de Oliveira JEE, Araújo AA, Deserno TM. Content-based image retrieval applied to BI-RADS tissue classification in screening mammography. World J Radiol 2011; 3(1): 24-31 [PMID: 21286492 DOI: 10.4329/wjr.v3.i1.24]
Corresponding Author of This Article
Júlia Epischina Engrácia de Oliveira, PhD, Department of Computer Science, Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627, 31270-901, Belo Horizonte, MG, Brazil. julia@dcc.ufmg.br
Article-Type of This Article
Brief Article
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
World J Radiol. Jan 28, 2011; 3(1): 24-31 Published online Jan 28, 2011. doi: 10.4329/wjr.v3.i1.24
Content-based image retrieval applied to BI-RADS tissue classification in screening mammography
Júlia Epischina Engrácia de Oliveira, Arnaldo de Albuquerque Araújo, Thomas M Deserno
Júlia Epischina Engrácia de Oliveira, Arnaldo de Albuquerque Araújo, Department of Computer Science, Universidade Federal de Minas Gerais, 31270-901, Belo Horizonte, MG, Brazil
Thomas M Deserno, Department of Medical Informatics, RWTH Aachen University, 52074, Aachen, Germany
Author contributions: de Oliveira JEE performed the majority of experiments, and provided major parts of the manuscript; Araújo A de A initiated the investigation and was involved in editing the manuscript; Deserno TM provided the data collection and contributed to the study design as well as writing of the manuscript.
Supported by CNPq-Brazil, Grants 306193/2007-8, 471518/2007-7, 307373/2006-1 and 484893/2007-6, by FAPEMIG, Grant PPM 347/08, and by CAPES; The IRMA project is funded by the German Research Foundation (DFG), Le 1108/4 and Le 1108/9
Correspondence to: Júlia Epischina Engrácia de Oliveira, PhD, Department of Computer Science, Universidade Federal de Minas Gerais, Av. Antônio Carlos, 6627, 31270-901, Belo Horizonte, MG, Brazil. julia@dcc.ufmg.br
Telephone: +55-31-34095854 Fax: +55-31-34095858
Received: November 8, 2010 Revised: December 8, 2010 Accepted: December 15, 2010 Published online: January 28, 2011
Abstract
AIM: To present a content-based image retrieval (CBIR) system that supports the classification of breast tissue density and can be used in the processing chain to adapt parameters for lesion segmentation and classification.
METHODS: Breast density is characterized by image texture using singular value decomposition (SVD) and histograms. Pattern similarity is computed by a support vector machine (SVM) to separate the four BI-RADS tissue categories. The crucial number of remaining singular values is varied (SVD), and linear, radial, and polynomial kernels are investigated (SVM). The system is supported by a large reference database for training and evaluation. Experiments are based on 5-fold cross validation.
RESULTS: Adopted from DDSM, MIAS, LLNL, and RWTH datasets, the reference database is composed of over 10 000 various mammograms with unified and reliable ground truth. An average precision of 82.14% is obtained using 25 singular values (SVD), polynomial kernel and the one-against-one (SVM).
CONCLUSION: Breast density characterization using SVD allied with SVM for image retrieval enable the development of a CBIR system that can effectively aid radiologists in their diagnosis.
