Feature Extraction for Computer-Aided Analysis of Mammograms

Abstract

A framework for computer-aided analysis of mammograms is described. General computer vision algorithms are combined with application specific procedures in a hierarchical fashion. The system is under development and is currently limited to detection of a few types of suspicious areas. The image features are extracted by using feature extraction methods where wavelet techniques are utilized. A low-pass pyramid representation of the image is convolved with a number of quadrature filters. The filter outputs are combined according to simple local Fourier domain models into parameters describing the local neighbourhood with respect to the model. This produces estimates for each pixel describing local size, orientation, Fourier phase, and shape with confidence measures associated to each parameter. Tentative object descriptions are then extracted from the pixel-based features by application-specific procedures with knowledge of relevant structures in mammograms. The orientation, relative brightness and shape of the object are obtained by selection of the pixel feature estimates which best describe the object. The list of object descriptions is examined by procedures, where each procedure corresponds to a specific type of suspicious area, e.g. clusters of microcalcifications.

Computer-Aided Diagnostic System based on Wavelet Analysis for Microcalcification Detection in Digital Mammograms

“…On the other hand, the classification rate of microcalcification cases achieves the bad performance with features extracted from levels 1-2 because the low frequency information which embodied in the lowest levels by wavelet decomposition. Table 6 shows results of classification rates for benign and malignant microcalcification by using K-NN classifier for testing set with varying the k value (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11) and at different levels of wavelet decomposition (levels 1-3, levels 1-2, and levels 2-3).…”

“…Netch [5] proposed a detection scheme for the automatic detection of clustered microcalcifications using multiscale analysis based on the Laplacian-of-Gaussian filter and a mathematical model describing a microcalcification as a bright spot of certain size and contrast. Barman, Granlund, and Haglund [6] used a low-pass filter to detect microcalcification by analyzing digital mammogram. Although the system based on their algorithm is still under development, good preliminary results have been produced with further modifications still to be made.…”