Classification of microcalcifications in digital mammograms using trend-oriented radial basis function neural network

PATTERN RECOGNITION

卷:32

Article

Tsujii, O ; Freedman, MT ; Mun, SK

关键词: mammograms;    microcalcification;    classification;    feature selection;    Karhunen-Loeve transformation;    Euclidean distance measure;    neural network;    radial basis function;    round-robin method;    receiver operating characteristic;   

DOI  :  10.1016/S0031-3203(98)00099-5

来源: Elsevier

【 摘 要 】

We proposed some novel classification features for the microcalcification of mammograms, and selected the effective combined features using Karhunen-Loeve (KL) transformation followed by the restricted Euclidean distance measure, and finally applied the proposed trend-oriented radial basis function neural network (TRBF-NN) to distinguish the benign group from the malignant group and evaluate the performance with the round-robin method. The two-dimensional KL features were more distinguishable than the raw two-dimensional features. The TRBF-NN was able to define the more generalized distribution than those distributions defined by the conventional RBF-NNs. According to the receiver operating characteristic analysis, the proposed system performed better than two trained radiologists. (C) 1999 Pattern Recognition Society. Published by Elsevier Science Ltd. All rights reserved.

【 授权许可】

Free