Abstract
Functional magnetic resonance imaging (fMRI) is a medical imaging technique used to characterize brain physiological activity, usually presented as 3D volumes in function of time. Our previous work in nonlinear association studies in electroencephalogram (EEG) time series enabled us to identify EEG features relevant for clinical diagnosis. The use of a similar approach in fMRI (adapted for 3D time series) is both appealing and new. Such time series analysis however imposes challenging requirements regarding computational power and medical image management. In this paper we propose the use of Grid computing to cope with the demanding fMRI multi-voxel association analysis workflow and present a working prototype. The system, implemented using the gLite middleware, provides the necessary support to manage brain images and run different non-linear fMRI analysis methods.

Abstract

Abstract
The present paper describes the Portuguese National Brain Imaging Network designed to join R&D efforts of four Portuguese universities (Aveiro, Coimbra, Minho and Porto) in this emergent scientific area. This is an open initiative, already funded in 81.3% of its predicted investment (similar to 4.3 million E) for the first 5 years of operation, opened to the participation of other national institutions. This area of neuroscience uses several types of datasets from different medical imaging modalities and biosignals. MRI/MRS and fMRI volumes along with high-resolution EEG are our main targets for the first 5 years of operation and can easily reach the GByte size for a patient study. The Brain Imaging Network Grid (BING) will provide the support to a "neuroscientist-friendly" web portal where neuroscientists can submit brain imaging datasets for different analysis protocols. We will focus the present paper on the description of the consortium, its objectives and the network and Grid services architecture designs that will provide both the computational resources and the federated large data repository for the Portuguese national wide neuroscience scientific community.

Abstract
In this paper we compare the classification accuracy of using compressed domain color (CDC) descriptors versus traditional full decoded images, for the purposes of topographic classification of wireless capsule endoscopy images. Results using a dataset of 26469 images, divided into stomach, small intestine and large intestine show a difference in classification accuracy below 1%. We also show that errors are mostly located near zone transitions (the pylorus and the ileocecal valve) and motivate the need for other visual descriptors (e. g. shape, motion) for addressing these specific areas. We conclude we can use the advantages of CDC in this type of classification with minor accuracy sacrifice.

Abstract
In this paper we associate features obtained from ECG signals with the expected levels of stress of real firefighters in action when facing specific events such as fires or car accidents. Five firefighters were monitored using wearable technology collecting ECG signals. Heart rate and heart rate variability features were analyzed in consecutive 5-min intervals during several types of events. A questionnaire was used to rank these types of events according to stress and fatigue and a measure of association was applied to compare this ranking to the ECG features. Results indicate associations between this ranking and both heart rate and heart rate variability features extracted in the time domain. Finally, an example of differences in inter personal responses to stressful events is shown and discussed, motivating future challenges within this research field.

Abstract
The endoscopic capsule is a recent medical technology with important clinical benefits but suffering from a practical handicap: long exam annotation times. This paper shows how support vector machines can be used to segment the gastrointestinal tract into its four major topographic areas, allowing the automatic estimation of the clinically relevant gastric and intestinal transit times. According to medical specialists, this can reduce exam annotation times by up to 12%.