Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-05-16T12:04:55Z No. of bitstreams: 1 alcindogandhibarretoalmeida.pdf: 12470862 bytes, checksum: 6af9613cce2cfd9264288cf2d6202d15 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-06-27T21:27:55Z (GMT) No. of bitstreams: 1 alcindogandhibarretoalmeida.pdf: 12470862 bytes, checksum: 6af9613cce2cfd9264288cf2d6202d15 (MD5) / Made available in DSpace on 2016-06-27T21:27:55Z (GMT). No. of bitstreams: 1 alcindogandhibarretoalmeida.pdf: 12470862 bytes, checksum: 6af9613cce2cfd9264288cf2d6202d15 (MD5) Previous issue date: 2015-04-24 / A divergência entre o parque de iluminação pública em funcionamento e o cadastro informado pelos municípios às concessionárias de energia elétrica pode resultar numa das parcelas das chamadas perdas comerciais. O procedimento usual adotado para minimizar esse problema consiste no envio de uma equipe de técnicos a campo para a inspeção dos pontos de iluminação. Este trabalho propõe um sistema de identificação automático de lâmpadas de iluminação pública (IP), que permite às concessionárias de energia elétrica elaborar um mapa dos pontos de iluminação dos municípios, estimando o consumo e evitando perdas comerciais. Inicialmente, o trabalho aborda o cenário da iluminação pública no Brasil, incluindo as principais tecnologias usadas em IP e a transferência dos ativos de IP para os municípios. Em seguida, são descritos os principais conceitos radiométricos que caracterizam o sistema de IP. Posteriormente, é apresentada a placa de aquisição de dados construída. Essa placa contém um conjunto de nove sensores radiométricos, que medem a radiação eletromagnética proveniente das lâmpadas. Também são apresentados os projetos do firmware de controle da placa e o software de aquisição de dados. É desenvolvida uma metodologia de medição baseada em sensores de posicionamento para garantir a reprodutibilidade das medidas. Além de descrever todos esses itens, este trabalho apresenta uma estrutura de iluminação pública ajustável, que permitiu o estudo comparativo das técnicas empregadas, e uma metodologia de calibração das placas de aquisição de dados. Por fim, tem-se o sistema de classificação de dados radiométricos, que permite inferir qual lâmpada está sendo medida. É feita uma comparação de diversos tipos de classificadores. Os resultados experimentais obtidos demonstram o bom funcionamento do sistema em diversas condições de operação. / The misinformation between the actual public lighting equipment and those reported by municipalities to the electricity companies may result in a kind of loss mainly referred as commercial loss. The usual adopted procedure to minimize this problem is sending technician teams to the field to do a low effective inspection of the lighting points. This paper proposes an automatic identification system of street lighting lamps, which allows the electricity companies to draw up a map of the lighting points, estimating the consumption and avoiding commercial losses. Initially, the work addresses the scenario of public lighting in Brazil, including the main technologies used in street lighting and the transfer of the assets to municipalities. Then, the main radiometric concepts which characterize the street lighting systems are described. After data acquisition board built is presented. This board contains a set of 9 radiometric sensors which measure the electromagnetic radiation from the lamps. The projects of the firmware and the data acquisition software are also presented. A measurement methodology based on positioning sensors to ensure the reproducibility of the measurements is developed. In addition to describing all of these items, this paper presents an adjustable street lighting structure, which allowed the comparative study of the techniques employed, and a calibration method of data acquisition boards. Finally, the radiometric data classification system is shown, which allows us to infer which lamp is being measured. A comparison of various types of classifiers is made. The experimental results demonstrate the proper functioning of the system in various operating conditions.

This thesis addresses the intersection of two important areas in epidemiology and statistics: genetic linkage analysis and missing data methods, respectively. Genetic linkage analysis is a promising method in viral epidemiology which involves learning about transmission patterns by studying clusters of similar gene sequences. For example, similar sequences found in a pair of geographically distinct communities may imply disease transmission between the two locations. However, this analysis is sensitive to missing data, which can introduce substantial bias. This thesis presents a multiple-imputation approach which corrects for much, though not all, of the bias in genetic linkage analysis. It also introduces a novel resampling-based approach that generates a weighted distribution of complete datasets and is even more effective than imputation for reducing bias. This work highlights the importance of missing data in genetic linkage studies and presents ways to provide more accurate epidemiological information by correcting for missing data. The new resampling-based approach presented in this paper is also general enough to be applied to many types of missing-data problems involving complex datasets; such broader applications are a promising avenue for future research. / Applied Mathematics

