Structural Informatics Group (SIG) logo
Home | Projects | Demos | Downloads | Publications | Local Info | About Us
Publications

Lightweight XML-based query, integration and visualization of distributed, multimodality brain imaging data

Brinkley, James F and Poliakov, Andrew V and Moore, Eider B and Detwiler, Landon T and Franklin, Joshua D and Corina, David P and Ojemann, George A (2006) Lightweight XML-based query, integration and visualization of distributed, multimodality brain imaging data . In Proceedings Society for Neuroscience Annual Meeting, pages 100.13, Atlanta.

Full text available as:
PDF

Abstract

A need of many neuroimaging researchers is to integrate multimodality brain data that may be stored in separate databases. To address this need we have developed a framework that provides a uniform XML-based query interface across multiple online data sources. The development of this framework is driven by the need to integrate neurosurgical and neuroimaging data related to language. The data sources for the language studies are 1) a web-accessible relational database of neurosurgical cortical stimulation mapping data (CSM) that includes patient-specific 3-D coordinates of each stimulation site mapped to an MRI reconstruction of the patient brain surface; and 2) an XML database of fMRI and structural MRI data and analysis results, created automatically by a batch program we have embedded in SPM. To make these sources available for querying each is wrapped as an XML view embedded in a web service. A top level web application accepts distributed XQueries over the sources, which are dispatched to the underlying web services. Returned results can be displayed as XML, HTML, CSV (Excel format), a 2-D schematic of a parcellated brain, or a 3-D brain visualization. In the latter case the CSM patient-specific coordinates returned by the query are sent to a transformation web-service for conversion to normalized space, after which they are sent to our 3-D visualization program MindSeer, which is accessed via Java WebStart through a generated link. The anatomical distribution of pooled CSM sites can then be visualized using various surfaces derived from brain atlases. As this framework is further developed and generalized we believe it will have appeal for researchers who wish to query, integrate and visualize results across their own databases as well as those of collaborators.

Item Type:In Collection
Projects:Brain Project
Subjects:SIG Publications
ID Code:199
Deposited By:Brinkley, James F.
Deposited On:20 January 2007