• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 82
  • 61
  • 24
  • 22
  • 17
  • 11
  • 7
  • 6
  • 6
  • 4
  • 4
  • 3
  • 3
  • 2
  • 2
  • Tagged with
  • 314
  • 124
  • 58
  • 43
  • 30
  • 30
  • 28
  • 28
  • 27
  • 26
  • 26
  • 24
  • 24
  • 24
  • 24
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
11

Software gerenciador de base de dados e imagens radiológicas para avaliação de monitores / Software to manage database and radiological images developed to monitor´s evaluation

Pires, Silvio Ricardo [UNIFESP] January 2007 (has links) (PDF)
Made available in DSpace on 2015-12-06T23:47:05Z (GMT). No. of bitstreams: 0 Previous issue date: 2007 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Objetivo: A proposta deste trabalho e identificar as caracteristicas dos monitores que podem influenciar na deteccao de sinais presentes na imagem e quantificar com que precisao eles sao detectados. Metodos: Foi desenvolvido um software capaz de gerenciar o processo de avaliacao do desempenho dos profissionais na deteccao de microcalcificacoes e fibras utilizando monitores comuns de 19 e 22, monitores especificos Clinton e Barco e um negatoscopio. Foram geradas 180 imagens em mamografos com diferentes condicoes tecnico-operacionais, utilizando o simulador estatistico ALVIM® e digitalizadas no scanner Lumiscan75. Foram avaliadas por 35 profissionais em negatoscopios e em diversas estacoes de trabalho, controladas por um servidor de banco de dados. As 180 imagens foram simuladas em espessura equivalente a de uma mama padrao e grande. A comparacao entre os dispositivos de visualizacao foi estatisticamente avaliada sob tres condicoes: dados originados das imagens que simulam mama padrao (4,5cm), dados originados das imagens que simulam mama grande (6,5 cm) e dados das imagens analisadas em conjunto. Para cada um destes grupos de imagens foi efetuada uma avaliacao estatistica, considerando a deteccao das estruturas de forma individual, bem como em conjunto. Para cada condicao foi gerada uma analise da concordancia na deteccao dos sinais presentes nas imagens (valores de Kappa), probabilidade de detectabilidade (P(det)) e a curva ROC. Uma fase de treinamento utilizando as ferramentas de manipulacao das imagens disponiveis no software foi realizada durante 8 horas. Para a analise estatistica foram considerados os resultados de 3.821 imagens avaliadas. Resultados: o treinamento utilizando os sistemas digitais demonstrou que os profissionais evoluem significativamente na deteccao de sinais na imagem com uma reducao no tempo de interpretacao com correlacao de -0,55 para os residentes e -0,05 para os especialistas. Nas imagens avaliadas em sistemas digitais os profissionais apresentaram maior sensibilidade e menor especificidade em comparacao ao negatoscopio. Os profissionais demonstraram menor subjetividade na interpretacao das imagens para sistemas digitais, bem como maior deteccao para fibras em relacao as microcalcificacoes. Os monitores Clinton e Barco apresentaram um desempenho de brilho 150 por cento maior do que o apresentado pelos monitores de 19 e 22. Ambos os monitores Clinton e Barco foram qualificados como apropriados para interpretacao de exames radiol6gicos em formato digital. Conclusao: Sistemas digitais permitem por meio do uso das ferramentas de manipulacao de imagens, maior concordancia na detectabilidade dos sinais presentes nas imagens simuladas. 0 treinamento dos profissionais alem de melhorar a detectabilidade de microcalcificacoes e fibras, agiliza a analise das imagens. A deteccao de fibras em imagens simuladas foi superior a deteccao de microcalcificacoes. As caracteristicas minimas dos monitores para analise de imagens mamograficas garantem que 0 desempenho do profissional seja equivalente ao obtido com o negatoscopio, sao: resolucao espacial de 3.0 Megapixels, resolucao de contraste de 10 bits/pixel e luminancia de 500 cd/m2 / Objective: The purpose of this work is to identify the characteristics of monitors that can modify the detection of signals into images and quantify the detection accuracy. Methods: It was developed a software able to manage the professional’s performance process at the microcalcification and fiber detection using viewbox and the following monitors: 19”, 22”, Clinton and Barco. A hundred and eighty images were generated in different mammographic equipment under several technical conditions using ALVIM statistical phantom. These images were randomly presented for thirty five professionals that evaluated them at viewbox and monitors in workstation controlled by a Database Server. The images were generated simulating standard and thickness breast (4.5 and 6.5 cm) and the data from image visualization devices were analyzed individually for each type of object and together. For each case was made a statistical analysis considering the microcalcifications and fibers separately and grouped. For each condition were generated the signals detectability agreements (Kappa values), detectability probability (Pdet) and ROC Curve. The professionals accomplished a training using software tools during eight hours previously to starting the objects detection into images. For the statistical analysis were considered 3.821 images. Results: The training using digital systems showed significant performance evolution while the time consumed decreased. The professionals presented higher sensibility and lower specificity for soft copy displays comparing with viewboxes. They also presented less subjectivity in the images readings using monitors. The Clinton and Barco monitors showed brightness 150% bigger than the others two. These were classified as appropriate to digital image readings. Conclusions: The digital readings using software tools generate higher agreements at the phantom image signals detection. The training can improve the microcalcifications and fibers detection along with timing saving. The fibers detection in the phantom image is higher than the microcalcification detection considering both phantom thicknesses (4,5 and 6,5cm). The minimum soft copy display specifications for mammograms readings are: 3 Megapixels for spatial resolution, 10 bits/pixel for contrast resolution and 500 cd/m2 for luminance. / BV UNIFESP: Teses e dissertações
12

