Extensible Architecture for Multimodal Information Retrieval in Medical Imaging Archives

Abstract

The challenge of medical information retrieval has attracted significant attention since the introduction of digital imaging in hospitals and other health facilities. Given the huge growth of medical imaging data produced in the past few years, studying new ways to index, process and retrieve medical images becomes an important subject to be addressed by the wider community of radiologists, scientists and engineers. At the moment, content-based image retrieval in medical imaging archives, although known to be beneficial to practitioners and researchers, is still rarely integrated into these archives. The aim of this paper is to present an overview of multimodal information retrieval for medical imaging studies, as well as the architecture of a solution for automatic medical image classification and retrieval using combinations of text and image queries. The complete solution was designed and implemented over an extensible open-source medical imaging archive software.