In the last decade, extraction of unconventional oil and gas (UOG) has rapidly increased in Canada. One of the major concerns related to UOG extraction is the risk of accidental releases during wastewater management. UOG extraction uses technologies of horizontal drilling and hydraulic fracturing. Flowback and produced water are generated as wastewater and may contain high concentrations of salts, metals, oil, grease, and organic compounds. Flowback water is stored on site before being transported for treatment, reuse or disposal. It is stored in containment pond, above ground walled storage systems (AWSS) or storage tanks. A comprehensive risk assessment has been carried out for the accidental release of flowback water during the storage. Two components of risk namely probability of failure and consequence assessment on the ecology have been examined using the frameworks of Backward Integrated Analysis (BIA) and Ecological Risk Assessment (ERA), respectively. In BIA, failure modes were identified for an uncontrolled release of flowback water due to AWSS failure by developing a fault tree. The probability of failure of the system was calculated and its failure modes were ranked by assigning risk priority number (RPN). To assess the consequence of the accidental release, the toxicity and exposure of the flowback water components to the aquatic ecology were examined through ERA. Toxicity of each constituent of flowback water was assessed by developing species sensitivity distribution curves. An exposure model using dilution factor and adsorption coefficient of the flowback water constituents is proposed and risk quotient was used to characterize ecological risk. To demonstrate the methodology, a case study in Montney unconventional play in Northern BC was carried out. The risk to the aquatic ecology was found to be very low, however, scenario analysis and uncertainty analysis prove that the risk cannot be completely overlooked. A review of the regulations for storage systems was carried out and they were assessed in light of the results of the study. Enforcing regulations pertaining to the quality of water stored, citing of the storage system with respect to the water body and making secondary system mandatory were realized to be the most beneficial. / Applied Science, Faculty of / Engineering, School of (Okanagan) / Graduate

The larger memory and channel bandwidth requirements for digital video transmission and storage make it mandatory to use compression techniques. Representation of the video signal in a pyramid format not only compresses the signal but also makes it suitable for specific applications such as packet video based on asynchronous transmission mode (ATM) and compatible advanced television (ATV). In this thesis, we propose to employ a pyramid data structure for video compression. A review of video coding schemes is first presented, followed by a review of the various 2-dimensional (2D) and 3-dimensional (3D) pyramid data structures from the perspectives of data compression. The performance of different configurations of temporal/spatial pyramid data structures is then measured for video compression in terms of the first order entropy. Based on this study, we introduce an efficient 3D adaptive temporal/spatial pyramid which selects either the temporal or spatial contractions using the temporal and spatial prediction differences, respectively. We propose a video codec that combines the adaptive temporal/spatial pyramid and an intra-frame coding technique. Simulation results on CCITT standard video sequences indicate that the adaptive pyramid reduced the lossless bit rate by a factor of two. For video conferencing applications, excellent subjective quality as well as objective quality (PSNR value of 36.6 db) are obtained at a bit rate less than T1 rate (i.e. 1.544 Mbits/s). Promising results have been obtained for CCIR resolution (720 x 480), high detail sequences at a bit rate of 6 Mbits/s. Furthermore, smooth transition is achieved in the case of scene changes without sacrificing picture quality. Finally, the algorithm is well suited for constant bit rate and constant quality applications. (Abstract shortened by UMI.)

Solutes are often most efficiently deployed in discrete pulses, for example in the delivery of herbicides or drugs. Manual application of each pulse can be labor-intensive, automated application of each pulse can be capital intensive, and both are often costly and impractical. Barrier-Mediated Pulsatile Release (BMPR) systems offer a materials-based alternative for automated pulsatile drug delivery, without pumps, power supplies, or complex circuitry. While earlier materials-based approaches such as delayed-release microcapsules are limited to two or three pulses due to the independent nature of each pulse’s timing control, BMPR systems link the timing of each pulse to the previous pulse. Each dose of drug is sequestered in its own stimuli-sensitive depot, releasing only upon contact with the stimulant. These depots are stacked with sacrificial barriers in between, each of which block the stimulant for a predetermined time. For instance, layers of soluble drug may be separated by degradable polymer layers. Water, as the stimulant, will erode the polymer layer over a fixed period of time, followed by quick dissolution and release of the underlying drug and the start of degradation for the next polymer layer. This example, however, is quickly limited by irregular polymer erosion, a single stimulant (water), and difficulty in scaling delay times. The research work presented in this thesis reports the development of a generalized BMPR system which overcomes those limitations. Model drugs (methylene blue and methyl orange) were immobilized in a pH-sensitive polymer [poly(methyl methacrylate-co-dimethylaminoethyl methacrylate)] which released only at low pH. Zinc oxide (ZnO) nanoparticles immobilized in a pH-insensitive matrix [poly(vinyl alcohol)] served as the barrier layer. The time required for acid to penetrate the barrier layer scaled with the ZnO concentration and with the square of the polymer thickness, allowing wide scaling of the delay time with only minor changes to the barrier layer. Harnessing the swelling pressure of the acid-sensitive hydrogel, each barrier/depot bilayer can delaminate upon solute release, directly exposing the next bilayer to the stimulant source. This system has demonstrated tuned release using a citric acid stimulant to produce up to ten pulses of model drug (methylene blue) over various preset timescales. This system has also demonstrated the alternate release of multiple solutes (methylene blue and methyl orange) at regular time intervals up to five pulses from a single BMPR device. For non-delaminating BMPR systems, spent bilayers impede stimulant diffusion to the inner layers and solute diffusion from the inner layers, increasing the delay time and the pulse width. To predict these changes, a computational model was constructed in FORTRAN. This model was extensively explored over a wide range of parameter space to understand the release behavior of various kinds of non-delaminating BMPR systems. The computer model also validates the performances of experimental delaminating BMPR system. This model can be used to guide the physical modeling of BMPR systems. The model also allows to incorporate variety of stimulants other than just acid. BMPR technology introduces efforts to further generalize the delivery strategy by incorporating glucose as a stimulant.

