Classicação de Regiões de Mamografias em Massa e Não Massa usando Estatistica Espacial e Máquina de Vetores de Suporte / Classification of Regions, in the Mamogra Mass and not using statistical Mass Space and Support Vector Machine

AUTOR(ES)

Geraldo Braz Junior

DATA DE PUBLICAÇÃO

2008

RESUMO

Breast cancer is a serious public health problem in several countries of the world. Computer-Aided Detection/Diagnosis systems (CAD/CADx) have been used with relative success in aid to health care professionals. The main contribution of this work is to present a methodology to discriminate and classify mammogram regions in mass and non-mass through spatial statistics. The spatial statistics techniques were used as texture descriptors. This work used Moran s Index, Geary s Coecient, Getis-Ord s Index and K of Ripley function for texture description. With this extracted features, a selection step is performed to get the most discriminating ones using Stepwise Discriminant Analysis. A Support Vector Machine is then used to classify the samples. The methodology reaches promising results for classication of masses and non-masses certifying that the features generated by the spatial statistical techniques generate a set of satisfactory discriminant features. Keywords: Pattern Recognition, Spatial Statistics, Support Vector Machine, Stepwise Discriminant Analysis

ASSUNTO(S)

engenharia biomedica reconhecimento de padrões support vector machine spatial statistics pattern recognition análise discriminante stepwise máquina de vetores de suporte stepwise discriminant analysis estatística espacial

Documentos Relacionados

• CLASSIFICAÇÃO DE TECIDOS DA MAMA A PARTIR DE IMAGENS MAMOGRÁFICAS EM MASSA E NÃO MASSA USANDO ÍNDICE DE DIVERSIDADE DE MCINTOSH E MÁQUINA DE VETORES DE SUPORTE
• Otimização e análise das máquinas de vetores de suporte aplicadas à classificação de documentos.
• Classification of injuries in the Mamogram by Components of Independent Review, Analysis Discriminant Linear and Vector Machine, Support
• SEGMENTAÇÃO AUTOMÁTICA DE NÓDULOS PULMONARES COM GROWING NEURAL GAS E MÁQUINA DE VETORES DE SUPORTE
• DETECÇÃO DE REGIÕES SUSPEITAS E CLASSIFICAÇÃO DE MASSAS EM MAMOGRAFIAS DIGITAIS UTILIZANDO DESCRIÇÃO ESPACIAL COM FUNÇÃO VARIOGRAMA