The FUSION project (Fuel cell Understanding through Semantic Inferencing, Ontologies and Nanotechnology) is a joint project between DSTC, the University of Queensland's Centre for Microscopy and Microanalysis (CMM) and Mathematics Department. It aims to apply, extend and combine Semantic Web technologies and image analysis techniques to develop a knowledge management system for optimizing the design of fuel cells. This problem domain produces large amounts of complex, varied, multi-level, mixed-media data and requires extensive tool support to analyse, extract and integrate the information necessary to improve the performance, efficiency and understanding of fuel cells.

Series of example microscopy images showing a fuel cell at different levels of magnification. Courtesy of CMM.

The main functions of the FUSION project are to enable:

1. the capture of quality data from all stages of the fuel cell design, manufacture, performance testing, microscopy and analysis phases;
2. the integration, via web services, of mathematical and scientific toolkits such as matlab, R and gnuplot;
3. the interactive development of semantic inferencing rules to semi-automatically label regions in microscopy images with high-level, domain-specific terms based on low-level image features;
4. the storage, annotation, application and processing of rules and mathematical models;
5. interfaces for hypothesis generation and testing using all available data and metadata;
6. tools to support experimental design.

The project is applying and extending emerging technologies and W3C standards such as:

• eXtensible Markup Language (XML), XML Schema, Resource Description Framework (RDF);
• Web Ontology Language (OWL);
• Web Services languages - SOAP, WSDL, BPEL4WS;
• Rule Languages for the Semantic Web - RuleML, SWRL;
• Mathematics Markup Language (MathML);
• Display formats such as JPEG2000, Synchronised Multimedia Integration Language (SMIL), Standard Vector Graphics (SVG);

While the current area of application is a materials science domain, the core concepts and tools of the FUSION project apply equally well to many eScience, bioinformatics and biomedical problem areas.

A number of prototypes, ontologies and metadata schemas have been produced during the course of this project. A list of those outcomes that are available online can be found here. For more detailed information see the publications list or contact Suzanne Little or Jane Hunter.