Abstract:

Detection of microcalcification based on textural image segmentation and classification is the most effective method for the early-diagnosis of breast cancer. In this paper, a proposed technique, Markov Random Field method hybrid with Ant Colony System, Genetic Algorithm and Backpropagation Network (MRF–ACSGA-BPN) is implemented for the detection of microcalcification in digital mammogram. Identification of microcalcification is performed in two steps: segmentation, feature extraction and classification. First, the mammogram image is segmented using MRF-ACSGA method to extract the suspicious region. Second, the conventional textural analysis method and the Spatial Gray Level Dependence Method (SGLDM) are used to extract the features from the segmented image and their performance is studied and a three-layer backpropagation neural network classifier, trained by jack knife method and round robin method, is used to classify the extracted features into benign or malignant. The results of the neural network for the texture-analysis methods are evaluated by using a Receiver Operating-Characteristics (ROC) analysis. The proposed algorithm and the techniques are tested on 161 pairs of digitized mammograms from MIAS database.

Keywords: Mammogram, Markov Random Field, Ant Colony Optimization, Genetic Algorithm, Backpropagation Neural Network.

Biographies:

	THANGAVEL KUTTIANNAN, received the Master of Science from Department of Mathematics, Bharathidasan University in 1986, and Master of Computer Applications Degree from Madurai Kamaraj University, India in 2001. He got his Ph. D. Degree from Mathematics department, Gandhigram Rural Institute-Deemed University in 1999. Currently he is working as Reader in Mathematics Department, Gandhigram Rural Institute-Deemed University, and his experience started from 1988; His area of interests includes medical image processing, artificial intelligence, neural network, and fuzzy logic.
	KARNAN MARCUS, received the Master of Computer Science and Engineering Degree from Computer Science and Engineering Department, from Government College of Engineering, Manonmaniam Sundaranar University, Tamil Nadu, India, in 2000. Currently he is working as Assistant Professor, Department of Computer Science & Engineering Department, RVS College of Engineering & Technology, Tamil Nadu, India. And doing part-time research in the Department of computer Science and Applications, Gandhigram Rural Institute-Deemed University, Tamil Nadu, India. His area of interests includes medical image processing, artificial intelligence, neural network, genetic algorithm, pattern recognition and fuzzy logic.
	SIVA KUMAR RAMAKRISHNAN,. Received the Master of Computer Science and Engineering Degree from Computer Science and Engineering Department, from Government College of Engineering, Manonmaniam Sundaranar University, Tamil Nadu, India, in 2000. Currently he is working as Assistant Professor, Department of Computer Science & Engineering, RVS College of Engineering & Technology, Tamil Nadu, India. His area of interests includes medical image processing, artificial intelligence, neural network, and fuzzy logic.
	KAJA MOHIDEEN ABUBAKKER, received the Master of Computer Applications Degree from Computer Science Department, Bharathidasan University, Tamil Nadu, India, in 2000. Currently he is working as Research Assistant, in Department of Computer Science & Engineering RVS College of Engineering & Technology, Tamil Nadu, India. His area of interests includes medical image processing, artificial intelligence, and neural network.
	* The following author designated as corresponding author: Karnan Marcus, E-mail id: karnanme@yahoo.com, rsk_siva@mailcity.com, ktvel@rediffmail.com Postal Address: 153, Palayam, Palani, Dindigul Dt, Tamil Nadu, India – 624601. Fax.No:91-4551-227229,91-4551-227230, Telephone.No:91-4551-227229,91-4551-2272.

BibTex:

@ARTICLE{P1120524001,

  AUTHOR = {K.Thangavel and M.Karnan and R.Sivakumar and A. Kaja Mohideen},

TITLE = {Ant Colony System for Segmentation and Classification of Microcalcification in Mammograms},

}

( Full Paper, 526 KB)