Content-based retrieval and data mining of a skin cancer image database

Skin cancer is the most common type of cancer in the USA. A large, shared skin cancer image database on the Internet would be quite valuable to both medical professionals and consumers. In this paper, such a database is created using a three-tier system: a client application implemented in Java applets, a Web server and a back-end database server. JDBC-ODBC is used for the Web server to communicate with the database server. Various browsing and content-based retrieval methods are supported for the skin cancer image database through Web-based graphical user interfaces. A data mining algorithm for finding association rules between different features of the skin cancer images is also implemented.