Neuro-Sharing is a one day satellite workshop of AIME-2009 devoted to the sharing of heterogeneous and distributed data and tools in the context of neuroimaging. It will be held in Verona on July 19th, in conjunction with AIME 09.
Neuroimaging gathers advanced imaging techniques to study in vivo brain anatomy and function.
It has a growing major impact on our understanding of complex brain processes under normal and pathological conditions.
Research in this domain is data intensive: data sets are large (many Gb per subject), heterogeneous, coming from various
imaging techniques with different spatial and temporal resolutions, and require, to be correctly exploited,
additional information about their provenance. Researchers should integrate these large rich sources of
brain data coming from disseminated centres to strengthen collaborative multi-centre studies,
perform meta-analysis and knowledge discovery via data mining techniques.
Moreover, to fully exploit neuroimaging data, many processing tools are currently developed in research laboratories.
Tool sharing will facilitate their dissemination, their evaluation face to various datasets and the construction of
robust heterogeneous data processing pipelines.
Recent advances have allowed overcoming many of the initial obstacles toward these objectives. Medical images are now in digital format and the DICOM standard facilitates the transfer of images and associated meta information. Grid architectures are now becoming available to healthcare communities enabling the management of large data sets. Specific recent projects have been launched in the US (BIRN, NA-MIC, ICBM, LONI) or Europe (NeuroLOG, IXI). Finally, suitable syntaxes (e.g. XML) and high level languages (e.g. OWL) have been designed to represent structured information such as metadata, related knowledge and data processing schemes.
However, many challenges remain for effective neuroimaging data and tools sharing and integration across multiple sources, such as ontology development for modelling data and methods in neuroscience, semantic interoperability for large-scale data integration, and management of data sharing policies and security issues.
In this perspective, the objective of the workshop is to gather researchers from different communities, especially from A.I., image processing, database, knowledge engineering and grid computing, interested to discuss the topics mentioned below in the context of Medical Image Computing.