High-aspect-ratio, molded microstructures with electrical isolation and embedded interconnects

 A thin film molding process was developed to enable the fabrication of monolithic micromechanical structures with built-in electrical isolation and embedded interconnects. High-aspect-ratio composite structures were created from undoped polysilicon, low stress nitride and doped polysilicon, in a dual micromolding process. These monolithic electro-mechanical microstructures are more resistant to thermal effects and misalignment errors compared to microsystems assembled from discrete elements. In addition, the microstructures are molded in a re-usable mold providing an economical advantage. A gimballed electrostatic microactuator was successfully fabricated using this process. Electrical isolation was achieved with a combination of low stress nitride and undoped polycrystalline silicon. Various isolation geometries were investigated. Current leakages of less than 1 nA at 30 V were measured for isolation structures 40 μm long and 80 μm tall. Received: 13 November 2000/Accepted: 16 November 2000     

The Shopping Jacket: Wearable computing for the consumer

As part of the Bristol Wearable Computing Initiative, we are exploring location-sensing systems suitable for use with wearable computing. In this paper we present our findings, and in particular a wearable application — the ‘Shopping Jacket’ — which relies on a minimal infrastructure to be effective. We use two positioning devices, ‘pingers’ and GPS. The pinger is used to signal the presence of a shop, and to indicate the type of shop and its website. The GPS is used to disambiguate which branch of a high street chain is being passed. The wearable uses this information to determine whether the wearer needs to be alerted that they are passing an interesting shop, or to direct the wearer around a shopping mall. The Shopping Jacket integrates a wearable CardPC, GPS and pinger receivers, a near-field radio link, hand-held display, GSM data telephone and a speech interface into a conventional sports blazer.     

Progressively weighted affine adaptive correlation matching for quasi-dense 3D reconstruction

Correlation matching has been widely accepted as a rudimentary similarity measure to obtain dense 3D reconstruction from a stereo pair. In particular, given a large overlapping area between images with minimal scale differences, the correlation results followed by a geometrically constrained global optimisation delivers adequately dense and accurate reconstruction results. In order to achieve greater reliability, however, correlation matching should correctly account for the geometrical distortion introduced by the different viewing angles of the stereo or multi-view sensors. Conventional adaptive least squares correlation (ALSC) matching addresses this by modifying the shape of a matching window iteratively, assuming that the distortion can be approximated by an affine transform. Nevertheless, since an image captured from different viewing angle is often not practically identical due to scene occlusions, the matching confidence normally deteriorates. Subsequently, it affects the density of the reconstruction results from ALSC-based stereo region growing algorithms. To address this, we propose an advanced ALSC matching method that can progressively update matching weight for each pixel in an aggregating window using a relaxation labelling technique. The experimental results show that the proposed method can improve matching performance, which consequently enhances the quality of stereo reconstruction. Also, the results demonstrate its ability to refine a scale invariant conjugate point pair to an affine and scale invariant point pair.     

Support and precursor effects on the preparation of new heterogenized Pt/Sn catalysts for the selective hydroformylation of 1-pentene

New heterogenized Pt/Sn catalysts selective for the hydroformylation of 1-pentene have been synthesized. The complex cis-[PtCl₂(PPh₃)₂] and the SnCl₂.2H₂O or SnC₂O₄ precursors have been anchored on silica-, magnesia- and alumina-carriers. X-ray photoelectron spectroscopy was used to determine the surface composition and the nature of the anchored species. The hydroformylation activity was found to depend on the type of support and tin precursor used. Only the silica supported catalysts were active in the hydroformylation reaction. Samples prepared from SnCl₂-2H₂O were 200-fold more active than those prepared from SnC₂O₄. Selectivity ton-hexanal of the silica-supported catalyst prepared from SnCl₂-2H₂O was as high as 94.4% at 39.2% conversion of 1-pentene.     

Maternally modulated infant separation responses are regulated by D2-family dopamine receptors.

