Multicluster Class-Based Classification for the Diagnosis of Suspicious Areas in Digital Mammograms

摘要: This chapter presents a multicluster class-based classification approach for the of suspicious areas extracted from digital mammograms into benign and malignant classes. The creates multiple clusters selects strong each class. created are used to form subclasses within classes training classifier. creation during process can improve accuracy system. experiments using standard classifier with single cluster per class have been conducted on benchmark database mammograms. results shown that makes significant impact improving accuracy.