The eradication of infectious diseases has been of key interest for many years. While the World Health Organization (WHO) and other health organizations typically track the progress of disease eradication based on whether regions are meeting their eradication targets, being able to quantify and/or visually track the eradication of a disease could prove beneficial. This thesis creates the “canonical path” to the elimination of measles in the United States (US), using similar methods as defined by Graham et al. [1]. We build on preliminary work conducted to fulfill the requirements of an Honours Bachelor of Science in Integrated Science at McMaster University, and the analysis conducted by Graham et al. [1], through the investigation of the sensitivity of the path to changes in its definition, as well how the path changes when we change the characteristics of the disease. This thesis demonstrates the ability to use a canonical path on a smaller, country-level scale, by using United States (US) state level data to create the US canonical path. We also determine the model structures necessary to simulate the canonical path, which suggests that the canonical path method is most useful for eradicable diseases for which we have ample knowledge of the disease, including the natural history of infection and vaccination. We also predict how the path is affected by the pattern of seasonality and by the natural history of infection. Overall, the analysis suggests that the more this method is implemented for other countries that have eliminated measles or for other diseases for which we have achieved elimination, we may gain insight of the successes and failures of elimination strategies. This knowledge could help the WHO and other organizations improve their disease elimination and eradication strategies in the future. / Thesis / Master of Science (MSc)

This research is an effort to improve the performance of successive approximation algorithm with a prime aim of solving finite states and actions, infinite horizon, stationary, discrete and discounted Markov Decision Processes (MDPs). Successive approximation is a simple and commonly used method to solve MDPs. Successive approximation often appears to be intractable for solving large scale MDPs due to its computational complexity. Action elimination, one of the techniques used to accelerate solving MDPs, reduces the problem size through identifying and eliminating sub-optimal actions. In some cases successive approximation is terminated when all actions but one per state are eliminated. The bounds on value functions are the key element in action elimination. New terms (action gain, action relative gain and action cumulative relative gain) were introduced to construct tighter bounds on the value functions and to propose an improved action elimination algorithm. When span semi-norm is used, we show numerically that the actual convergence of successive approximation is faster than the known theoretical rate. The absence of easy-to-compute bounds on the actual convergence rate motivated the current research to try a heuristic action elimination algorithm. The heuristic utilizes an estimated convergence rate in the span semi-norm to speed up action elimination. The algorithm demonstrated exceptional performance in terms of solution optimality and savings in computational time. Certain types of structured Markov processes are known to have monotone optimal policy. Two special action elimination algorithms are proposed in this research to accelerate successive approximation for these types of MDPs. The first algorithm uses the state space partitioning and prioritize iterate values updating in a way that maximizes temporary elimination of sub-optimal actions based on the policy monotonicity. The second algorithm is an improved version that includes permanent action elimination to improve the performance of the algorithm. The performance of the proposed algorithms are assessed and compared to that of other algorithms. The proposed algorithms demonstrated outstanding performance in terms of number of iterations and omputational time to converge.

This research is an effort to improve the performance of successive approximation algorithm with a prime aim of solving finite states and actions, infinite horizon, stationary, discrete and discounted Markov Decision Processes (MDPs). Successive approximation is a simple and commonly used method to solve MDPs. Successive approximation often appears to be intractable for solving large scale MDPs due to its computational complexity. Action elimination, one of the techniques used to accelerate solving MDPs, reduces the problem size through identifying and eliminating sub-optimal actions. In some cases successive approximation is terminated when all actions but one per state are eliminated. The bounds on value functions are the key element in action elimination. New terms (action gain, action relative gain and action cumulative relative gain) were introduced to construct tighter bounds on the value functions and to propose an improved action elimination algorithm. When span semi-norm is used, we show numerically that the actual convergence of successive approximation is faster than the known theoretical rate. The absence of easy-to-compute bounds on the actual convergence rate motivated the current research to try a heuristic action elimination algorithm. The heuristic utilizes an estimated convergence rate in the span semi-norm to speed up action elimination. The algorithm demonstrated exceptional performance in terms of solution optimality and savings in computational time. Certain types of structured Markov processes are known to have monotone optimal policy. Two special action elimination algorithms are proposed in this research to accelerate successive approximation for these types of MDPs. The first algorithm uses the state space partitioning and prioritize iterate values updating in a way that maximizes temporary elimination of sub-optimal actions based on the policy monotonicity. The second algorithm is an improved version that includes permanent action elimination to improve the performance of the algorithm. The performance of the proposed algorithms are assessed and compared to that of other algorithms. The proposed algorithms demonstrated outstanding performance in terms of number of iterations and omputational time to converge.

