• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 14
  • 9
  • 2
  • 2
  • 2
  • 2
  • 2
  • 2
  • 1
  • 1
  • Tagged with
  • 35
  • 35
  • 12
  • 10
  • 9
  • 8
  • 8
  • 8
  • 8
  • 8
  • 8
  • 7
  • 7
  • 7
  • 7
  • 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

Evidence-based detection of spiculated lesions on mammography

Sampat, Mehul Pravin 28 August 2008 (has links)
Not available / text
12

Automated analysis of mammography phantom images

Brooks, Kenneth W. 08 1900 (has links)
No description available.
13

Automated acceptance criteria for the American College of Radiology (ACR) mammographic accreditation phantom images

Peng, Jinghong P. 12 1900 (has links)
No description available.
14

The impact of mammography utilization on breast cancer incidence in Hawaii

Maskarinec, Gertraud January 1996 (has links)
Thesis (Ph. D.)--University of Hawaii at Manoa, 1996. / Includes bibliographical references (leaves 136-146). / Microfiche. / xiv, 146 leaves, bound ill., maps 29 cm
15

Hierarchical segmentation of mammograms based on pixel intensity /

Masek, Martin. January 2004 (has links)
Thesis (Ph.D.)--University of Western Australia, 2004.
16

Synthesis of 2-D Images From the Wigner Distribution with Applications to Mammography and Edge Extraction

Pettit, Elaine J. (Elaine Joyce) 12 1900 (has links)
A new method for the general application of quadratic spatial/spatial frequency domain filtering to imagery is presented in this dissertation. The major contribution of this research is the development of an original algorithm for approximating the inverse psuedo Wigner distribution through synthesis of an image in the spatial domain which approximates the result of filtering an original image in the DPWD domain.
17

Three stage level set segmentation of mass core, periphery, and spiculations for automated image analysis of digital mammograms

Ball, John E., January 2007 (has links)
Thesis (Ph.D.)--Mississippi State University. Department of Electrical and Computer Engineering. / Title from title screen. Includes bibliographical references.
18

Otimização de imagens de mamografia computadorizada

Alvarez, Matheus [UNESP] 30 June 2012 (has links) (PDF)
Made available in DSpace on 2014-06-11T19:29:11Z (GMT). No. of bitstreams: 0 Previous issue date: 2012-06-30Bitstream added on 2014-06-13T18:58:43Z : No. of bitstreams: 1 alvarez_m_me_botib.pdf: 425512 bytes, checksum: ffc36e9f37fd09af33193b987cfefdc4 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O propósito deste trabalho é apresentar uma metodologia de análise computacional de radiografias do fantoma de acreditação mamográfico (FAM) do colégio americano de radiologia (CAR) a ser utilizado em serviços de mamografia. Processamentos via transformada discreta de wavelet (TDW) são usados de maneira a realçar o contraste dos objetos teste a serem analisados, possibilitando condições automáticas da avaliação do desempenho de sistemas mamográficos. Os algoritmos desenvolvidos localizam a região do fantoma na radiografia de maneira totalmente automatizada, onde são aplicados processamentos via TDW, que realçam características geométricas e morfológicas do fantoma. Cada região da imagem é submetida a testes de reconhecimento de detalhes estruturais, visando a minimização de detecção de falsos positivos. Em seguida, parâmetros morfológicos e geométricos das imagens processadas são utilizados na quantificação do nível de qualidade final da imagem. Também foram investigados as possibilidades de otimizar a dose utilizada na obtenção de mamografias e o potencial de otimização dos algoritmos desenvolvidos. A análise das tabelas de contingência revelou uma associação significativa ( p < 0.001 ) para as avaliações de massas, fibras e microcalcificações, revelando um grau de concordância satisfatório entre algoritmos e radiologistas. As médias das diferenças para simuladores de massas, fibras e microcalcificações obtidas foram, respectivamente, -0,11±0,86, -0,01±0,82 e 0,14±0,88. Foi avaliado o potencial de diminuição de dose na obtenção da mamografia de uma mama padrão e um potencial de redução de dose de aproximadamente 50% da dose utilizada atualmente foi encontrado. A metodologia desenvolvida e validada pode ser utilizada na detecção de pequenas variações da... / The purpose of this paper is to present a methodology for computed analysis of the mammographic accreditation phantom (MAP) recommended by the American College of Radiology (ACR) radiographs. Wavelet enhancement by the discrete wavelet transform (DWT) are used in order to highlight the test objects to be analyzed. Enabling automatic conditions of evaluating the performance of mammographic systems. The algorithms presented here locate the region of the ACR MAP in the radiograph by a fully automatic process. DWT is used to enhance the contrast between the objects and the background this allows morphological operator to detect the test objects in each region. Then, Morphological and geometrical pparameters of the processed images are used to quantify the final level of image quality presented by the system. We also investigated the possibilities of optimizing the dose used in the acquisition of mammograms. Our analysis revealed a strong correlation (p<0.001) in the computed evaluation of the test objects and radiologists scores.The mean difference between the human and the computed score were -0.11±0.86, -0.01±0.82, 0.14±0.88. We have found a potential of dose reduction in approximated 50% of the currently radiation level without prejudicing the image quality level. The methodology were developed and validated to show that they can be used to detect small variations in the radiographs levels. Our... (Complete abstract click electronic access below)
19

