The use of the nomination stage as the first step in the identification process is pervasive across the field of gifted education. In many cases, nominations are used to limit the number of students who will need to be evaluated using costly and time-consuming assessments for the purpose of gifted program identification and placement. This study evaluated the effect of the nomination stage on the overall efficacy of a gifted identification system. Results showed that in nearly all conditions, identification systems that require a nomination before testing result in a large proportion of gifted students being missed. Under commonly implemented conditions, the nomination stage can cause the false negative rate to easily exceed 60%. Changes to identification practices are urgently needed in order to ensure that larger numbers of gifted students receive appropriate educational placement and to maintain the integrity of gifted education services.

No description available.

Over the past few decades, visualization and application researchers have been investigating vortices and have developed several algorithms for detecting vortex-like structures in the flow. These techniques can adequately identify vortices in most computational datasets, each with its own degree of accuracy. However, despite these efforts, there still does not exist an entirely reliable vortex detection method that does not require significant user intervention. The objective of this research is to solve this problem by introducing a novel vortex analysis technique that provides more accurate results by optimizing the threshold for several computationally-efficient, local vortex detectors, before merging them using the Bayesian method into a more robust detector that assimilates global domain knowledge based on labeling performed by an expert. Results show that when choosing the threshold well, combining the methods does not improve accuracy; whereas, if the threshold is chosen poorly, combining the methods produces significant improvement.

No description available.

The aim of this thesis is to investigate the use of hyperspectral reflectance signals for the discrimination of cogongrass (Imperata cylindrica) from other subtly different vegetation species. Receiver operating characteristics (ROC) curves are used to determine which spectral bands should be considered as candidate features. Multivariate statistical analysis is then applied to the candidate features to determine the optimum subset of spectral bands. Linear discriminant analysis (LDA) is used to compute the optimum linear combination of the selected subset to be used as a feature for classification. Similarly, for comparison purposes, ROC analysis, multivariate statistical analysis, and LDA are utilized to determine the most advantageous discrete wavelet coefficients for classification. The overall system was applied to hyperspectral signatures collected with a handheld spectroradiometer (ASD) and to simulated satellite signatures (Hyperion). A leave-one-out testing of a nearest mean classifier for the ASD data shows that cogongrass can be detected amongst various other grasses with an accuracy as high as 87.86% using just the pure spectral bands and with an accuracy of 92.77% using the Haar wavelet decomposition coefficients. Similarly, the Hyperion signatures resulted in classification accuracies of 92.20% using just the pure spectral bands and with an accuracy of 96.82% using the Haar wavelet decomposition coefficients. These results show that hyperspectral reflectance signals can be used to reliably detect cogongrass from subtly different vegetation.

