Paper Title
Breast Abnormality Detection in Mammograms Using Neural Network

Abstract
Breast cancer is one in all predominant diseases in women, and might be known improvement some examinations that contain digital mammogram, ultrasound, and magnetic resonance imaging and diagnostic test, etc. However, the accurate classification of breast abnormalities remains a medical challenge tested by researchers. Problems are typically bumped into within the examination for teams of arrangements that offer reasonable quality essential for classifying breast tissues into collections of normal and abnormal. Digital mammogram is one of the low energy screening tools for breast. Consequently, the aim of this study is to propose a system for prediction of breast abnormality discrimination using Neural Network (NN) models supported 2-dimensional digital mammograms. The main stages of this research paper are preprocessing, feature extraction, segmentation and neural network classifier. The primary technique needs the preprocessing step for breast profile extraction, approved out by removing the low-frequency parts of the digital mammogram, giving up sub bands comprising high-frequency coefficients, supported the thought that small calcifications indicate high-frequency coefficients. The sequential approach involves options extraction derived from current fragmentation analysis. The final approach is said because the classification stage that utilizes back propagation neural network to communicate apart abnormal tissue from normal ones. Digital diagnostic procedure may be the finestprocedure for early detection of breast abnormalities. Computer Aided system (CAD) system might be the best methodology for early identification of breast abnormalities in digital mammogram. This research paper accomplishes the locality of abnormal tissue on digital mammographic images. Breast tissues might be a tiny bit, which can't be plainly particular acknowledgments to low contrast inside the advanced 2D-digital mammographic images. The most goal is to utilize image process approach to improve the advanced 2D- mammographic images and accordingly remove the selections from the suspicious area with best exactness and dependableness to care regardless of whether breast masses is affected or not. Indexterms- Neural Networks (NN), Thresholding, Digital mammogram, Distinction Stretching Improvement, Segmentation, Mass, Micro calcification, Ripple Analysis, micro Calcification.