Certain investigations on application Of multiresolution analysis and Advanced neural networks on Mammograms for early detection of Breast cancer

Abstract

Breast cancer which continues to be a significant public health problem newlinearound the world is the most prevalent cancer among women Breast cancer is most newlineeffectively treated when detected at an early stage and the survival probability of newlinethe patient is dependent on the stage at which it is diagnosed Digital X ray newlinemammographic method is a specialized radiographic imaging technique for newlinediagnosis of breast diseases It identifies the morphological differences that indicate newlinethe presence of breast cancer such as masses microcalcifications, and architectural newlinedistortions Detection of breast cancer at an early stage requires mammographic newlineimages which have high sensitivity and specificity with a relatively low radiation newlinedose This imposes challenging requirements for interactive and intelligent medical newlineimage analysis Computerized medical image analysis method can provide effective newlinetools to help differential diagnosis intervention and treatment monitoring newlineIn the literature various Computer Aided Diagnostic CAD systems newlineare described to detect the presence of breast cancer and to classify them as newlinebenign or malignant A detailed review of existing methods is presented in order to newlineprovide an insight about the state of the art The objective of this research is to newlinedesign advanced image processing techniques and algorithms that can aid breast newlinecancer detection at an early stage newlineMammography remains the most effective diagnostic technique for newlineearly breast cancer detection however not all types of breast cancer can be newlinedetected by mammograms newline newline