Search

Background: According to the 2019 CDC Antibiotic Resistance Threats Report, more than 2.8 million antibiotic-resistant infections occur in the United States each year, leading to more than 35,000 deaths. Among the most urgent threats identified by the CDC are carbapenem-resistant Enterobacterales (CRE). Despite efforts to control the spread of these organisms, the number of estimated cases between 2012 and 2017 remained stable. In 2017, an estimated 13,100 hospitalized cases of CRE led to approximately 1,100 deaths and $130 million attributable healthcare costs. This dissertation seeks to address this issue from both a pharmacokinetic/pharmacodynamic and epidemiological perspective. Methods: We evaluated the susceptibility of 140 CRE clinical isolates against novel agents eravacycline and plazomicin using techniques standardized by the Clinical and Laboratory Standards Institute. We performed in-vitro static time-kill assays in 8 Verona Integron-encoded metallo-beta-lactamase (VIM)-producing CRE using single and combination exposures of cefepime, meropenem, piperacillin/tazobactam, amikacin, and plazomicin along with aztreonam and aztreonam/avibactam. Additionally, we performed a 10-year, inverse probability of treatment weighting adjusted retrospective cohort study comparing the risk in observing a composite outcome of all-cause mortality or discharge to hospice in patients having CRE vs. carbapenem-susceptible Enterobacterales (CSE) infections after 14 and 30 days. In this cohort, we also reported on the prevalence of CRE across the decade. Additionally, we compared the organism composition and susceptibilities of isolates cultured in both the CRE and CSE groups. Results: Plazomicin showed higher susceptibility than eravacycline against our CRE isolates. In time kill studies, plazomicin was bactericidal against 5/8 isolates as monotherapy. Meropenem/amikacin or meropenem/plazomicin were bactericidal in all experiments, except for one isolate which regrew against meropenem/plazomicin. Aztreonam/avibactam was bactericidal in all experiments tested. Neither cefepime nor piperacillin/tazobactam improved the activity of plazomicin against our isolates. Cefepime with amikacin showed inconsistent activity. In the retrospective cohort study, the overall incidence of CRE infections was 1.8%. CRE isolates exhibited higher resistance across all routinely tested antimicrobials classes compared to CSE. The CRE population appeared to be largely non-carbapenemase-producing given the high susceptibility of meropenem and the high prevalence of E. cloacae, a known AmpC-producer. Overall, the risk of composite outcome only appeared to be increased among patients with a bloodstream infection on the index date and could only be assessed when utilizing an exposure of carbapenem-non-susceptible Enterobacterales (CNSE) due to insufficient sample size. However, the results were inconclusive as they were not statistically significant. Conclusions: Novel antimicrobial agents plazomicin and aztreonam/avibactam were highly active against a collection of CRE including both Klebsiella pneumoniae carbapenemase (KPC) and VIM. Aztreonam/avibactam, meropenem/amikacin, and meropenem/plazomicin all exhibited comparably bactericidal activity. Furthermore, at an academic medical center in a non-endemic region for CRE, it appears that CRE infection may have increased the risk of experiencing the composite outcome after both 14 and 30 days, but definitive conclusions may not be drawn given the lack of statistical significance and imprecision in the estimation of the effect. The difficulties in drawing definitive conclusions from this study owing to limited sample size in the CRE or CNSE group stresses the importance of developing novel strategies and performing larger, multicenter studies when investigating highly resistant infections with low prevalence.