<div><div><div><p>Electronic health records collect an enormous amount of data about patients. However, the information about the patient’s illness is stored in progress notes that are in an un- structured format. It is difficult for humans to annotate symptoms listed in the free text. Recently, researchers have explored the advancements of deep learning can be applied to pro- cess biomedical data. The information in the text can be extracted with the help of natural language processing. The research presented in this thesis aims at automating the process of symptom extraction. The proposed methods use pre-trained word embeddings such as BioWord2Vec, BERT, and BioBERT to generate vectors of the words based on semantics and syntactic structure of sentences. BioWord2Vec embeddings are fed into a BiLSTM neural network with a CRF layer to capture the dependencies between the co-related terms in the sentence. The pre-trained BERT and BioBERT embeddings are fed into the BERT model with a CRF layer to analyze the output tags of neighboring tokens. The research shows that with the help of the CRF layer in neural network models, longer phrases of symptoms can be extracted from the text. The proposed models are compared with the UMLS Metamap tool that uses various sources to categorize the terms in the text to different semantic types and Stanford CoreNLP, a dependency parser, that analyses syntactic relations in the sentence to extract information. The performance of the models is analyzed by using strict, relaxed, and n-gram evaluation schemes. The results show BioBERT with a CRF layer can extract the majority of the human-labeled symptoms. Furthermore, the model is used to extract symptoms from COVID-19 tweets. The model was able to extract symptoms listed by CDC as well as new symptoms.</p></div></div></div>

Målet med detta projekt var att utforma ett tidsrapporteringssystem åt Online CC AB för att effektivisera deras kunders tidsrapportering. Systemet är en webbapplikation som ska användas till att rapportera in tid som framdeles kan exporteras till valfritt lönesystem för lönehante-ring av personal. Detta system är grunden för ett framtida, fulländat system som har utökad funktionalitet. Produkten togs fram med Ex-treme Programming samt testdriven utveckling. Under utvecklingen jobbade utvecklingsgruppen med välkända och beprövade metoder för att säkerställa ett system av hög kvalité. Webbapplikationen nyttjar moderna teknologier och ramverk för webbutveckling – inklusive Microsofts ASP.NET MVC 4 och Entity Framework. Det visade sig att apputveckling är ett diffust område där de senaste verktygen inom verksamhetsgrenen inte förhållandevis förenklade arbetet. Ett system som fungerar såväl på mobila enheter, i form av en hybridapplikation, som stationära enheter, som webbapplikation, krävde att utvecklings-gruppen var erfarna inom respektive områden. I slutet av projektet var inte alla ställda krav uppfyllda - men eftersom vi använder oss av testdriven utveckling så är systemet fullt operationsdugligt. De krav som implementerades, gjordes det till fullo. Till sist så kan det visa sig att de senaste teknologierna och ramverken inte alltid är de bästa att nyttja. Mer beprövade metoder och teknologier kan i vissa fall vara mer lämpliga. / The goal of this project was to design a timesheet system for Online CC AB in order to make time reporting more efficient for their customers. The system is a web application that is to be used for time reporting in-which, later on can be exported to a salary system of their choice for salary transactions of personnel. This system is the foundation for a future, all-in-one system with extended functionality. The product was produced using Extreme Programming and Test-Driven Development. During development the development team worked with well-renowned and well-tried methods to ensure a system with the utmost quality. The web application utilizes modern technologies and frame-works for web development – including Microsoft’s ASP.NET MVC 4 and Entity Framework. It’s shown that app development is a diffuse field in which the latest tools within the field do not comparatively simplify the work. A system that works on as-well as mobile units, in the form of a hybrid application, as stationary units, in the form of a web application, demands the development team to be experienced within respective fields. At the end of the project not all requirements are met – however due to us using Test-Driven Development, the system is fully operational. Those requirements that were implemented are done so fully. Furthermore, it’s shown that the latest technologies and frame-works not always are best-suited for usage. More well-tried methods and technologies can in some cases be more appropriate