Influencia de tempos alternativos de dessensibilização das placas de fosforo do sistema denoptix na qualidade da imagem digital / Performance of alternative DenOptix PSPs erasing times on subjective digital image quality

Melo, Daniela Pita de 28 February 2007 (has links)
Orientador: Guilherme Monteiro Tosoni / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Odontologia de Piracicaba / Made available in DSpace on 2018-08-08T15:57:15Z (GMT). No. of bitstreams: 1 Melo_DanielaPitade_M.pdf: 2421264 bytes, checksum: 450ce050ad326e2d47d3b89064367748 (MD5) Previous issue date: 2007 / Resumo: O objetivo, neste trabalho, foi verificar se tempos alternativos de dessensibilização das placas de fósforo do sistema DenOptix causam interferência na qualidade da imagem radiográfica e no diagnóstico de cáries proximais. Foram obtidas imagens de 10 phantoms, constituídos de dentes humanos, utilizando-se placas de fósforo parcialmente dessenbilizadas. Os dez tempos de dessensibilização estudados foram em ordem decrescente: 130s, 98s, 66s, 34s, 25s, 20s, 15s, 10s, 5s e 0s. As imagens foram posteriormente avaliadas, quanto à sua qualidade subjetiva, por dois examinadores. Cada avaliador atribuiu escores para contraste, densidade e nitidez da imagem; para definição das estruturas - esmalte, dentina e junção amelo-dentinária; e para a presença ou ausência de imagem dupla. Os resultados foram submetidos à correlação de Spearman, análise de variância e teste de Tukey. Os tempos de 20s a 130s apresentaram qualidade satisfatória em todos os quesitos abordados. Após esta primeira avaliação, os seis tempos considerados estatisticamente eficazes nos quesitos referentes à qualidade geral da imagem foram submetidos a uma nova avaliação, quanto à presença de cárie proximal. As imagens correspondentes aos tempos selecionados foram interpretadas por cinco avaliadores. Cada avaliador atribuiu um escore para cada uma das faces dentárias interpretadas. Esses escores foram posteriormente comparados ao padrão ouro. Os tempos de dessensibilização alternativos foram avaliados, quanto à sua acurácia no diagnóstico de cárie, por meio da curva ROC. As médias das curvas ROC correspondentes aos tempos de dessensibilização avaliados variaram de 0,61 a 0,66 e não apresentaram diferença estatística significativa entre elas. O desempenho dos tempos de dessensibilização alternativos de 20s a 130s foi satisfatório tanto nos quesitos de qualidade geral da imagem como no diagnóstico de cáries proximais, demonstrando a possibilidade da utilização de tempos alternativos de dessensibilização das placas do sistema DenOptix na redução do tempo de trabalho na clínica diária / Abstract: The aim on this study was to verify if DenOptix¿s phosphor plates alternative erasing times would cause any influence on the image quality and on the diagnosis of approximal caries. Ten human teeth phantoms were radiographed using phosphor plates, partially erased. The chosen erasing times were: 130s, 98s, 66s, 34s, 25s, 20s, 15s, 10s, 5s e 0s. Two observers, independently, evaluated the general quality of each image, scoring the images on contrast, density, sharpness and structures definition (enamel, dentine and enamel-dentine junction). The presence of double image was also evaluated. The resultant data was evaluated using the Spearmen¿s correlation, variance analysis, and Tukey¿s test. The erasing times from 20s to 130s had a satisfactory quality in all the evaluated items. After the first evaluation, the erasing times that showed statistical efficiency in the evaluated quality items of the image went through another evaluation regarding the presence of approximal caries. The images of the selected erasing times were evaluated by five observers. Each observer scored the approximal surfaces of each tooth. The scores were compared to the golden standard. The alternative erasing times were evaluated by their accuracy on predicting caries by means of the ROC curve analysis. The means of the ROC curves of each erasing time evaluated varied from 0,61 to 0,66 and did not show significant statistical difference between them. The alternative erasing times from 20s to 130s was satisfactory both on the quality analysis as on approximal caries diagnosis, what makes the use of those alternative erasing times of DenOptix¿s phosphor plates feasible during clinical practice / Mestrado / Radiologia Odontologica / Mestre em Radiologia Odontológica
13

