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.
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.
Entities:
Keywords:
Computer-aided diagnosis; Content-based image retrieval; Image processing; Screening mammography; Singular value decomposition; Support vector machine
Authors: Xiao Hui Wang; Walter F Good; Brian E Chapman; Yuan-Hsiang Chang; William R Poller; Thomas S Chang; Lara A Hardesty Journal: AJR Am J Roentgenol Date: 2003-01 Impact factor: 3.959
Authors: T M Lehmann; M O Güld; C Thies; B Fischer; K Spitzer; D Keysers; H Ney; M Kohnen; H Schubert; B B Wein Journal: Methods Inf Med Date: 2004 Impact factor: 2.176
Authors: Thomas M Lehmann; Mark O Güld; Thomas Deselaers; Daniel Keysers; Henning Schubert; Klaus Spitzer; Hermann Ney; Berthold B Wein Journal: Comput Med Imaging Graph Date: 2005 Mar-Apr Impact factor: 4.790
Authors: Cyril Castella; Karen Kinkel; Miguel P Eckstein; Pierre-Edouard Sottas; Francis R Verdun; François O Bochud Journal: Acad Radiol Date: 2007-12 Impact factor: 3.173
Authors: Thomas M Deserno; Mark O Güld; Bartosz Plodowski; Klaus Spitzer; Berthold B Wein; Henning Schubert; Hermann Ney; Thomas Seidl Journal: J Digit Imaging Date: 2007-05-12 Impact factor: 4.056