You are here
A visual concept-based interactive biomedical image retrieval using entropy and spatial information.
This paper presents a novel approach to biomedical image retrieval by mapping image regions to local concepts and represent images in a weighted entropy-based concept feature space. The term concept refers to perceptually distinguishable visual patches that are identiﬁed locally in image regions and can be mapped to a glossary of imaging terms. Further, the visual signiﬁcance (e.g., visualness) of concepts is measured as Shannon entropy of pixel values in image patches and is used to reﬁne the feature vector. Moreover, the system can assist user in interactively select a Region-Of-Interest (ROI) and search for similar image ROIs. Further, a spatial veriﬁcation step is used as a post-processing step to improve retrieval results based on location information. The hypothesis that such approaches would improve biomedical image retrieval, is validated through experiments on a data set of 450 lung CT images extracted from journal articles from four diﬀerent collections. © (2015) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.