INTEGRATING CONTENT-BASED RETRIEVAL IN A MEDICAL IMAGE REFERENCE DATABASE

Image reference databases (RDBs) are a recurrent research topic in med ical imaging. Most IRDBs are designed to help experienced physicians i n diagnostic tasks and require that users have prior extensive knowled ge of the field for their use to be fruitful. Therefore, the education al potential of such image collections cannot be exploited thoroughly. In this paper we propose an image-indexing method to extend the funct ionalities of an existing medical IRDB and allow for ifs use in educat ional applications, as well as in computer-assisted diagnosis. Our met hod, based on the Kahrunen-Loeve transform, has been used to develop a content-based search engine for tomographic image databases on which we are presently experimenting and which we aim to integrate into a wo rking radiological IRDB installed at the University of Florence. Resul ts achieved in our preliminary tests are also reported. Copyright (C) 1996 Elsevier Science Ltd.