Adaptive Gaze Control for Object Detection.

We propose a novel gaze-control model for detecting objects in images. The model, named act-detect, uses the information from local image samples in order to shift its gaze towards object locations. The model constitutes two main contributions. The first contribution is that the model's setup makes it computationally highly efficient in comparison with existing window-sliding methods for object detection, while retaining an acceptable detection performance. act-detect is evaluated on a face-detection task using a publicly available image set. In terms of detection performance, act-detect slightly outperforms the window-sliding methods that have been applied to the face-detection task. In terms of computational efficiency, act-detect clearly outperforms the window-sliding methods: it requires in the order of hundreds fewer samples for detection. The second contribution of the model lies in its more extensive use of local samples than previous models: instead of merely using them for verifying object presence at the gaze location, the model uses them to determine a direction and distance to the object of interest. The simultaneous adaptation of both the model's visual features and its gaze-control strategy leads to the discovery of features and strategies for exploiting the local context of objects. For example, the model uses the spatial relations between the bodies of the persons in the images and their faces. The resulting gaze control is a temporal process, in which the object's context is exploited at different scales and at different image locations relative to the object.

From clinical practice guidelines to computer-interpretable guidelines. A literature overview.

BACKGROUND: Guidelines are among us for over 30 years. Initially they were used as algorithmic protocols by nurses and other ancillary personnel. Many physicians regarded the use of guidelines as cookbook medicine. However, quality and patient safety issues have changed the attitude towards guidelines. Implementing formalized guidelines in a decision support system with an interface to an electronic patient record (EPR) makes the application of guidelines more personal and therefore acceptable at the moment of care. OBJECTIVE: To obtain, via a literature review, an insight into factors that influence the design and implementation of guidelines. METHODS: An extensive search of the scientific literature in PubMed was carried out with a focus on guideline characteristics, guideline development and implementation, and guideline dissemination. RESULTS: We present studies that enable us to explain the characteristics of high-quality guidelines, and new advanced methods for guideline formalization, computerization, and implementation. We show how the guidelines affect processes of care and the patient outcome. We discuss the reasons of low guideline adherence as presented in the literature and comment upon them. CONCLUSIONS: Developing high-quality guidelines requires a skilled team of people and sufficient budget. The guidelines should give personalized advice. Computer-interpretable guidelines (CIGs) that have access to the patient's EPR are able to give personal advice. Because of the costs, sharing of CIGs is a critical requirement for guideline development, dissemination, and implementation. Until now this is hardly possible, because of the many models in use. However, some solutions have been proposed. For instance, a standardized terminology should be imposed so that the terms in guidelines can be matched with terms in an EPR. Also, a dissemination model for easy updating of guidelines should be established. The recommendations should be based on evidence instead of on consensus. To test the quality of the guideline, appraisal instruments should be used to assess the guideline as a whole, as well as checking the quality of the recommendations individually. Only in this way optimal guideline advice can be given on an individual basis at a reasonable cost.

Information retrieval: an overview of system characteristics.

The paper gives an overview of characteristics of information retrieval (IR) systems. The characteristics are identified from the descriptions of 23 IR systems. Four IR models are discussed: the Boolean model, the vector model, the probabilistic model and the connectionistic model. Twelve other characteristics of IR models are identified: search intermediary, domain knowledge, relevance feedback, natural language interface, graphical query language, conceptual queries, full-text IR, field searching, fuzzy queries, hypertext integration, machine learning, and ranked output. Finally, the relevance of IR systems for the World Wide Web is established.

Robust feedforward processing in synfire chains.

The rapidity of time-constrained visual identification suggests a feedforward process in which neural activity is propagated through a number of cortical stages. The process is modeled by using a synfire chain, leading to a neural-network model which involves propagating activation waves through a sequence of layers. Theory and analysis of the model's behavior, especially in the presence of noise, predict enhancement of wave propagation for a range of noise intensities. Simulation studies confirm this prediction. The results are discussed in terms of (spatio-temporal) stochastic resonance. It is concluded that feedforward processes such as time-constrained visual identification may benefit from moderate levels of noise.