Content based MRSI image indexing and retrieval involving a fuzzy descriptor

Abstract

Magnetic resonance spectroscopic imaging (MRSI) integrates both spectroscopic and imaging methods to produce spatially localized spectra from within the sample or patient. Although MRSI is a relatively new imaging technology for clinical applications and relevant databases still do not exist, the rapid advances made in the field of NMR and the associated scanning technologies, the increased frequency in which oncological diseases appear may soon lead to the creation of large databases of such images at least in main medical centers. Therefore, the need for indexing and retrieval of MRSI images might soon emerge. This paper proposes the application of a new method for content based MRSI medical image retrieval. The description of MRSI images relies on a new Descriptor which includes global image features, capturing both brightness and texture characteristics at the same time, based on an evaluation of MRS spectrum per voxel. Image information is extracted using a set of fuzzy approaches. Experiments demonstrate the feasibility of the proposed technique using synthetic images.