Virtual EZ Grid - Applications

Two medical applications are gridified and deployed on the Virtual EZ-Grid platform: MedGift and NeuroWeb.

MedGift

MedGIFT is a project for analysing medical images using the GIFT (GnuImage Finding Tool) software. GIFT recognises similarities in images based on "local" and "global" features like colours and edges. Extracting these features from a large set of images is a time consuming task, but fortunately it can be parallelised quite easily,since different subtasks do not need to communicate with each other.

ImageCLEFmed is a collection of medical images, created for testing the performance and accuracy of image processing. The current collection consists of about 50 000 images. Indexing these on a single-cpu computer takes typically several days. By parallelising the indexing process it can be done in a few hours.

Figure: An interface for a medical image analysis and retrieval application

NeuroWeb

NeuroWeb is a medical application used to generate temporal dynamic neuronal maps (Figure 1). One of its objectives is to better understand the connectivity and activity of neurons. The input data are obtained from a MEG scanner (Figure 2) to establish the cartography of the electric neuronal state of the cortex.

These dynamic neuronal maps are complex to create and require intensive computations. One of the objectives of Virtual EZ Grid is to gridify the NeuroWeb application and deploy it on the XtremWeb-CH desktop Grid platfom.

Figure 1: A dynamic neuronal map	Figure 2: Magnetoencephalography (MEG)