How Good Is "Good"? Making Better Use of Subjective Information in Bank Internal Credit Scoring Systems /

Lehmann, Bina. January 2008 (has links)
Konstanz, Univ., Diss., 2008.
14

Semiparametric least squares analysis of the receiver operating characteristic curve /

Zhang, Zheng, January 2004 (has links)
Thesis (Ph. D.)--University of Washington, 2004. / Vita. Includes bibliographical references (p. 91-94).
15

ASYMPTOTIC PROPERTIES OF PARTIAL AREAS UNDER THE RECEIVER OPERATING CHARACTERISTIC CURVE WITH APPLICATIONS IN MICROARRAY EXPERIMENTS

Liu, Hua 01 January 2006 (has links)
Receiver operating characteristic (ROC) curves are widely used in medical decision making. It was recognized in the last decade that only a specific region of the ROC curve is of clinical interest, which can be summarized by the partial area under the ROC curve (partial AUC). Early statistical methods for evaluating partial AUC assume that the data are from a specified underlying distribution. Nonparametric estimators of the partial AUC emerged recently, but there are theoretical issues to be addressed. In this dissertation, we propose two new nonparametric statistics, partially integrated ROC and partially integrated weighted ROC, for estimating partial AUC. We show that our partially integrated ROC statistic is a consistent estimator of the partial AUC, and derive its asymptotic distribution which is distribution free under the null hypothesis. In the partially integrated ROC statistic, when the ROC curve crosses the Uniform distribution function (CDF) and if the partial area evaluated contains the crossing point, or when there are multiple crossing, the partially integrated ROC statistic might not perform well. To address this issue, we propose the partially integrated weighted ROC statistic. This statistic evaluates the partially weighted AUC, where larger weight is given when the ROC curve is above the Uniform CDF and smaller weight is given when the ROC curve is below the Uniform CDF. We show that our partially integrated weighted ROC statistic is a consistent estimator of the partially weighted AUC. We derive its asymptotic distribution which is distribution free under the null hypothesis. We propose to apply our two nonparametric statistics to functional category analysis in microarray experiments. We define the functional category analysis to be the statistical identification of over-represented functional gene categories in a microarray experiment based on differential gene expression. We compare our statistics with existing methods for the functional category analysis both via simulation study and application to a real microarray data, and demonstrate that our two statistics are effective for identifying over-represented functional gene categories. We also emphasize the essential role of the empirical distribution function plots and the ROC curves in the functional category analysis.
16

Methods for Designing and Forming Predictive Genetic Tests

Lu, Qing 24 June 2008 (has links)
No description available.
17

The ROC Curve and the Area under the Curve (AUC)

Zheng, Shimin 17 February 2017 (has links)
No description available.
18

Predicción del parto vaginal a través de la puntuación Bishop en nulíparas inducidas con oxitocina de 41 y 42 semanas de gestación, atendidas en el Hospital Docente Madre Niño “San Bartolomé” – 2014

