Computer-aided detection system based on PCA/SVM for diagnosis of breast cancer lesions

Breast cancer is the most common female cancer and the leading cause of cancer-related deaths among women worldwide. Early detection of breast cancer increases treatment options and recovery rates. Mammography (MG) is the only broadly accepted and used screening test for early breast cancer detection. Several nationwide MG screening programs decreased breast cancer mortality in many developed countries. However, the difficulty of MG interpretation leads to high rates of missed cancers. Therefore, a second opinion is usually required to reduce erroneous rates. The computer-aided diagnosis (CAD) systems are commonly used to reduce the expense and to assist radiologists in MG interpretation. Typically, a CAD system is composed of three steps: preprocessing, feature extraction and classification where the extraction of a representative features is a key stage in enhancing the diagnosis performance. This talk will provide an overview of the fundamentals of CAD for diagnostic via detecting lesions such as microcalcifications and masses, providing discussion and comparison of existing CAD systems and illustrating their performance via experimental results.