Effective recognition of MCCs in mammograms using an improved neural classifier

J. C. Ren, D. Wang, J. M. Jiang

Research output: Contribution to journalArticle

30 Citations (Scopus)

Abstract

Computer-aided diagnosis is one of the most important engineering applications of artificial intelligence. In this paper, early detection of breast cancer through classification of microcalcification clusters from mammograms is emphasized. Although artificial neural network (ANN) has been widely applied in this area, the average accuracy achieved is only around 80% in terms of the area under the receiver operating characteristic curve A(z). This performance may become much worse when the training samples are imbalanced. As a result, an improved neural classifier is proposed, in which balanced learning with optimized decision making are introduced to enable effective learning from imbalanced samples. When the proposed learning strategy is applied to individual classifiers, the results on the DDSM database have demonstrated that the performance from has been significantly improved. An average improvement of more than 10% in the measurements of F(1) score and A(z) has fully validated the effectiveness of our proposed method for the successful classification of clustered microcalcifications.
Original languageEnglish
Pages (from-to)638-645
Number of pages8
JournalEngineering Applications of Artificial Intelligence
Volume24
Issue number4
DOIs
Publication statusPublished - Jun 2011

Fingerprint

Classifiers
Computer aided diagnosis
Artificial intelligence
Decision making
Neural networks

Keywords

  • optimized decision making
  • breast-cancer
  • mammography
  • IMBA
  • digital mammography
  • artificial neural network
  • feature-extraction
  • image retrieval
  • clustered microcalcifications
  • balanced learning
  • microcalcification clusters (MCC)
  • computer-aided diagnosis
  • effective recognition
  • mcc's
  • mammograms
  • improved
  • neural classifier

Cite this

@article{346cc0cac2a4411b9497e9edb7fa3fbb,
title = "Effective recognition of MCCs in mammograms using an improved neural classifier",
abstract = "Computer-aided diagnosis is one of the most important engineering applications of artificial intelligence. In this paper, early detection of breast cancer through classification of microcalcification clusters from mammograms is emphasized. Although artificial neural network (ANN) has been widely applied in this area, the average accuracy achieved is only around 80{\%} in terms of the area under the receiver operating characteristic curve A(z). This performance may become much worse when the training samples are imbalanced. As a result, an improved neural classifier is proposed, in which balanced learning with optimized decision making are introduced to enable effective learning from imbalanced samples. When the proposed learning strategy is applied to individual classifiers, the results on the DDSM database have demonstrated that the performance from has been significantly improved. An average improvement of more than 10{\%} in the measurements of F(1) score and A(z) has fully validated the effectiveness of our proposed method for the successful classification of clustered microcalcifications.",
keywords = "optimized decision making, breast-cancer, mammography, IMBA, digital mammography, artificial neural network, feature-extraction, image retrieval, clustered microcalcifications, balanced learning, microcalcification clusters (MCC), computer-aided diagnosis, effective recognition, mcc's, mammograms, improved, neural classifier",
author = "Ren, {J. C.} and D. Wang and Jiang, {J. M.}",
year = "2011",
month = "6",
doi = "10.1016/j.engappai.2011.02.011",
language = "English",
volume = "24",
pages = "638--645",
journal = "Engineering Applications of Artificial Intelligence",
issn = "0952-1976",
number = "4",

}

Effective recognition of MCCs in mammograms using an improved neural classifier. / Ren, J. C.; Wang, D.; Jiang, J. M.

In: Engineering Applications of Artificial Intelligence , Vol. 24, No. 4, 06.2011, p. 638-645.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Effective recognition of MCCs in mammograms using an improved neural classifier

AU - Ren, J. C.

AU - Wang, D.

AU - Jiang, J. M.

PY - 2011/6

Y1 - 2011/6

N2 - Computer-aided diagnosis is one of the most important engineering applications of artificial intelligence. In this paper, early detection of breast cancer through classification of microcalcification clusters from mammograms is emphasized. Although artificial neural network (ANN) has been widely applied in this area, the average accuracy achieved is only around 80% in terms of the area under the receiver operating characteristic curve A(z). This performance may become much worse when the training samples are imbalanced. As a result, an improved neural classifier is proposed, in which balanced learning with optimized decision making are introduced to enable effective learning from imbalanced samples. When the proposed learning strategy is applied to individual classifiers, the results on the DDSM database have demonstrated that the performance from has been significantly improved. An average improvement of more than 10% in the measurements of F(1) score and A(z) has fully validated the effectiveness of our proposed method for the successful classification of clustered microcalcifications.

AB - Computer-aided diagnosis is one of the most important engineering applications of artificial intelligence. In this paper, early detection of breast cancer through classification of microcalcification clusters from mammograms is emphasized. Although artificial neural network (ANN) has been widely applied in this area, the average accuracy achieved is only around 80% in terms of the area under the receiver operating characteristic curve A(z). This performance may become much worse when the training samples are imbalanced. As a result, an improved neural classifier is proposed, in which balanced learning with optimized decision making are introduced to enable effective learning from imbalanced samples. When the proposed learning strategy is applied to individual classifiers, the results on the DDSM database have demonstrated that the performance from has been significantly improved. An average improvement of more than 10% in the measurements of F(1) score and A(z) has fully validated the effectiveness of our proposed method for the successful classification of clustered microcalcifications.

KW - optimized decision making

KW - breast-cancer

KW - mammography

KW - IMBA

KW - digital mammography

KW - artificial neural network

KW - feature-extraction

KW - image retrieval

KW - clustered microcalcifications

KW - balanced learning

KW - microcalcification clusters (MCC)

KW - computer-aided diagnosis

KW - effective recognition

KW - mcc's

KW - mammograms

KW - improved

KW - neural classifier

UR - http://www.scopus.com/inward/record.url?scp=79953654774&partnerID=8YFLogxK

U2 - 10.1016/j.engappai.2011.02.011

DO - 10.1016/j.engappai.2011.02.011

M3 - Article

VL - 24

SP - 638

EP - 645

JO - Engineering Applications of Artificial Intelligence

JF - Engineering Applications of Artificial Intelligence

SN - 0952-1976

IS - 4

ER -