Analysis and Diagnosis of Mammograms for Early Breast Cancer Detection

Abstract

Breast cancer is the most frequently diagnosed cancer in women. Although mammography screening is the most effective method currently available for the early detection of breast cancer, it is far from being an infallible procedure. Mammography reading is error prone, partly because of the complexity of the task and partly because of the variability in human performance. The goal of this research was to analyses and diagnoses of Mammograms for early breast cancer detection by used statistical features and dynamic features, eight statistical features that include mean, standard deviation, kurtosis, skeweness, maximum, minimum, median absolute deviation and range were estimated from each region of interest texture, Also used dynamic method (wavelet decomposition coefficient and create gray level co matrix by image processing). The graphic user interface was created to enables physicians dealt with in a flexible manner for decision-making. This program written by MATLAB version 2010 software, functions and toolboxes attached to it. The code detected forty five normal images from Seventieth and sixteen cancer images from twenty six.