Typescript (photocopy). / Digitized by Kansas Correctional Industries

Controlling and eliminating malaria in Indonesia is a challenging endeavour. Evidence-based strategic plans should be critically formulated to overcome a complex mosaic of infection risk across the 5000-km-long archipelago of thousand islands and distinctive habitats. This project aimed to thoroughly explore the challenges and opportunities for controlling/eliminating malaria and present the application of malaria cartographic tools to allow malaria control agencies and their partners to comprehensively assess the prospects for reaching pre-elimination, monitor and evaluate the effectiveness of future strategies against the baseline generated. First, the historical context of malaria in Indonesia and important methods of control and their impact in the context of the political systems that supported them was comprehensively described. A series of distribution maps of twenty Anopheles malaria vector mosquitoes in Indonesia were also produced, supported by comprehensive reviews of each species’ bionomics and susceptibility to insecticides. Then, the application of malaria cartographic tools for malaria control/elimination in Indonesia was explored. The first high spatial resolution (1 x 1 km) baseline endemicity maps of Plasmodium falciparum and P. vivax malaria were generated, together with corresponding estimates of population at risk and clinical burdens of each species in 2010. Low malaria prevalences of these parasites were predicted in western areas, with high prevalences in eastern Indonesia. Over 132 million people in Indonesia lived at risk of P. falciparum transmission with 70% of them in areas of unstable transmission and 30% in stable transmission. There were an estimated 7.7 million P. falciparum clinical cases across the populations at risk. Meanwhile, nearly 130 million people lived at risk of P. vivax malaria with 79% living in unstable and 21% in stable transmission areas. This infection caused 1.5 million estimated clinical cases in 2010. Both estimates were 30-fold and 6-fold higher than routinely reported numbers, respectively. Finally, this project revealed the substantial multi-faced problems that impede current efforts towards the pre-elimination agenda. High rates of undiagnosed clinical cases, insufficient competence of malaria microscopy, inadequate primaquine dosing against P. vivax malaria infections, insufficient evidence of vector control interventions, wide diversities of vector mosquitoes and their bionomics, mosquito resistance against insecticides and inadequate malaria surveillance systems are challenging the task of controlling and eliminating malaria. A diverse range of strategies enabling locally-specific approaches must be implemented for controlling and eliminating malaria in Indonesia. Strategic recommendations are listed and future research priorities are proposed for further study.

Many intervention methods theoretically have the potential to eliminate microorganisms. However, they do not perform efficiently once applied to fruits and vegetables. In this study Salmonella Typhimurium LT2 and hydrogen peroxide (H2O2) were used as model systems on 6 different types of produce to determine the effect of produce surface characteristics on sanitizer effectiveness. Microbial attachment on produce surfaces was induced after 3 h of drying at 24°C and high relative humidity (RH). Afterwards, produce was stored for 3 subsequent days and washed with sterilized tap water for 5, 10 and 15 minutes to separate weakly from strongly attached microorganisms from the produce surface. The strongly attached microorganisms were then treated with 3% H2O2 for 1, 3, and 5 minutes. These results were compared to the log reduction curves obtained with a pure liquid culture and 3% H2O2. Additionally, contact angle of water and diiodomethane on each type of surface were measured and used as indicator of wettability, and for calculating surface tension characteristics of the produce surfaces. Then these surface characteristics were related to the bacterial attachment and population reduction values obtained after applying the treatments. In general, the geometric mean equation was the most useful in predicting the surface tension values of produce surface and the polar and non-polar components of produce surface tension. Our results suggest that surface properties, such as roughness and surface tension of fruits and vegetables are important factors limiting decontamination. These surface properties allow the formation of micro-air pockets within the rough surface, thus contributing to create a protective environment for microorganisms and reducing the effectiveness of the chemical aqueous based intervention methods applied. Wettable surfaces (water contact angle < 90°) allowed more bacteria to attach after the washing and H2O2 chemical treatments. Roughness and surface polarity are intrinsic characteristics of produce surfaces which affect wettability and the spreading and penetration of the sanitizer treatment on the produce surface. Rough surface and porosity, considered an extreme case of roughness, enhance a deeper cell internalization and a protective environment for bacteria.

Thesis (Ph. D.)--University of Wisconsin--Madison, 1957. / Typescript. Vita. eContent provider-neutral record in process. Description based on print version record. Bibliography: leaves 110-113.

Zugl.: Darmstadt, Techn. Hochsch., Diss., 1995. / Zugl.: Darmstadt, Techn. Hochsch., Diss., 1995.

Zugl.: Dresden, Techn. Universiẗat, Diss., 2004.

The development of a quantum computer presents one of the greatest challenges in science and engineering to date. The promise of more ecient computing based on entangled quantum states and the superposition principle has led to a worldwide explosion of interest in the elds of quantum information and computation. Decoherence is one of the main problems that gives rise to dierent errors in the quantum system. However, the discovery of quantum error correction and the establishment of the accuracy threshold theorem provide us comprehensive tools to build a quantum computer. This thesis contributes to this eort by investigating a particular class of quantum error correcting codes, called Decoherence free subsystems. The passive approach to error correction taken by these encodings provides an ecient means of protection for symmetrically coupled system-bath interactions. Here I will present methods for determining the subsystem-preserving evolutions for noiseless subsystem encodings and more importantly implementing a Universal quantum computing over three-quantum dots.