Explorando informações texturais das visões crânio-caudal e médio-lateral oblíqua para extração e classificação Fuzzy de imagens mamográficas / Explorando informações texturais das visões crânio-caudal e médio-lateral oblíqua para extração e classificação Fuzzy de imagens mamográficas

AUTOR(ES)
FONTE

IBICT - Instituto Brasileiro de Informação em Ciência e Tecnologia

DATA DE PUBLICAÇÃO

25/11/2010

RESUMO

The computerized system to support detection and diagnosis (Computer Aided Detection or Diagnosis - CAD) is a tool for medical help that has been used to identify and diagnose abnormalities in breast tissue. Although CAD systems have provided a large number of research and increased sensitivity, the majority analyzes the MLO and CC views independently. Experts believe that there is an inconsistency if a particular injury is similar in both views and the system does not have the ability to find it. This paper presents a method for extraction and selection of texture-related attributes and classification using the fusion of information from the visions Crânio-Caudal (CC) and Medio-Lateral Oblique (MLO). In the extraction stage, the wavelet transform method (TW) and the technique of Singular Value Decomposition (SVD) provided an initial reduction of 99.27% in the number of attributes. The application of analysis of variance also contributes to the reduction of information, enabling minimize this set of information in approximately 88% compared to using only the SVD. In classification, a system using a classifier with fuzzy logic was applied to the separation of data into two classes: Normal and Abnormal, provided at a rate 81% for ROC area. The results show that the fusion of the characteristics of CC and MLO views are relevant to the classification stage of the proposed method. Moreover, the method of reduction contributed to a breakthrough in data classification

ASSUNTO(S)

transformada wavelet imagem mamográfica diagnóstico auxiliado por computador lógica fuzzy raios x wavelet transform mammogram image computer systems to aided detection fuzzy logic x-ray outros

ACESSO AO ARTIGO

Documentos Relacionados