The placement of cryogenic fuel/propellant depot stations in Earth orbit has the potential to transform the nature and operations for many types of spaceflight missions. Today, spaceflight missions are almost universally required to carry the entire amount of fuel required for the mission, for the entire duration of the mission, from the point of launch. This is the rough equivalent of making a drive from Ohio to California, requiring the traveler to bring along the total sum of gasoline required for the entire trip, without being able to `fill-up’ anywhere along the route. Obviously, this framework of travel greatly encumbers the breadth, scope, and efficiency of potential journeys. Cryogenic fuel/propellant depots have not been implemented because many technical, operational, and engineering challenges still exist. These must be overcome prior to the placement of usable on-orbit propellant depots. This thesis investigates three specific engineering challenges related to on-orbit propellant depots, and presents the current state, technological challenges, and ultimate benefits of on-orbit cryogenic refueling.This thesis begins with a literature review of past and present research endeavors being undertaken to realize on-orbit refueling depots, focusing on the technologies necessary for and the orbital mechanics associated with cryogenic fuel depot operations. Then, an orbital dynamics study is conducted, and a method for computing refueling orbits to optimize total mission architecture mass savings over a no-refueling, single rocket case is presented. A MATLAB script has been written that allows for calculation and assessment of optimal refueling orbits around the Earth and Moon for deep space missions, utilizing specific impulses of engines and mass ratios of stages as inputs. Python is also used in conjunction with this MATLAB script to compute launch windows and to create dedicated plots to find optimal mission windows for minimizing mission energy requirements (“porkchop” plots). Next, the results of a parametric study analyzing the effects of staging, mass ratio, and specific impulse on optimal refueling orbit placement and mass savings are shown and discussed. Specifically, this parametric study confirms that orbital refueling can offer significant launch vehicle mass savings, potentially providing equivalent missions for 1.4-7.3 times less total mass than the traditional single rocket architecture for two-stage rockets and enabling utilization of single-stage to orbit (SSTO) launch vehicles for more demanding missions. Additionally, upcoming missions, such as NASA’s Artemis 1 mission and a SpaceX Starship Mars mission are assessed with refueling in mind, and potential mass savings are tabulated for applicable optimal refueling architectures. Finally, the idea of sustainable, on-orbit cryogenic refueling infrastructures is discussed as a whole, with long-term effects on the human exploration of the solar system theorized and presented.The second topic of research in this thesis concerns itself with developing technologies and methods needed to achieve on-orbit refueling. Specifically, the storage and transfer of cryogenic fuels between spacecraft tanks has been identified as a key issue, and so is addressed in this thesis. In the environment of microgravity, propellants are unsettled in their tanks and cannot be transferred with the receiving tank’s vent valve open. For normal (storable) fuels few problems would arise. However, for cryogenic fuels, rapid conversion to a vaporized state can occur, causing the pressure to build within a tank. This process can ultimately force the vent valve on the receiving tank to open, relieving pressure but allowing for liquid fuel to escape. One method of solution here is to pre-chill the receiver tank to some “target temperature” that is sufficiently cold to then allow a non-vented fill (NVF) to take place. Predicting this “target temperature” is a key goal of this work, and a derivation based on the 1st Law of Thermodynamics is undertaken. This derivation is similar to what was done in Kim et al. (2016) but is extended to include heat leak from the space environment as well account for initial fill levels in the receiver tank (a no-vent top off fill). Accounting for these specific factors subsequently allows for application of this prediction parameter to the entire no-vent fill and no-vent top off experimental database. The “target temperature” is computed for 158 historical tests over a wide range of fluids, injection methods, and tank geometries. Additionally, a parametric study is conducted to determine the influential factors that affect NVF, and an efficiency parameter is derived to determine the efficiency of a given no-vent fill process. Results indicate that the predicted “target temperature” (also known as the prediction parameter) can always predict the failure of a non-vented transfer if the thermal energy needing to be taken from the tank metal is too large for the transferred fluid to absorb. It is concluded here that the resulting pressure in the receiver tank depends on the specific filling method used and as such is path dependent. Therefore, transient modeling of the no-vent fill process is needed to sufficiently predict how a tank/injector/cryogen pair will behave.The third technical chapter of this thesis presents work done on a trajectory subroutine for such a model, allowing a user to select tank geometry, fuel injection method, and specify dimensions of each. The program subsequently outputs parameters necessary for a transient 1st order no-vent fill model, including the height of the liquid-vapor interface in the tank, the average angle that injected fuel impacts the tank wall, and the average distance the fuel travels through vaporized fuel. Additionally, the subroutine is then applied to the available NVF historical tests for validation. In each research area of this thesis, limitations of the research and recommended future work to improve and supplement the findings in this thesis are addressed.