Otimização de imagens de mamografia computadorizada /

Alvarez, Matheus. January 2012 (has links)
Orientador: José Ricardo de Arruda Miranda / Coorientador: Diana Rodrigues de Pina / Banca: Sérgio José Barbosa Duarte / Banca: André Petean Trindade / Resumo: O propósito deste trabalho é apresentar uma metodologia de análise computacional de radiografias do fantoma de acreditação mamográfico (FAM) do colégio americano de radiologia (CAR) a ser utilizado em serviços de mamografia. Processamentos via transformada discreta de wavelet (TDW) são usados de maneira a realçar o contraste dos objetos teste a serem analisados, possibilitando condições automáticas da avaliação do desempenho de sistemas mamográficos. Os algoritmos desenvolvidos localizam a região do fantoma na radiografia de maneira totalmente automatizada, onde são aplicados processamentos via TDW, que realçam características geométricas e morfológicas do fantoma. Cada região da imagem é submetida a testes de reconhecimento de detalhes estruturais, visando a minimização de detecção de falsos positivos. Em seguida, parâmetros morfológicos e geométricos das imagens processadas são utilizados na quantificação do nível de qualidade final da imagem. Também foram investigados as possibilidades de otimizar a dose utilizada na obtenção de mamografias e o potencial de otimização dos algoritmos desenvolvidos. A análise das tabelas de contingência revelou uma associação significativa ( p < 0.001 ) para as avaliações de massas, fibras e microcalcificações, revelando um grau de concordância satisfatório entre algoritmos e radiologistas. As médias das diferenças para simuladores de massas, fibras e microcalcificações obtidas foram, respectivamente, -0,11±0,86, -0,01±0,82 e 0,14±0,88. Foi avaliado o potencial de diminuição de dose na obtenção da mamografia de uma mama padrão e um potencial de redução de dose de aproximadamente 50% da dose utilizada atualmente foi encontrado. A metodologia desenvolvida e validada pode ser utilizada na detecção de pequenas variações da... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: The purpose of this paper is to present a methodology for computed analysis of the mammographic accreditation phantom (MAP) recommended by the American College of Radiology (ACR) radiographs. Wavelet enhancement by the discrete wavelet transform (DWT) are used in order to highlight the test objects to be analyzed. Enabling automatic conditions of evaluating the performance of mammographic systems. The algorithms presented here locate the region of the ACR MAP in the radiograph by a fully automatic process. DWT is used to enhance the contrast between the objects and the background this allows morphological operator to detect the test objects in each region. Then, Morphological and geometrical pparameters of the processed images are used to quantify the final level of image quality presented by the system. We also investigated the possibilities of optimizing the dose used in the acquisition of mammograms. Our analysis revealed a strong correlation (p<0.001) in the computed evaluation of the test objects and radiologists scores.The mean difference between the human and the computed score were -0.11±0.86, -0.01±0.82, 0.14±0.88. We have found a potential of dose reduction in approximated 50% of the currently radiation level without prejudicing the image quality level. The methodology were developed and validated to show that they can be used to detect small variations in the radiographs levels. Our... (Complete abstract click electronic access below) / Mestre
20

Application of ultrasound characteristics in the accurate prediction of benign versus malignant solid breast nodules

Janse van Rensburg, Mariska 16 October 2012 (has links)
M.Tech. / To determine whether a combination of real-time B-Mode ultrasound, Doppler Color flow and Power Doppler flow mapping would be reliable in differentiating benign from malignant breast nodules in an attempt to avoid unnecessary biopsies, where after ultrasound guidelines would be formulated. A quantitative cross-sectional comparative descriptive design in a study population which consisted of 62 women over the age of 35 years who came to Klerksdorp Radiology services for mammography. Both breast ultrasound imaging and mammography was used as a routine procedure as part of the workup for the classification of breast nodules, before histologic specimens were obtained. All nodules were classified according to the ultrasonographic BI-RADS lexicon and compared with the pathologic results. Of the 63 patients, 63 breast nodules were detected and confirmed by biopsy. Thirty seven (59%) nodules were found to be malignant and 26 (41%) were benign according to biopsy results. Mammography had 87% sensitivity and ultrasound 60% sensitivity in detecting malignancy. It is recommended that B-mode, Color Doppler flow and Power Doppler flow mapping be used in combination with mammography for screening as a gold standard.

Page generated in 0.0554 seconds