Although dopamine is necessary for mammalian adult pair-bond formation and maternal behavior, its function in infant social behavior and attachment has been less thoroughly explored. The vocalization rate of an isolated rat pup is influenced by recent social contact. Interactions with the dam potentiate vocalization rate. Interactions with littermates or adult males do not. Systemic administration of the D2-family agonist quinpirole specifically blocked maternal potentiation at doses that did not alter vocalization rate in an isolation prior to dam contact. This result was not explained by quinpirole's effects on body temperature or locomotion. The results are consistent with a role for dopamine in infant social behavior. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Impairment in long-term retention but not short-term performance on a water maze reversal task following hippocampal or mediodorsal striatal n-methyl-d-aspartate receptor blockade.

Male Long-Evans rats were injected with 32 ng/μl of the N-methyl-D-aspartate (NMDA) receptor antagonist 3-(2-carboxypiperazin-4-yl) propyl-1-phosphonic acid (CPP) or vehicle and trained to locate a hidden platform in a different location (reversal training) than used on the initial 4 days of training. Rats treated with vehicle or CPP into the dorsal hippocampus, basolateral amygdala, or mediodorsal striatum had similar latencies to locate the platform on the reversal day. Rats infused with CPP into the dorsal hippocampus or mediodorsal striatum failed to search preferentially in the novel location during a 24-hr, drug-free retention test, whereas all other groups searched preferentially in this location. Therefore, blocking dorsal hippocampal or mediodorsal striatal NMDA receptors selectively blocked long-term spatial retention without producing short-term performance deficits. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Constructing descriptive and discriminative nonlinear features: Rayleigh coefficients in kernel feature spaces

We incorporate prior knowledge to construct nonlinear algorithms for invariant feature extraction and discrimination. Employing a unified framework in terms of a nonlinearized variant of the Rayleigh coefficient, we propose nonlinear generalizations of Fisher's discriminant and oriented PCA using support vector kernel functions. Extensive simulations show the utility of our approach.     

Deformable M-Reps for 3D Medical Image Segmentation

M-reps (formerly called DSLs) are a multiscale medial means for modeling and rendering 3D solid geometry. They are particularly well suited to model anatomic objects and in particular to capture prior geometric information effectively in deformable models segmentation approaches. The representation is based on figural models, which define objects at coarse scale by a hierarchy of figures—each figure generally a slab representing a solid region and its boundary simultaneously. This paper focuses on the use of single figure models to segment objects of relatively simple structure.A single figure is a sheet of medial atoms, which is interpolated from the model formed by a net, i.e., a mesh or chain, of medial atoms (hence the name m-reps), each atom modeling a solid region via not only a position and a width but also a local figural frame giving figural directions and an object angle between opposing, corresponding positions on the boundary implied by the m-rep. The special capability of an m-rep is to provide spatial and orientational correspondence between an object in two different states of deformation. This ability is central to effective measurement of both geometric typicality and geometry to image match, the two terms of the objective function optimized in segmentation by deformable models. The other ability of m-reps central to effective segmentation is their ability to support segmentation at multiple levels of scale, with successively finer precision. Objects modeled by single figures are segmented first by a similarity transform augmented by object elongation, then by adjustment of each medial atom, and finally by displacing a dense sampling of the m-rep implied boundary. While these models and approaches also exist in 2D, we focus on 3D objects.The segmentation of the kidney from CT and the hippocampus from MRI serve as the major examples in this paper. The accuracy of segmentation as compared to manual, slice-by-slice segmentation is reported.     

Asymptotic statistical theory of overtraining and cross-validation

A statistical theory for overtraining is proposed. The analysis treats general realizable stochastic neural networks, trained with Kullback-Leibler divergence in the asymptotic case of a large number of training examples. It is shown that the asymptotic gain in the generalization error is small if we perform early stopping, even if we have access to the optimal stopping time. Based on the cross-validation stopping we consider the ratio the examples should be divided into training and cross-validation sets in order to obtain the optimum performance. Although cross-validated early stopping is useless in the asymptotic region, it surely decreases the generalization error in the nonasymptotic region. Our large scale simulations done on a CM5 are in good agreement with our analytical findings.