The budding yeast Saccharomyces cerevisiae is an important model organism for cellular aging. A common metric for determining the lifespan of budding yeast cells is the replicative lifespan (RLS), how many times a mother cell divides in its lifetime. Traditionally, determining the RLS of yeast cells is a tedious manual process. To address this challenge, our long-term goal is to develop an automated RLS estimation process. Recently microfluidics-based methods have been developed, which generate time- series of images of individual cells. This work is focused on classifying these images into categories which can be used to estimate the RLS. We test three different deep learning models and found that all of the models have diverse and complementary errors, so we developed an ensemble of models that combine the best single models which led to high overall accuracy, precision and recall.

Corynespora cassiicola, causal agent of Corynespora leaf spot, causes severe epidemics in African violet production facilities. Recent concerns over loss of fungicide efficacy has led to this investigation of baseline sensitivity distributions to thiophanate-methyl, iprodione and fludioxonil fungicides and temperature sensitivity study to evaluate fitness of high and low sensitivity groups. During a disease outbreak, 325 single lesion isolates were collected and 40 isolates were selected randomly for an in vitro assay to determine EC[Subscript50] (the point at which 50% of mycelial growth was inhibited) values on potato dextrose agar (PDA) or PDA amended with thiophanate-methyl, iprodione and fludioxonil. EC[Subscript50] values for iprodione and thiophanate-methyl ranged from 0.0833 to 0.6478 [Mu]g/ml and 0.0157 to 0.1539 [Mu]g/ml with mean values of 0.2828 [Mu]g/ml (Figure 1-1, see Chapter 2 appendix) and 0.0553 [Mu]g/ml (Figure 1-2), respectively.Fludioxonil EC[Subscript50] values ranged from 0.0013 to 0.0103 [Mu]g/ml and the mean value was 0.0075 [Mu]g/ml (Figure 1-3). A resistance factor for each fungicide was calculated by dividing the least sensitive isolate's EC[Subscript50] value to the mean EC[Subscript50] for that fungicide. A lower resistance factor for fludioxoni-amended plates (1.37) than for iprodione (2.39) or thiophanate-methyl (2.78) (Table 1-1) indicates a tendency toward less insensitivity to this fungicide in this population. Correlation coefficients were calculated to determine cross-sensitivity between fungicides. Of the three fungicides, iprodione and fludioxonil had a moderately significant correlation (r = 0.38686) (P = 0.0125) (Table 1-2), indicating moderate cross-resistance.To determine fitness ranges of isolates to each fungicide, three isolates least sensitive to each fungicide along with three isolates most sensitive to each fungicide were used with mean radial growth area recorded every three days for twelve days across 10, 15, 20, 25, 30 and 35° C temperatures. Across all fungicide sensitivity groups, growth did not occur at 10° C and was very limited at 35° C. Optimum growth for all isolates within sensitivity groups was 25° C (Figure 2-1 to Figure 2-3, see Chapter 3 appendix) across all fungicide and observation times. Differences in growth area (mm²) between the least and most sensitive isolate groups did not differ for the thiophanate-methyl isolate group (P = 0.2246), the iprodione isolate group (P = 0.0512), or the fludioxonil isolate group (P = 0.6070) based on linear mixed model analysis.Even though significant differences did not exist in this analysis, temperature sensitivity information is an important part of fungicide resistance management. Developing baseline and temperature sensitivity data is the first step in determining fungal population sensitivity shifts for better resistance management strategies.

The purpose of this work is to provide experimental results of a real-world multi-vehicle coordination application, perimeter detection and tracking, not to prove optimality, convergence, stability, etc. The tools were provided to experimentally verify the hybrid system (a dynamical system composed of discrete and continuous states). The algorithm has been extensively tested in simulation and experiments. A decentralized, cooperative hybrid system was designed and implemented that allows a group of nonholonomic robots to successfully search for, detect, and track a dynamic perimeter with limited communication, while avoiding collisions and reconfiguring on-the-fly. Furthermore, an assessment of advantages and disadvantages was drawn concerning the simulators (Matlab and Gazebo) used from this testing. Finally, experimental results were promising, but further testing is needed. Future areas of research might include more realistic outdoor tests, methods to estimate the dynamic perimeter as it evolves, and a formal analysis of the hybrid system, i.e., cycling, stability, etc.