Abstract
A versatile time-domain OCT system is presented which, for the first time, can generate cross-section images (B-scans) by using either transverse priority (T-scans) or depth priority (A-scans). Images from the optic nerve are obtained with either regime, with the same system. In different scanning regimes, different values are allowed for the maximum power to be launched to the eye. We present the maximum exposure level for a variety of scanning procedures, such as generation of cross section images and 3D volumes employing either A or T scanning.

Abstract
This paper develops measures, based on the Malmquist index, that enable the decision making units' internal inefficiencies to be distinguished from those associated with their group (or program) characteristics. The applicability of these measures is illustrated with the assessment of bank branches' performance. The analysis involves the construction of an index reflecting the relative performance of branches in four different regions, which can be decomposed into an index for the comparison of within-group efficiency spread, evaluating internal managerial efficiencies, and an index for the comparison of frontier productivity, reflecting the impact of environmental factors and regional managerial policies on branches' productivity.

Abstract
Many software systems can be viewed as organizational Systems, where the different components are seen as autonomous entities, interacting with each other, collaborating toward system's aims. In such systems we may not have full control over the behavior of all its components. Normative specification of an organizational system, provides a way of describing the norms that regulate the behavior of a system and of its components, stating how they are expected to behave, assuming however, that they may deviate from that ideal behavior. In this paper we use an action and deontic modal logic for the normative specification of organizational systems. This logical framework allows us to describe expected behavior of agents, detect non-ideal behavior and identify the agents that, direct or indirectly, are responsible for it. We argue that normative specification can be an useful tool to increase trust and security in complex computational systems and propose a responsibility-based trust concept.

Abstract

Abstract
The Support Vector Machine (SVM) algorithm is sensitive to the choice of parameter settings, which makes it hard to use by non-experts. It has been shown that meta-learning can be used to support the selection of SVM parameter values. Previous approaches have used general statistical measures as meta-features. Here we propose a new set of meta-features that are based on the kernel matrix. We test them on the problem of setting the width of the Gaussian kernel for regression problems. We obtain significant improvements in comparison to earlier meta-learning results. We expect that with better support in the selection of parameter values, SVM becomes accessible to a wider range of users. Copyright 2006 ACM.

Abstract
A refractometric sensor based on a phase-shifted long-period fiber grating written by electric-arc discharges is presented. Transmission and reflective configurations for refractive index measurements are studied. It is observed that the reflective topology permits better performance compared with the transmission one, which is the approach normally utilized in the context of long-period fiber sensing. The resolution achieved in the measurement of refractive index enables the application of this sensing head structure in demanding situations, such as the measurement of the level of salinity of water. (c) 2006 Optical Society of America.