Computer-aided diagnosis of digital mammography images using unsupervised clustering and biclustering techniques

Abstract

A new methodology for computer aided diagnosis in digital mammography using unsupervised classification and classdependent feature selection is presented. This technique considers unlabeled data and provides unsupervised classes that give a better insight into classes and their interrelationships, thus improving the overall effectiveness of the diagnosis. This technique is also extended to utilize biclustering methods, which allow for definition of unsupervised clusters of both pathologies and features. This has potential to provide more flexibility, and hence better diagnostic accuracy, than the commonly used feature selection strategies. The developed methods are applied to diagnose digital mammographic images from the Mammographic Image Analysis Society (MIAS) database and the results confirm the potential for improving the current diagnostic rates.

Identification of microcalcifications in digital mammogram using Modified k-mean Clustering

“…The k-means clustering algorithm partitions a given dataset into k mutually exclusive clusters such that the sum of the distances between data and the corresponding cluster centroid is minimized. The clustering can be applied to either rows or columns of the feature matrix separately but biclustering methods perform clustering in the two dimensions simultaneously [11].…”

Biclustering Analysis for Pattern Discovery: Current Techniques, Comparative Studies and Applications

“…In [74], biclustering is applied to the computer-aided diagnosis of digital mammography images. With the rows representing the set of images and columns representing the set of features, biclustering can find a subset of images participating in a common pathology of interest while defining a subset of features that best describe this pathology.…”

Unsupervised Breast Masses Classification through Optimum-Path Forest

“…Kim et al [13], Al-Olfe et al [14] and Choi et al [15] designed a new type of classification combining unsupervised and supervised models to classify malignant and benign masses on mammograms. Abdel-Qader and AbuAmara [16] presented a hybrid Computer-Aided Detection and Diagnosis (CADD) system for breast cancer identification based on PCA, Independent Component Analysis and a fuzzy classifier to identify and label suspicious regions.…”