Technique for Producing Mullite and Other Mixed-Oxide Systems

Mixed-oxide systems are of interest in a variety of technical ceramic applications. The key in many systems is to achieve a high degree of homogeneity, approaching the molecular level. A technique for producing mixed oxides via direct oxidation of mixtures of organometallic compounds is discussed as a potential route. Preliminary work performed on producing mullite is discussed.     

Cognitive categorization and quality of performance ratings.

The effects of cognitive categorization of raters on accuracy, leniency, and halo of performance evaluations were investigated in a field setting. One hundered seventy-four subordinates evaluated the performance of their managers on three performance dimensions. Managers were categorized as congruent or incongruent based on subordinates' perceptions of the extent to which the manager's behavior met the subordinates' expectations. The results indicated that the quality of ratings assigned by subordinates was related to the cognitive categories used. As hypothesized, ratings of managers who were categorized as congruent were found to be more accurate and also to contain more leniency and halo tendency than the ratings of managers who were categorized as incongruent. Implications of these findings for performance-appraisal research are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Photodegradation of ethylene–propylene copolymer and ethylene–propylene–ethylidenenorbornene terpolymer

Significant progress has been made in recent years regarding the photooxidation of olefin copolymers, but questions still remain. This paper reviews the progress and probes the photooxidative chemistry of ethylene–propylene (EP) and ethylene/propylene/diene monomer (EPDM) copolymers. Both stabilized and unstabilized polymer plaques were irradiated in a xenon are and the surface chemistry followed using infra-red spectroscopy. Model compounds were used to help elucidate the chemistry caused by unique structural features present in the copolymers. Volatile products evolved during photooxidation were determined giving valuable insight into the degradation chemistry.     

Inductive transfer with context-sensitive neural networks

Context-sensitive Multiple Task Learning, or csMTL, is presented as a method of inductive transfer which uses a single output neural network and additional contextual inputs for learning multiple tasks. Motivated by problems with the application of MTL networks to machine lifelong learning systems, csMTL encoding of multiple task examples was developed and found to improve predictive performance. As evidence, the csMTL method is tested on seven task domains and shown to produce hypotheses for primary tasks that are often better than standard MTL hypotheses when learning in the presence of related and unrelated tasks. We argue that the reason for this performance improvement is a reduction in the number of effective free parameters in the csMTL network brought about by the shared output node and weight update constraints due to the context inputs. An examination of IDT and SVM models developed from csMTL encoded data provides initial evidence that this improvement is not shared across all machine learning models.     

Computing linear approximations to nonlinear neuronal response

We present an approach to obtain nonlinear information about neuronal response by computing multiple linear approximations. By calculating local linear approximations centered around particular stimuli, one can obtain insight into stimulus features that drive the response of highly nonlinear neurons, such as neurons highly selective to a small set of stimuli. We implement this approach based on stimulus-spike correlation (i.e., reverse correlation or spike-triggered average) methods. We illustrate the benefits of these linear approximations with a simplified two-dimensional model and a model of an auditory neuron that is highly selective to particular features of a song.     

Managing empowerment and control in an intranet environment

Abstract. An intranet increases in sophistication and complexity as it evolves. This evolution leads to an increasing need for control over the intranet. However, this is a contentious issue, as an intranet is deemed to be an empowering technology. Consequently, intranet control systems must balance empowerment and control so as not to negate each other. This paper investigates intranet control activities and their effect on users' perceptions of empowerment throughout the evolution of an intranet in Hewlett Packard (Ireland). The growth of the intranet is charted as a six-stage model that illustrates an evolution of purpose, control and empowerment. The control strategies for managing the intranet implemented at each stage are investigated, and their resultant effects on empowerment are evaluated. The study reveals the importance of balancing control strategies with empowerment initiatives in managing intranet environments. Based on the evidence available, the study recommends the implementation of specific controls at particular stages in the evolution of an intranet in order to achieve control systems that balance empowerment and control.     

Multi-agent segmentation of IVUS images

A novel multi-agent image interpretation system has been developed which is markedly different from previous approaches in especially its elaborate high-level knowledge-based control over low-level image segmentation algorithms. Agents dynamically adapt segmentation algorithms based on knowledge about global constraints, contextual knowledge, local image information and personal beliefs. Generally agent control allows the underlying segmentation algorithms to be simpler and be applied to a wider range of problems with a higher reliability.The agent knowledge model is general and modular to support easy construction and addition of agents to any image processing task. Each agent in the system is further responsible for one type of high-level object and cooperates with other agents to come to a consistent overall image interpretation. Cooperation involves communicating hypotheses and resolving conflicts between the interpretations of individual agents.The system has been applied to IntraVascular UltraSound (IVUS) images which are segmented by five agents, specialized in lumen, vessel, calcified-plaque, shadow and sidebranch detection. IVUS image sequences from 7 patients were processed and vessel and lumen contours were detected fully automatically. These were compared with expert-corrected semiautomatically detected contours. Results show good correlations between agents and expert with r=0.84 for the lumen and r=0.92 for the vessel cross-sectional areas, respectively.     

Automated contour detection in X-ray left ventricular angiograms using multiview active appearance models and dynamic programming

This paper describes a new approach to the automated segmentation of X-ray left ventricular (LV) angiograms, based on active appearance models (AAMs) and dynamic programming. A coupling of shape and texture information between the end-diastolic (ED) and end-systolic (ES) frame was achieved by constructing a multiview AAM. Over-constraining of the model was compensated for by employing dynamic programming, integrating both intensity and motion features in the cost function. Two applications are compared: a semi-automatic method with manual model initialization, and a fully automatic algorithm. The first proved to be highly robust and accurate, demonstrating high clinical relevance. Based on experiments involving 70 patient data sets, the algorithm's success rate was 100% for ED and 99% for ES, with average unsigned border positioning errors of 0.68 mm for ED and 1.45 mm for ES. Calculated volumes were accurate and unbiased. The fully automatic algorithm, with intrinsically less user interaction was less robust, but showed a high potential, mostly due to a controlled gradient descent in updating the model parameters. The success rate of the fully automatic method was 91% for ED and 83% for ES, with average unsigned border positioning errors of 0.79 mm for ED and 1.55 mm for ES.     

Design and analysis of a 512×768 current-mediated activepixel array image sensor

Current-mediated, current-output active pixels offer the advantages of compact size and simple operation in designing large format CMOS image sensors, with performance limited by spatial fixed pattern noise. In this paper, a thorough noise analysis is made of the expected performance of a current-output pixel image sensor. This analysis is compared with experimental results from a 512×768 array imager fabricated in a 0.7 μm process, and the effectiveness of basic error correction techniques are explored. The goal of this study was to determine the performance limits of this device and to gain insight into the design issues needed to develop a high-quality current-output imager