A multi-modality neuroimaging research data informatics system
Filed under:
Neuroimaging
Gary Egan (Monash University), David Barnes (Monash University), Neil Killeen (University of Melbourne), Jason Lohrey (Arcitecta Pty Ltd), Wilson Liu (University of Melbourne), Wojtek Goscinksi (Monash University), Nicholas McPhee (Monash University)
Modern data-intensive, distributed and collaborative research depends critically on the ability to manage, move and interpret information. Monash Biomedical Imaging (MBI) operates five biomedical imaging scanners, and is linked to the Australian Synchrotron Imaging and Medical Beam Line. Our objective was to apply the Framework data [1] model and the Distributed and Reflective Informatics System (DaRIS) imaging data management tool to develop a multi-modality neuroinformatics system for use at MBI.
DaRIS compels scientific users to adopt a data model that has been designed specifically for biomedical imaging research. The Framework object model draws on and extends pre-existing object models in the public domain (e.g. DICOM and XCEDE). The Java and "web 2.0" interfaces to the DaRIS system are driven by the data model so that they are reflective of the data tree, rather than hand-coding an interface to match the data and meta-data. The system supports federation through the underlying Digital Asset Management system Mediaflux™ (www.arcitecta.com) and a citable identification scheme so that data and queries can be distributed over multiple nodes.
The MBI data management infrastructure is underpinned by a local “staging post” through which the acquired imaging data is routed. Imaging data tagged as belonging to a DaRIS project is automatically sent onwards (in DICOM format) to the server, and the data is ingested and attached to the appropriate research project. User-level access to data is then provided via the DaRIS web-based portal that provides browsing, viewing, download and transfer capabilities for biomedical imaging data. The MBI implementation of DaRIS is now operational for multi-modality biomedical imaging research applications, and has broader application to other fields of subject-centric research.
References:
[1] J. Lohrey, N. Killeen, G.F. Egan, “An Integrated Object Model and Method Framework for Subject-Centric e-Research Applications”, Frontiers in Neuroinformatics 3 (2009) 19, 1-10.
DaRIS compels scientific users to adopt a data model that has been designed specifically for biomedical imaging research. The Framework object model draws on and extends pre-existing object models in the public domain (e.g. DICOM and XCEDE). The Java and "web 2.0" interfaces to the DaRIS system are driven by the data model so that they are reflective of the data tree, rather than hand-coding an interface to match the data and meta-data. The system supports federation through the underlying Digital Asset Management system Mediaflux™ (www.arcitecta.com) and a citable identification scheme so that data and queries can be distributed over multiple nodes.
The MBI data management infrastructure is underpinned by a local “staging post” through which the acquired imaging data is routed. Imaging data tagged as belonging to a DaRIS project is automatically sent onwards (in DICOM format) to the server, and the data is ingested and attached to the appropriate research project. User-level access to data is then provided via the DaRIS web-based portal that provides browsing, viewing, download and transfer capabilities for biomedical imaging data. The MBI implementation of DaRIS is now operational for multi-modality biomedical imaging research applications, and has broader application to other fields of subject-centric research.
References:
[1] J. Lohrey, N. Killeen, G.F. Egan, “An Integrated Object Model and Method Framework for Subject-Centric e-Research Applications”, Frontiers in Neuroinformatics 3 (2009) 19, 1-10.
Preferred presentation format:
Poster
Topic:
Neuroimaging
Latest news for Neuroinformatics 2011
Follow INCF on Twitter
