A case study of image retrieval on lung cancer chest x-ray pictures.
This paper presents a case study of an image retrieval system based on a notion of similarity between images in a multimedia database and where a user request can be an image file or a keyword. The CBIR (Content Based Image Retrieval) system, the current System of Search for Information (SSI) --e.g. PEIR, MIRC, MIR, IRMA, and Pathopic-- and the Current Search Engines (CSE) --e.g. Google, Yahoo and Alta Vista-- make image search possible only when the query is a keyword. This type of search is limited because keywords are not expressive enough to describe all important characteristics of an image. For example, an exact match request cannot be formulated in such systems and in SSI system, users should know natural language (e.g. English, French or German) used. We used XIRS (an XML Image Retrieval System) to set up a similarity distance between images, then to compare the request image with those in a database. An experimentation of XIRS on lung cancer diagnosis is presented. The statistics show that our system is more efficient than leading CBIR systems such as ERIC7, PEIR, PathoPic and CSE.
Citation : Gile Narcisse Fanzou, T. et al. (2008) A case study of image retrieval on lung cancer chest x-ray pictures. 9th International Conference on Signal Processing, ICSP 2008. pp. 924-927.
ISBN : 9781424421794
Research Group : Software Technology Research Laboratory (STRL)
Research Institute : Cyber Technology Institute (CTI)
Peer Reviewed : Yes