Region of Interest (ROI) and Feature Extraction from the images of mammograms for the diagnosis of Breast Cancer

ANNAVARAPU JAGADISH KUMAR, A V DATTATREYA RAO, K KARTEEKA PAVAN

Abstract


Now a days the diagnosis of breast cancer is done through the analysis of mammograms. The primary object of breast imaging is to optimize the early and
accurate diagnosis of breast abnormalities based on mammogram. Here the term
diagnosis refers to classifying a given mammogram into one of the following classes,
1. Calcifications (CALC) 2. Circumscribed (CIRC) 3. Speculated (SPIC) masses 4. III-
defined (MISC) masses 5. Architectural distortions (ARCH) 6. Asymmetry(s) (ASYM) 7.
Normal (NORM) masses. The captured image having been pre-processed for removing
digitization noise, suppression of artifacts and separating back-ground and segmenting
pectoral muscles in order to find Region of Interest (ROI), texture features are to be extracted from the Region of Interest (ROI). Here an attempt is made to find ROI using fixed window segmentation and hence to obtain statistical features based on GLCM.

Key words: ROI, Fixed Window Segmentation, textural feature extraction and GLCM.

پاراگلایدر Full Text: PDF

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

ISSN : 2251-1563