Chávez Tacas, Liseth Rosaura January 2015 (has links)
OBJETIVO: Evaluar la puntuación Bishop que predice el parto vaginal en nulíparas inducidas con oxitocina de 41 y 42 semanas de gestación, atendidas en el Hospital Docente Madre Niño “San Bartolomé” – 2014. METODOLOGÍA: Estudio de tipo observacional, descriptivo correlacional, retrospectivo, longitudinal. Para el estudio se trabajó con el total de gestantes atendidas en el Hospital Docente Madre Niño “San Bartolomé” de enero a diciembre del 2014, lo cual fue de 152 pacientes. Los datos se registraron en el programa Statistics SPSS v.21. Para el análisis descriptivo de las variables cuantitativas se estimó medidas de tendencia central (media) y medidas de dispersión (desviación estándar) y para el análisis de variables cualitativas (nominal) se estimó frecuencias absolutas y porcentajes (frecuencias relativas). Para el análisis inferencial se utilizaron pruebas paramétricas (T* Student) y no paramétricas (Chi cuadrado). Asimismo para obtener el punto de corte de mayor rendimiento se elaboró la curva ROC. RESULTADOS: El 55.9% de la población gestante con 41 y 42 semanas de gestación finalizó en cesárea y el 44.1% culminó por parto vaginal. La duración promedio de la fase latente en el grupo de gestantes que finalizó en parto vaginal con puntuación Bishop >= a 7 puntos fue 8.35 horas y en el grupo de gestantes con puntuación Bishop < 7 puntos fue 4.88 horas. La duración de la fase activa promedio en las gestantes con puntuación Bishop >= 7 fue 5.73 horas similar a la duración de la fase activa promedio en las gestantes con puntuación Bishop < 7 puntos (4.84 horas). La duración promedio del expulsivo en las gestantes del primer grupo con puntuación Bishop >= 7 puntos fue 19.86 minutos y en el grupo con puntuación Bishop < 7 puntos fue 15.54 minutos. Según puntuación Bishop >= 7, el 67.2% culminaron en parto vaginal y 60% de las gestantes finalizó en cesárea. Asimismo, en el grupo de gestantes con puntuación Bishop < 7, el 32.8% de las gestantes finalizaron por parto vaginal a diferencia del 40% de gestantes que culminaron por cesárea. La estimación de los valores diagnósticos para los puntos de corte de la puntuación Bishop 7 en la predicción del tipo de parto son: sensibilidad 67%, especificidad 40%, VPP 47% y VPN 61%. Se Evaluó el área bajo la curva (AUC) el cual fue 0.601, planteándose que el puntaje Bishop es un test diagnóstico, aceptable para predecir el parto vaginal, estimándose un nuevo punto de corte de 6 con una sensibilidad de 70% y la especificidad de 40%. Al evaluar con puntuación Bishop >= 6 puntos, el 100% culminaron en parto vaginal y 71.4% de las gestantes finalizó en cesárea. Asimismo, en el grupo de gestantes con puntuación Bishop < 6 puntos, el 28.6% de las gestantes finalizaron por parto cesárea, observándose que existe relación significativa entre la puntuación Bishop >= 6 puntos y la culminación por parto vaginal. CONCLUSIÓN: La puntuación Bishop con mayor rendimiento diagnóstico para la predicción del parto vaginal en nulíparas inducidas con oxitocina de 41 y 42 semanas de gestación, fue 6 con una sensibilidad de 70% y especificidad del 40%. / Tesis
19

Application of ROC curve analysis to metabolomics data sets for the detection of cancer in a mouse model

Moroz, Jennifer Unknown Date
No description available.
20

Application of ROC curve analysis to metabolomics data sets for the detection of cancer in a mouse model

Moroz, Jennifer 11 1900 (has links)
The goal of this study was to show that quantifiable metabolic changes may be used to screen for cancer. NIH III nude mice (n=22) were injected with human GBM cells, with daily urine samples collected pre and post-injection. 14 mice were injected with saline to serve as controls. The measurement of metabolite concentrations took place on an 800 MHz NMR spectrometer. 34 metabolites were identified and quantified, through targeted profiling, with Chenomx Suite 5.1. Univariate statistical analysis showed that 3 metabolites (2-oxoglutarate, glucose and trimethylamine n-oxide) were significantly altered in the presence of tumour, while PCA and PLS-DA models found the maximum variance between the healthy and tumour-bearing groups. Receiver operating characteristic (ROC) curve analysis was applied to the data set to provide a measure of clinical utility. ROC statistics were as high as 0.85 for the analysis of individual metabolites, 0.939 for the analysis of metabolite pairs and 0.996 for PLS-DA models. These results show that metabolomics has potential as a screening tool for cancer. / Medical Physics

Page generated in 0.0471 seconds