Data webs for locating research images stored in heterogeneous distributed repositories
Images play vital roles in academic research and teaching. While they may be easily interpreted by humans, they typically have no internal semantics that are easily extractable by present computing technologies. Descriptive metadata about the images are therefore essential to bridge this ‘semantic gap’. However, even with good metadata, finding images is not easy. Despite some commonality of the mechanisms of metadata exposure, for example using OIA-PMH protocols, data repositories exhibit varying degrees of syntactic and semantic incompatibility, making it impossible to find relevant images scattered across several repositories reliably, without searching each source individually.
As a solution to this problem, we propose the development of subject-specific data webs, involving lightweight software tools to harvest core metadata describing distributed research data into searchable registries that permit discovery and provide links back to the original data sources to allow data delivery. Within each data web, integration of the heterogeneous distributed resources is achieved by semantic mappings between their individual metadata schemas and a core ontology, permitting automated harvesting and indexing of core metadata.
Dr David Shotton
Dr David Shotton received bachelor, masters and doctoral degrees from the University of Cambridge during the late 1960s, and subsequently worked at Bristol University, Berkeley, Harvard, the Weizmann Institute of Science, and Imperial College London, before joining the Department of Zoology of the University of Oxford in 1981, where he is now Reader in Image Bioinformatics within the Mathematics, Physical and Life Sciences Division. He is a Fellow of Wolfson College, Oxford.
He teaches both elementary and advanced cell biology and informatics to Oxford undergraduates and graduates. Since its inception in 1995 he has been Degree Director for the University of Oxford M. Sc. in Biology Degree, and he also teaches for the M. Sc. in Bioinformatics and the EPSRC Life Sciences Interface Doctoral Training Centre. His past research has included molecular and cell biological investigations of protein and membrane structure in health and disease, investigations that have employed advanced light and electron microscopy techniques, particularly video and confocal light microscopy and freeze fracture electron microscopy. He has published extensively on these techniques, and has taught on many international microscopy courses.
He currently heads the Image Bioinformatics Research Group, part of the distributed Oxford e-Research Centre within the Department of Zoology, which is dedicated to research and the development of best practice for the sharing and reuse of biological research images. His present interests and R&D activities include the development of biological image databases, the development of data webs to enable integration of distributed information resources, the creation and community development of biological ontologies, the semantic content analysis of biological research videos, and the provision of related Semantic Web services to assist life sciences research. He is also interested in issues of semantic interoperability between the sciences, arts and humanities.