A constante evolução da tecnologia em saúde permite diagnósticos clínicos cada vez mais rápidos e precisos, porém eles não são isentos de danos. Conhecer e calcular a capacidade preditiva de classificadores de risco é um passo fundamental no desenvolvimento de políticas de saúde e na criação de protocolos clínicos. O processo de avaliação de um teste diagnóstico normalmente tem início com o preenchimento da tabela de contingência. Testes diagnósticos com elevado número de falso positivos geram ansiedade, exames adicionais e tratamentos desnecessários. Testes diagnósticos com elevado número de falso negativos podem contribuir para demoras desnecessárias no tratamento de uma possível doença, podendo dificultar sua cura. A partir da experiência adquirida com a análise da capacidade preditiva do partograma da OMS, esta tese apresenta o desenvolvimento de um novo indicador de capacidade preditiva e diagnóstica de testes. Justifica-se este novo desenvolvimento pela necessidade de simplificar, integralizar e exibir de forma conjunta os principais parâmetros necessários à avaliação da capacidade preditiva e diagnóstica de testes. A avalição dos partogramas de 9.995 mulheres da Nigéria e Uganda mostrou que a linha de alerta deste instrumento possui uma baixa capacidade diagnóstica como preditor para desfechos adversos graves no trabalho de parto. A partir desses e outros resultados, em fevereiro de 2018 a OMS recomendou mudanças no partograma. Com isso, foi explorado o uso do gráfico-radar como uma ferramenta para facilitar a comunicação dos parâmetros já existentes utilizados na avaliação da capacidade diagnóstica dos testes. Sete documentos com diretrizes em saúde materna publicados pela OMS entre os anos de 2016 a 2018 foram utilizados nesse processo. Utilizando o cálculo da área inscrita do gráfico-radar apresenta-se também um novo indicador integrativo para avaliação da capacidade preditiva de testes diagnósticos, o Índice de Máxima Acurácia (IMA) / The constant evolution of health technology allows for faster and more accurate clinical diagnoses, but they are not harmless. Knowing and calculating the predictive capacity of risk classifiers is a fundamental step in the development of health policies and in the creation of clinical protocols. The process of evaluating a diagnostic test usually begins with the completion of the contingency table. Diagnostic tests with high false positives generate avoidable distress, additional tests and unnecessary treatments. Diagnostic tests with high false negatives may contribute to unnecessary delays in the treatment of a possible disease. Based on the experience gained with the analysis of predictive capacity of the WHO partograph, this thesis presents the development of a new indicator of predictive and diagnostic capacity of tests. This new development is justified by the need to simplify, integrate and display together the main parameters necessary for the evaluation of the predictive and diagnostic capacity of tests. The assessment of the partographs of 9,995 women from Nigeria and Uganda showed that the alert line of this instrument has a low diagnostic capacity as a predictor for serious adverse outcomes in labor. From these and other results, in February 2018 the WHO recommended changes in the partograph. The use of the radar chart was explored as a tool to facilitate the communication of the existing parameters used in the evaluation of the diagnostic capacity of the tests. Seven documents with guidelines on maternal health published by WHO between 2016 and 2018 were used in this process. Using the calculated area of the radar chart, a new integrative indicator for the predictive capacity of diagnostic tests, the Index of Maximum Accuracy (IMA), is also presented

Este trabalho apresenta como parte de resultados, uma ampla pesquisa que permitiu identificar os grupos de pesquisas mais importantes do mundo, os quais possuem em comum o processamento de imagens médicas, mais especificamente o processamento de imagens que busca a identificação de microcalcificações mamárias. O vasto levantamento, a seleção e organização culminou na reunião de mais de cem artigos, publicados nos mais importantes periódicos da área, que mostram claramente as formas utilizadas pelos grupos de pesquisa para apresentação dos resultados encontrados pelos seus algoritmos. Esses resultados devem auxiliar o médico no diagnóstico do câncer de mama. Demonstramos neste trabalho porque as técnicas utilizadas para apresentação dos resultados são insatisfatórias e propusemos um novo método de avaliação desses resultados. O método proposto no trabalho baseia-se no teste do X&sup2 (Qui-Quadrado), nas curvas ROC (Receiver Operating Characteristic) e no teste de concordância, que juntos permitem apresentar de forma clara e objetiva as relações entre verdadeiros positivos e falsos positivos, verdadeiros negativos e falsos negativos, sensibilidade e especificidade do algoritmo analisado. O novo método é preciso e tem bases estatísticas conhecidas pelos médicos e pelos pesquisadores, facilitando sua aceitação. / This work presents as part of results, a wide investigation that it allowed to identify the principal research groups of the world, which possess in common the processing of medical images, more specifically the processing of images that search for the identification of mammary microcalcifications. The vast collection, selection and organization culminated in the meeting of more than a hundred articles, published in the most important newspapers of the area, that show the forms used by the research groups to present the results found clearly by its algorithms. Those results should assist the doctor in the diagnosis of the breast cancer. We demonstrated in this work that the techniques used for presentation of the results are unsatisfactory and we proposed a new method of evaluation of those results. The proposed method bases on the test of the X&sup2 (Qui-square), in ROC curve (Receiver Operating Characteristic) and in the agreement test, that take together allow to present in a clear and objective way the relationships among true positive and false positive, true negative and false negative, sensibility and specificity of the analyzed algorithm. The new method is precise and has statistical bases known by the clinicians and researchers, facilitating its acceptance.

