In a typical fMRI scanning session, the accumulated data can reach several hundreds of brain image volumes. The computational demand is usually very high that functional analyses are often delayed and activation maps are obtained offline, long after the scanning session is completed and the subject is no longer inside the MR scanner. In this work, the feasibility of using remote computational resources to enable real-time capability to a functional MR imaging site will be explored. The idea is to use the Globus toolkit for remote job submission so that the analysis can be performed in a remote computational server. Data transfer will be done volume-by-volume and immediately after a volume is acquired by the MR scanner. The results will be retrieved immediately to have a real-time update of the activation map.

Several tools for the analysis of functional MRI data will also be developed. One of these is independent component analysis (ICA), an fMRI data analysis method equally effective as the general linear model. These tools will be installed in several recognized remote analysis servers. An integration platform will be established so that the complexity of calling these tools remotely will be transparent to the users. Software development and maintenance will therefore be distributive. This means that one research group can contribute and maintain its own analysis tools within its own analysis servers which can be accessed by other research groups through the integration paltform.

For the system to be used widely, it must be accessible to a wide range of users, even by non-technical researchers. For this, a web-based interface, based on the concept of computational web services (CWS), will be developed. The idea is to make job submission as simple as uploading the data, submitting a job request form in the medgrid website, and downloading the results. CWS will offer a simplified interface for the user with a fast and powerful computational backend.