En el Perú aún existen muchas áreas naturales protegidas (ANP) por explotar su potencial turístico. La infraestructura con la que cuenta el ANP, es de gran importancia para la visita de nuestros turistas. Por esta razón, dotar de infraestructura necesaria para implementar y mejorar la capacidad receptiva y oferta turística de la Zona de Uso turístico y Recreativo del Parque Nacional de Cutervo (PNC), es nuestro principal objetivo. Las obras que conforman esta infraestructura van desde una carretera hacia el PNC, camino peatonal hacia un mirador externo, recorrido y mirador interno de la gruta de los guácharos. El proyecto consta de 3 fases; en la primera se efectuaron las coordinaciones con las autoridades de SERNANP, se realizaron los estudios de: tráfico que correspondió a un T0, topográfico con una orografía accidentada tipo 3 y escarpada tipo 4, mecánica de suelos donde se determinó un suelo de arcillas de alta plasticidad con expansividad e hidrológico donde se obtuvieron caudales de diseño, medidas de temperatura(12°C) y humedad(80%) dentro de la gruta; en la segunda fase evaluamos los datos de campo para realizar los diseños geométricos de la carretera y el camino peatonal, así como la ubicación de las obras de arte; en la tercera fase diseñamos el pavimento vehicular, para el cual se realizaron 3 estructuras de pavimento para un óptimo diseño y peatonal que se eligió un pavimento articulado, así mismo las obras de arte y para el diseño del mirador externo e interno se optó por gaviones por su estructura flexible.

Attempts were made to adapt the microtiter hemagglutination inhibition assay technique for the assay of enterotoxin A. The presence of a potent hemagglutinin in crude and partially purified preparations and the instability of sensitized erythrocytes prevented its use for routine analysis of enterotoxin from culture media and foods. A capillary tube immunological assay was developed in which 1 μ g of enterotoxin/ml was detected in less than 1 hr . Interfacial reaction of antisera and enterotoxin solutions in a 1 mm internal diameter capillary tube allowed rapid detection and serological typing of enterotoxins. Staphylococcus aureus growth and enterotoxin A development in Cheddar cheese slurry was evaluated. S. aureus growth and enterotoxin production occurred at 32 C. in 45 and 60% moisture cheese slurries following inoculation with 10 3 to 10 5 bacteria/gram. Hydrogen peroxide (0. 5%) treatment of slurry at 37 C did not inhibit S. aureus and enterotoxin A development. Heating slurry at 72 C for 30 min eliminated staphylococci but reinoculation with ripening organisms was essential. Addition of sorbic acid (0. 2 to 0. 3%) to a slurry adjusted to pH 5. 0 with lactic acid, inhibited staphylococci. in milk and slurry. Cheese flavor development was retarded due to inhibition of micrococci and lipolysis. Non-protein nitrogen increases paralleled that of sorbate-free controls. Sorbate treatment was preferred over other treatments .

This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. / Thesis: Ph. D., Massachusetts Institute of Technology, Department of Physics, 2019 / Cataloged from student-submitted PDF version of thesis. / Includes bibliographical references (pages 91-98). / Spatial expansions occur across multiple scales, from the expanding range of a species to the growth of tumors and microbial biofilms. In ecology, range expansions are becoming more frequent due to environmental changes and rare long distance dispersal, often facilitated by anthropogenic activities. Simple models in theoretical ecology explain many emergent properties of range expansions, such as a constant expansion velocity, in terms of organism-level properties such as growth and dispersal rates. Moreover, the evolution and potentially even the survival of an expanding population depends on its genetic diversity, which is also predicted to reduce drastically during range expansions. However, testing these quantitative predictions in natural populations is difficult because of large environmental variability and the inability of replicating historical processes. / In this thesis, I describe the use of a microbial model system to gain a deeper understanding of spatial range expansions in a controlled and replicable setting. In particularly, I study the role of cooperative growth in spatial expansions. Given the prevalence of cooperative growth in nature, understanding the effects of cooperativity is essential to managing invading species and understanding their evolution. For non-cooperative growth, the expansion dynamics are dominated by population growth at the low-density front, which pulls the expansion forward. I find these expansions to be in close quantitative agreement with the classical theory of pulled waves by Fisher and Skellam, suitably adapted to my experimental system. However, as cooperativity increases, the expansions transition to being pushed, i.e. controlled by growth in the bulk as well as in the front. / In addition to the population dynamics, cooperation within populations is also predicted to significantly alter the evolutionary fate of expanding populations. This difference in evolutionary dynamics within pulled and pushed waves is also studied experimentally. / by Saurabh Rajendra Gandhi. / Ph. D. / Ph.D. Massachusetts Institute of Technology, Department of Physics