Made available in DSpace on 2016-08-10T10:55:28Z (GMT). No. of bitstreams: 1 Paula Lorenzoni Paste de Oliveira.pdf: 1048800 bytes, checksum: 578b18c5642e39d43d7fcc9b72357fc4 (MD5) Previous issue date: 2006-12-15 / Nowadays the high level of cholesterol is one of the illnesses that reach millions of people causing about 12 million deaths in many developed countries, being the main cause of deaths in developing countries. Various researches in the world had defined the reference values for the total cholesterol levels in the blood. The reference values for the cholesterol exam are: desirable smaller than 200; smaller than 200 and larger or equal to 239 moderate risk and high risk larger ou equal to 240. Currently the dosage is carried through by spectrophotometer that it is based on the analysis of the color concentration, such concentration is directly proportional to the dosage of cholesterol (mg/dL) found in the sample. With computers, several techniques are being improved to aid clinical diagnosis. The aim of our work was to develop a technique using Digital Image Processing to dose the cholesterol level in a sample. The ELISA microtiter plate was used to capture a digital image with cholesterol exams samples. From this image, techniques of image processing were applied to find out the cholesterol concentration. The results were compared by statistical analysis with the results found for the current technique (spectrophotometer). The correlation coefficient analysis was applied and the results found was -0,94 (IC 95% = -0,9668 the -0,9036, P smaller than 0,0001), demonstrating that the process of capture and analysis of the cholesterol images has a strong correlation with the current method. To validate the experiment a ROC curve of the test was constructed. The area under the ROC curve was of 0,868 (IC 95% = 0,798 the 0,920, P smaller than 0,0001) evidencing to be a method with good discriminatory power. These results demonstrate that the analysis and image capture process obtained the desired results. The method proposed is an alternative way to assist in the dosage of the total cholesterol and can be used to develop equipments at low cost. / Nos dias de hoje a hipercolesterolemia é um dos distúrbios metabólicos que atinge milhões de pessoas, causando cerca de 12 milhões de óbitos em diversos países desenvolvidos, sendo a principal causa de óbitos em países em desenvolvimento. Diversas pesquisas em várias partes do mundo definiram os valores de referência para os níveis de colesterol total no sangue. Os valores de referência para o exame de colesterol são: desejável menor que 200; risco moderado de 200 à 239 e alto risco maior ou igual a 240. Atualmente a dosagem é realizada pelo espectrofotômetro que se baseia na análise da concentração de cor, sendo diretamente proporcional à dosagem de colesterol (mg/dL) encontrada na amostra. Com a evolução da informática, várias técnicas estão sendo aprimoradas por meio do uso de computadores no auxílio ao diagnóstico clínico. Neste trabalho propôs-se desenvolver uma técnica utilizando processamento de imagens para dosar o nível de colesterol existente nas amostras. Utilizou-se uma placa de ELISA para realizar a captura de uma imagem digital com amostras de colesterol. A partir desta captura, aplicaram-se técnicas de processamento de imagens para encontrar a concentração de colesterol existente. Estes resultados então, foram comparados estatisticamente com os resultados encontrados pela técnica atual (espectrofotômetro). A análise de correlação foi efetuada e os resultados encontrados com os experimentos foram de -0,94 (IC 95% = -0,9668 a -0,9036, p menor que 0,0001). Esse trabalho sugere que o processo de captura e análise das imagens de colesterol tem uma correlação forte com o método atual. Para validar o experimento, construiu-se a curva ROC do teste. A área abaixo da curva foi de 0,868 (IC 95% = 0,798 a 0,920, p menor que 0,0001) evidenciando um bom poder discriminatório. Estes resultados demonstram que o processo de captura e de análise obtiveram os resultados esperados. O modelo proposto é um método alternativo para auxiliar na dosagem do colesterol total, podendo ser utilizado para desenvolver um equipamento de baixo custo.

