Linguraru, M: Detection of Early Features of Cancer in Mammo

Approximately one in nine women in the Western world will develop breast cancer, making it one of the most common and terrifying cancers. Screening programs reduce the mortality rate, but introduce an enormous amount of information that radiologists must process on a daily basis. Computer-aided diagnosis can assist clinicians sort through this data and help detect and classify cancer. This monograph introduces original results for the detection of early features of breast cancer from X-ray mammography analysis. Some of the earliest signs of breast cancer are clusters of microcalcifications, and a method for their detection based on a model of the human visual system is presented. The detection of breast masses in temporal mammographic pairs is also investigated. Image normalization and its benefits for cancer detection is another key concept discussed in this book, along with screening programs and other imaging modalities for breast cancer detection. The subject is of immense importance, not just to radiologists and clinicians, but to physicists, researchers, students and anyone who has been touched by this terrible disease.