Review of medical image retrieval systems and future directions

摘要: This paper presents a review of online systems for content-based medical image retrieval (CBIR). The objective this is to evaluate the capabilities and gaps in these determine ways improving relevance multi-modal (text image) information iMedline system, being developed at National Library Medicine (NLM). Seven systems: Figuresearch, BioText, GoldMiner, Yale Image Finder, Yottalook, Retrieval Medical Applications (IRMA), have been evaluated here using system defined [1]. Not all take advantage visual contained biomedical literature as figures illustrations. However, attempt extract metadata about from full-text articles retrieve figures/images response query. aims advance state-of-the-art multimodal by unifying text features computing relevance. We discuss shortcomings current future directions next steps toward context-based retrieval.