Modelos estatísticos com objetivos preditivos são frequentemente aplicados como ferramentas no processo de tomadas de decisão em diversas áreas. Uma classe importante de modelos estatísticos é composta por modelos de análise de sobrevivência. Duas quantidades são de interesse nessa classe: o tempo até o instante do evento de interesse ou o status para um determinado instante de tempo fixado. Aplicações importantes desses modelos incluem a identificação de novos marcadores para certas doenças e definição de qual terapia será mais adequada de acordo com o paciente. Os marcadores utilizados podem ser dados por biomarcadores, assim como por marcadores baseados em modelos de regressão. Um exemplo de marcador baseado em modelos de regressão é dado pelo preditor linear. Ainda que a utilização de modelos de sobrevivência com objetivos preditivos seja de suma importância, a literatura nesse assunto é muito esparsa e não há consenso na forma de se avaliar o desempenho preditivo desses. Esse trabalho pretende reunir e comparar diferentes abordagens de se avaliar o desempenho preditivo de modelos de sobrevivência. Essa avaliação é feita principalmente utilizando-se funções de perda para o tempo de sobrevivência e quantidades associadas a diferentes definições de curva ROC para o status. Para a comparação dessas diferentes metodologias foi feito um estudo de simulação e no final aplicou-se essas técnicas em um conjunto de dados de um estudo do Instituto do Câncer de São Paulo. / In many fields, predictive models are often applied as a helpful tool in the decision making process. An important class of predictive models is composed by survival models. Two quantities of special interest in these class are: time until the occurrence of a specified event and survival status for a fixed moment of time. Important applications of these models include new markers identification for certain diseases, as well as defining which therapy is the most appropriated for a patient. Markers can be given by biomarkers, but they can also be derived from regression models. An example of regression models based markers is the linear predictor. Despite the importance of survival models applications with predictive goals, literature is this subject is very sparse and there is no agreement on the best methodology to evaluate predictive performance of these models. In this work we intend to assemble and to compare different methodologies for assessing the predictive performance of survival models. This assessment is made mainly with loss functions for the survival time and ROC curve associated quantities for status. An simulation study was done in order to compare these different methodologies, which were also applied to a study about survival of patients at ICU of ICESP (Instituto do Câncer de São Paulo)

En recherche médicale, les cohortes permettent de mieux comprendre l'évolution d'une pathologie et d'améliorer la prise en charge des patients. La mise en évidence de liens de causalité entre certains facteurs de risque et l'évolution de l'état de santé des patients est possible grâce à des études étiologiques. L'analyse de cohortes permet aussi d'identifier des marqueurs pronostiques de l'évolution d'un état de santé. Cependant, les facteurs de confusion constituent souvent une source de biais importante dans l'interprétation des résultats des études étiologiques ou pronostiques. Dans ce manuscrit, nous présentons deux travaux de recherche en Biostatistique dans la thématique des scores de propension. Dans le premier travail, nous comparons les performances de différents modèles permettant d'évaluer la causalité d'une exposition sur l'incidence d'un événement en présence de données censurées à droite. Dans le second travail, nous proposons un estimateur de courbes ROC dépendantes du temps standardisées et pondérées permettant d'estimer la capacité prédictive d'un marqueur en prenant en compte les facteurs de confusion potentiels.En cohérence avec l'objectif de fournir des outils statistiques adaptés, nous présentons également dans ce manuscrit une application nommée Plug-Stat®. En lien direct avec la base de données, elle permet de réaliser des analyses statistiques adaptées à la pathologie afin de faciliter la recherche épidémiologique et de mieux valoriser les données de cohortes observationnelles. / In medical research, cohorts help to better understandthe evolution of a pathology and improve the care ofpatients. Causal associations between risk factors andoutcomes are regularly studied through etiological studies. Cohorts analysis also allow the identification of new markers for the prediction of the patient evolution.However, confounding factors are often source of bias in the interpretation of the results of etiologic or prognostic studies.In this manuscript, we presented two research works in Biostatistics, the common topic being propensity scores.In the first work, we compared the performances of different models allowing to evaluate the causality of an exposure on an outcome in the presence of rightc ensored data. In the second work, we proposed anestimator of standardized and weighted time-dependentROC curves. This estimator provides a measure of theprognostic capacities of a marker by taking into accountthe possible confounding factors. Consistent with our objective to provide adapted statistical tools, we also present in this manuscript an application, so-calledPlug-Stat®. Directly linked with the database, it allows toperform statistical analyses adapted to the pathology in order to facilitate epidemiological studies and improve the valorization of data from observational cohorts.