Abstract Views :209 |
PDF Views:122
Authors
Affiliations
1 Department of Computer Application, Madhav Institute of Technology and Science, Gwalior, IN
Source
AIRCC's International Journal of Computer Science and Information Technology, Vol 3, No 1 (2011), Pagination: 199-211
Abstract
Nowadays the most common type of cancer in women is breast cancer. This is the second main cause of cancer deaths in women. Digital mammography is the technique which is used to examine the breast. This is very much useful for the detection of breast diseases in women. The automatic detection of tumor or some type of deformity in the medical imaging is done by many researchers to develop some algorithms and methods. In this paper we are using SOM and Fuzzy c-means clustering techniques for tumor detection in digital mammography images. We then further calculate the statistical features of tumor like location of tumor, area, energy, entropy, idm, mean, contrast, mean and standard deviation which helps the radiologist to study the statistical information regarding breast cancer, so that the doctors can give better treatment to patients. For calculating these statistical properties we use region growing and region merging techniques.
Keywords
Gray Level Cooccurrence Matrix (GLCM), best Matching Unit (BMU), Epoch Number, Idm (iInverse Difference Moment), Micro-Calcifications.
Full Text
