Detection and Identification of Mass Structure in Digital Mammogram

Abstract

Breast Cancer is the rampant issue facing by most of the women these days. Mammography is the most successful modus operandi for verdict of breast cancer. Radiologists view mammograms to perceive the abnormalities. In this paper, we urbanized an algorithm to isolate and extract the malignant masses in mammograms for detection of breast cancer. This exertion is based on the following course of action: (a) Confiscate the background information from the DICOM image. (b)Refurbish the RGB image to gray image (c) Apply thresholding to remove the background information (ROI). (d)Apply median filter on the image to reduce random noise and preserve the edges. (e)Extract the binary image contours. (f)Carry out Area-open on the image and. (g)To end with act upon filling on the resultant image. This method was tested over the real time images of various patients taken from a cancer hospital and implemented using Matlab code. Thus, capable in executing the pre-processed image efficiently and detected the segmentation region which helped in retrieving the mass present in the digital mammogram and hence documented the malignant mass by calculating different features like Area, perimeter, Compactness, Width, Height, Isotropic factor and Eccentricity of the extracted mass.

Advanced Region Detection Technique to Enhance Accurate Abnormality Isolation of Cancer in Digital Images

“…The Proposed technique is based on the Fisher information measure. Bethapudi et al, proposed a detection and identification method of mass structure in digital images [14] which detect malignant tissues in following steps: (1) Thresholding to remove the background information, (2) Apply median filter for random noise removal, (3) Extract the binary image contours [7]. Thereafter, morphological open and close operations to fill the gaps in holes inside the image region.…”