Tree-based modelling for the classification of mammographic benign and malignant micro-calcification clusters

Abstract

Computer Aided Detection (CAD) systems are being developed to assist radiologists in diagnosis. For breast cancer the emphasis is shifting from detection to classification of abnormalities. The presented work concentrates on the benign versus malignant classification of micro-calcification clusters, which are a specific type of mammographic abnormality associated with the early development of breast cancer. After segmentation (automatic or manual), tree-based representations were used to distinguish between benign and malignant clusters, which takes into account clinical criteria such as the number of micro-calcifications in the clusters and their distribution and is based on the topology of the trees and the connectivity of the micro-calcifications. The idea of using tree structure based on the distance of individual calcifications for the classification of benign and malignant micro-calcification clusters is novel and closely related to clinical perception. Tree structures used in this study are distinct from decision trees classifiers being used in many machine learning approaches. Initial evaluation on the Digital Database for Screening Mammography (DDSM) data shows promising results, with an accuracy equal to 91 %, which is comparable to state of the art CAD systems and is in line with clinical perception of the morphology and appearance of benign and malignant micro-calcification clusters.

Classification of mammographic microcalcification clusters with machine learning confidence levels

“…Yang et al 14 employed the Simplified Pulse Coupled Neural Network (SPCNN) to classify each case based on wavelet high-frequency coefficients and reported an AU C = 0.97 with an accuracy of over 93% tested on both the MIAS and Japanese Society of Medical Imaging Technology datasets. Suhail et al 15 used a binary tree-based approach to model the clinical perception features such as the number of microcalcifications and the distribution pattern; which can be mathematically computed based on the topology of the trees and the connectivity of the microcalcifications. The authors reported 91% accuracy based on 288 patches extracted from the DDSM dataset.…”

Classification of Microcalcification Clusters in Digital Mammograms Using a Stack Generalization Based Classifier

“…The proposed method for MC cluster classification was compared with other relevant publications (see Table 7 ). Akram et al [ 12 ] proposed a tree-based representations for MC clusters, where scale-invariant topological features of MC were extracted showing 91% accuracy for cluster classification. Although high accuracy was achieved, the performance for MC cluster classification on digital mammogram was not reported in this study.…”

“…Double reading can improve sensitivity, but a lack of experienced radiologists can be a challenge [ 9 ]. CADx can assist radiologists in detecting abnormalities in an efficient way [ 10 , 11 ] and systems have been developed to provide a second opinion for diagnosis [ 12 ]. Previous studies have developed computerized methods to aid the diagnosis of MC clusters.…”