Detección de microcalcificaciones utilizando discriminantes lineales de Fisher

The main objective of this thesis is to develop a new method for detecting microcalcifications in digital mammograms, using machine learning and computer vision techniques; the method detects different shapes, sizes and intensities of microcalcifications, and also it´s able to recognize them not onl...

ver descrição completa

Detalhes bibliográficos
Autor: GABRIELA ALEJANDRA RODRIGUEZ RUIZ
Formato: tesis de maestría
Estado:Versión aceptada para publicación
Fecha de publicación:2009
País:México
Recursos:Instituto Nacional de Astrofísica, Óptica y Electrónica
Repositorio:Repositorio Institucional del INAOE
Idioma:español
OAI Identifier:oai:inaoe.repositorioinstitucional.mx:1009/442
Acesso em linha:http://inaoe.repositorioinstitucional.mx/jspui/handle/1009/442
Access Level:acceso abierto
Palavra-chave:info:eu-repo/classification/Segmentación de imagen/Image segmentation
info:eu-repo/classification/Imagen de clasificación/Classification image
info:eu-repo/classification/Visión/Vision
info:eu-repo/classification/cti/1
info:eu-repo/classification/cti/12
info:eu-repo/classification/cti/1203
Descrição
Resumo:The main objective of this thesis is to develop a new method for detecting microcalcifications in digital mammograms, using machine learning and computer vision techniques; the method detects different shapes, sizes and intensities of microcalcifications, and also it´s able to recognize them not only in fat breast but also in dense breast. To achieve this objective, the breast tissue is taken into account to detect microcalcifications even in dense breast, where the contrast difference between tissue and microcalcification is almost null. In the segmentation step the Fisher Linear Discriminants are used to segment the possible microcalcifications, to reduce the false positives generated in this step, some characteristics (morphological and intensity) are extracted from these regions. The method is tested; using ISSSTEP and MIAS databases, the ROC and FROC analysis were used as performance measures.