Detection and Classification of Mammographic Calcifications

Abstract

We propose a detection and classification system for the analysis of mammo-graphic calcifications. First, a new multi-tolerance region growing method is proposed for the detection of potential calcification regions and extraction of their contours. The method employs a distance metric computed on feature sets including measures of shape, centre of gravity, and size obtained for various growth tolerance values in order to determine the most suitable parameters. Then, shape features from moments, Fourier descriptors, and compactness are computed based upon the contours of the regions. Finally, a two-layer perceptron is utilized for the purpose of classification of calcifications with the shape features. A new leave-one-out algorithm-based parameter determination procedure is included in the neural network training step. In our preliminary study, detection rates were 81% and 85±3%, and correct classification rates were 94% and 87% with a test set of 58 benign calcifications and 241±10 malignant calcifications, respectively. The proposed system should provide considerable help to radiologists in the diagnosis of breast cancer.

Classifier scheme for clustered microcalcifications in digitized mammograms by using Artificial Neural Networks

“…We could note from this work that only with the clusters characteristics -without considering individual microcalcifications features -good results could be obtained in testing the classifier, which are comparable to several others presented in literature ( [2], [5], [6], [7] and [13]). …”

“…For training the network, such curves were compared to those more used according to the literature ( [2], [3], [4], [5] and [7]) in order to select the most adequate characteristics to be used. The red curve present the "suspect" class and the blue curve present the "non-suspect" class.…”

“…Artificial Neural Networks (ANNs) are often used as classifiers in CAD schemes, with good results ( [2], [5] to [7]). Characteristics extracted from digitized mammograms are the input data for lesion classification.…”

“…For classifying the cases in "suspect" and "nonsuspect" from the clusters analysis, geometric descriptors, as first, second and third orders moments [9], were extracted, the Hu Invariant Moments ( [7], [9], [10]) and the cluster radius gyration were calculated, and also some information from the patient medical report was used, as patient age and a "density" measure. Characteristics extracted from the clusters did not have considered the individual microcalcifications, but only the cluster itself.…”

Mass Description for Breast Cancer Recognition

“…Where P and A are the mass perimeter and area respectively. A mass with a rough contour will have a higher compactness than a mass with a smooth boundary [15].…”