A Novel Cascade Classifier for Automatic Microcalcification Detection

Abstract

In this paper, we present a novel cascaded classification framework for automatic detection of individual and clusters of microcalcifications (mu C). Our framework comprises three classification stages: i) a random forest (RF) classifier for simple features capturing the second order local structure of individual mu Cs, where non-mu C pixels in the target mammogram are efficiently eliminated; ii) a more complex discriminative restricted Boltzmann machine (DRBM) classifier for mu C candidates determined in the RF stage, which automatically learns the detailed morphology of mu C appearances for improved discriminative power; and iii) a detector to detect clusters of mu Cs from the individual mu C detection results, using two different criteria. From the two-stage RF-DRBM classifier, we are able to distinguish mu Cs using explicitly computed features, as well as learn implicit features that are able to further discriminate between confusing cases. Experimental evaluation is conducted on the original Mammographic Image Analysis Society (MIAS) and mini-MIAS databases, as well as our own Seoul National University Bundang Hospital digital mammographic database. It is shown that the proposed method outperforms comparable methods in terms of receiver operating characteristic (ROC) and precision-recall curves for detection of individual mu Cs and free-response receiver operating characteristic (FROC) curve for detection of clustered mu Cs.