appear to have biological functions similar to those previously identified. We also recognized power differences between EB and permutation tests, possibly due to nonnormality of the data but also because permutation tests do not make use of all available information in the data.