A completely automated CAD system for mass detection in a large mammographic database

Abstract

Mass localization plays a crucial role in computer-aided detection (CAD) systems for the classification of suspicious regions in mammograms. In this article we present a completely automated classification system for the detection of masses in digitized mammographic images. The tool system we discuss consists in three processing levels: (a) Image segmentation for the localization of regions of interest (ROIs). This step relies on an iterative dynamical threshold algorithm able to select iso-intensity closed contours around gray level maxima of the mammogram. (b) ROI characterization by means of textural features computed from the gray tone spatial dependence matrix (GTSDM), containing second-order spatial statistics information on the pixel gray level intensity. As the images under study were recorded in different centers and with different machine settings, eight GTSDM features were selected so as to be invariant under monotonic transformation. In this way, the images do not need to be normalized, as the adopted features depend on the texture only, rather than on the gray tone levels, too. (c) ROI classification by means of a neural network, with supervision provided by the radiologist's diagnosis. The CAD system was evaluated on a large database of 3369 mammographic images [2307 negative, 1062 pathological (or positive), containing at least one confirmed mass, as diagnosed by an expert radiologist]. To assess the performance of the system, receiver operating characteristic (ROC) and free-response ROC analysis were employed. The area under the ROC curve was found to be Az = 0.783 +/- 0.008 for the ROI-based classification. When evaluating the accuracy of the CAD against the radiologist-drawn boundaries, 4.23 false positives per image are found at 80% of mass sensitivity.

Identification of Micro-calcification in Mammogram for Breast Cancer Analysis using SVM Classifier

“…Same half DDSM database was used for learning and other half was used for testing final test result was 92%, 88% and 81%. Bellotti et al [25] implemented edge based segmentation algorithm to select the suspicious region and then GLCM matrix used to identify the feature of ROI, finally classification of masses is achieve by neural network which was based on gradient descent learning rule. A database of 3369 mammogram are used with 2307 and 1062 are negative and positive cases respectively.…”

DeepCAD: A Computer-Aided Diagnosis System for Mammographic Masses Using Deep Invariant Features

“…An accurate classification allows medical experts to accurately differentiate [10,11] between benign and malignant masses. Unfortunately, the classification step is difficult due to the overlap between dense breast tissue and irregular mass shape.…”

Detection of Breast Masses in Mammogram Images Using Growing Neural Gas Algorithm and Ripley’s K Function

“…Rezai-rad and Jamarani [6] present an approach for the detection of microcalcifications on screening mammograms using artificial neural networks combined with the wavelet transform, with sensitivity of around 94%. Bellotti et al [7] proposed a tool for detection of masses using an algorithm based on edge segmentation for the selection of suspicious regions. Second-order measures obtained from co-occurrence matrices are used to describe the texture of these regions.…”