MammoSys: A content-based image retrieval system using breast density patterns

被引：58

作者：

de Oliveira, Julia E. E. ^{[1
]}

Machado, Alexei M. C. ^{[2
]}

Chavez, Guillermo C. ^{[1
]}

Lopes, Ana Paula B. ^{[1
]}

Deserno, Thomas M. ^{[3
]}

Araujo, Arnaldo de A. ^{[1
]}

机构：

[1] Univ Fed Minas Gerais, Dept Ciencia Computaccao, BR-31270901 Belo Horizonte, MG, Brazil

[2] Pontificia Univ Catol Minas Gerais, Dept Ciencia Computacaao, Inst Informat, BR-30535610 Belo Horizonte, MG, Brazil

[3] Aachen Univ Technol RWTH, Dept Med Informat, D-52057 Aachen, Germany

来源：

COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE | 2010年 / 99卷 / 03期

关键词：

Medical images; Breast density; Content-based image retrieval; Two-dimensional principal component analysis; Support vector machine; CLASSIFICATION; RECOGNITION;

D O I：

10.1016/j.cmpb.2010.01.005

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

In this paper, we present a content-based image retrieval system designed to retrieve mammographies from large medical image database. The system is developed based on breast density, according to the four categories defined by the American College of Radiology, and is integrated to the database of the Image Retrieval in Medical Applications (IRMA) project, that provides images with classification ground truth. Two-dimensional principal component analysis is used in breast density texture characterization, in order to effectively represent texture and allow for dimensionality reduction. A support vector machine is used to perform the retrieval process. Average precision rates are in the range from 83% to 97% considering a data set of 5024 images. The results indicate the potential of the system as the first stage of a computer-aided diagnosis framework. (C) 2010 Elsevier Ireland Ltd. All rights reserved.

引用

页码：289 / 297

页数：9

共 28 条

[11]

Dudda R.O., 2001, Pattern Classification

[12]

Gonzalez R. C., 2004, Digital image processing using MATLAB, VSecond

[13]

Heath M, 1998, COMP IMAG VIS, V13, P457

[14] A comparison of methods for multiclass support vector machines [J].

Hsu, CW ;

Lin, CJ .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2002, 13 (02) :415-425

[15] Content-based retrieval of mammograms using visual features related to breast density patterns [J].

Kinoshita, Sergio Koodi ;

de Azevedo-Marques, Paulo Mazzoncini ;

Pereira, Roberto Rodrigues, Jr. ;

Heisinger Rodrigues, Jose Antonio ;

Rangayyan, Rangaraj Mandayam .

JOURNAL OF DIGITAL IMAGING, 2007, 20 (02) :172-190

[16] The IRMA code for unique classification of medical images [J].

Lehmann, TM ;

Schubert, H ;

Keysers, D ;

Kohnen, M ;

Wein, BB .

MEDICAL IMAGING 2003: PACS AND INTEGRATED MEDICAL INFORMATION SYSTEMS: DESIGN AND EVALUATION, 2003, 5033 :440-451

[17]

Lehmann TM, 2004, METHOD INFORM MED, V43, P354

[18] A review of content-based image retrieval systems in medical applications -: clinical benefits and future directions [J].

Müller, H ;

Michoux, N ;

Bandon, D ;

Geissbuhler, A .

INTERNATIONAL JOURNAL OF MEDICAL INFORMATICS, 2004, 73 (01) :1-23

[19] A novel breast tissue density classification methodology [J].

Oliver, Arnau ;

Freixenet, Jordi ;

Marti, Robert ;

Pont, Josep ;

Perez, Elsa ;

Denton, Erika R. E. ;

Zwiggelaar, Reyer .

IEEE TRANSACTIONS ON INFORMATION TECHNOLOGY IN BIOMEDICINE, 2008, 12 (01) :55-65

[20]

Rijsbergen C. J. V, 1979, Information Retrieval

← 1 2 3 →