A Novel Cascade Classifier for Automatic Microcalcification Detection

Authors

Seung Yeon Shin
Soochahn Lee
Il Dong Yun
Ho Yub Jung
Yong Seok Heo
Sun Mi Kim
Kyoung Mu Lee

Abstract

In this paper, we present a novel cascaded classification framework for automatic detection of individual and clusters of microcalcifications (mC). Our framework comprises three classification stages: i) a random forest (RF) classifier for simple features capturing the second order local structure of individual mCs, where non-mC pixels in the target mammogram are efficiently eliminated; ii) a more complex discriminative restricted Boltzmann machine (DRBM) classifier for mC candidates determined in the RF stage, which automatically learns the detailed morphology of mC appearances for improved discriminative power; and iii) a detector to detect clusters of mCs from the individual mC detection results, using two different criteria. From the two-stage RF-DRBM classifier, we are able to distinguish mCs 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 mCs and free-response receiver operating characteristic (FROC) curve for detection of clustered mCs.

Paper

PDF

Dataset

1. The Seoul National University Bundang Hospital Mammographic Database (SNUBH-MDB)
- SNUBH-MDB-mCi
- SNUBH-MDB-mCc: Available by request, to researchers who meet the criteria for access to the confidential data. please contact prof. Sun Mi Kim (kimsmlms@daum.net), of the SNUBH.
2. The original Mammographic Image Analysis Society (MIAS) Database
- Images
- Annotations of individual mCs
- Annotations of mC clusters
3. The mini-MIAS Database
- Images and annotations of mC clusters
- Annotations of individual mCs

Citation

Seung Yeon Shin, Soochahn Lee, Il Dong Yun, Ho Yub Jung, Yong Seok Heo, Sun Mi Kim, and Kyoung Mu Lee, "A Novel Cascade Classifier for Automatic Microcalcification Detection," PLoS ONE, 2015. Bibtex