The cranial MRI in severe cerebral palsy: a comparative study with clinical data

The magnetic resonance examination was performed in 38 patients with severe cerebral palsy (CP; 15 males and 23 females) who had both motor delay (unable to move anywhere) and mental retardation (I. Q or D. Q below 30). Neuroimaging findings were compared with the CP type, etiology, and grade of understanding of language. Cranial magnetic resonance imagings (MRI) in CP were divided into five types. Type 1 : nine predominantly showed cyst-liked ventricles and periventricular hyperintensity on T2-weighted imaging (PVH) and only scarred basal ganglia and thalamus were visible. All suffered from neonatal asphyxia and the clinical type was rigospastic tetraplegia (RST). Type 2: eleven predominantly showed PVH and hyperintensity on T2-weighted (HT2) in basal ganglia and thalamus. All suffered from neonatal asphyxia and the clinical type was RST or rigospastic diplegia. Type 3: five showed PVH and three had cortical atrophy. All suffered from neonatal asphyxia and the clinical type was spastic diplegia. Type 4: four predominantly showed HT 2 in putamen and thalamus. Three had cortical atrophy. All suffered from neonatal asphyxia. The clinical type was athetotic CP (ATH). Type 5: nine predominantly showed HT 2 in globus pallidus. Four had cortical atrophy and two had hippocampal atrophy. All suffered from neonatal jaundice and the clinical type was ATH. All patients who suffered from neonatal asphyxia and spastic CP had MRI in PVH. All patients who suffered from neonatal asphyxia and ATH showed HT 2 in putamen and thalamus. Almost patients who suffered neonatal jaundice and ATH showed HT 2 in globus pallidus. With athetotic CP, cases with atrophy of the cerebral cortex or/and hippocampus were lower grade of understanding of language than no atrophy of both. The result of studies of MRI are in agreement with neuropathological findings.     

A procedure to analyze surface profiles of the protein molecules visualized by quick-freeze deep-etch replica electron microscopy

Quick-freeze deep-etch replica electron microscopy gives high contrast snapshots of individual protein molecules under physiological conditions in vitro or in situ. The images show delicate internal pattern, possibly reflecting the rotary-shadowed surface profile of the molecule. As a step to build the new system for the "Structural analysis of single molecules", we propose a procedure to quantitatively characterize the structural property of individual molecules; e.g. conformational type and precise view-angle of the molecules, if the crystallographic structure of the target molecule is available. This paper presents a framework to determine the observed face of the protein molecule by analyzing the surface profile of individual molecules visualized in freeze-replica specimens. A comprehensive set of rotary-shadowed views of the protein molecule was artificially generated from the available atomic coordinates using light-rendering software. Exploiting new mathematical morphology-based image filter, characteristic features were extracted from each image and stored as template. Similar features were extracted from the true replica image and the most likely projection angle and the conformation of the observed particle were determined by quantitative comparison with a set of archived images. The performance and the robustness of the procedure were examined with myosin head structure in defined configuration for actual application.     

A new learning method using prior information of neural networks

In this paper, we present a new learning method using prior information for three-layer neural networks. Usually when neural networks are used for identification of systems, all of their weights are trained independently, without considering interrelated weights values. Thus, the training results are usually not good. The reason for this in that each parameter has its influence on others during learning. To overcome this problem, we first give an exact mathematical equation that describes the relation between weight values given a set of data conveying prior information. The we present a new learning method that trains part of the weights and calculates the others using these exact mathematical equations. This method often a priori keeps the given mathematical structure exactly the same during learning; in other words, training is done so that the network follows a predetermined trajectory. Numerical computer simulation results are provided to support this approach. This work was presented, in part, at the Fourth International Symposium on Artificial Life and Robotics, Oita, Japan, January 19–22, 1999.     

Highly accurate SNR measurement using the covariance of two SEM images with the identical view

Quality of an SEM image is strongly influenced by the extent of noise. As a well-known method in the field of SEM, the covariance is applied to measure the signal-to-noise ratio (SNR). This method has potential ability for highly accurate measurement of the SNR, which is hardly known until now. If the precautions discussed in this article are adopted, that method can demonstrate its real ability. These precautions are strongly related to "proper acquisition of two images with the identical view," "alignment of an aperture diaphragm," "reduction of charging phenomena," "elimination of particular noises," and "accurate focusing," As necessary, characteristics in SEM signal and noise are investigated from a few standpoints. When using the maximum performance of this measurement, SNR of many SEM images obtained in a variety of the SEM operating conditions and specimens can be measured accurately.     

A Quasi-ARMAX approach to modelling of non-linear systems

This paper proposes a class of quasi-ARMAX models for non-linear systems. Similar to ordinary non-linear ARMAX models, the quasi-ARMAX models are flexible black-box models, but they have various linearity properties similar to those of linear ARMAX models. A modelling scheme is introduced to construct models consisting of two parts: a macro-part and a kernel-part. By using Taylor expansion and other mathematical transformation techniques, it is first constructed as a class of quasi-ARMAX interfaces (macro-parts) that have various linearity properties but contain some complicated coefficients. MIMO neurofuzzy models (kernel-parts) are then introduced to represent the complicated coefficients. It is shown that the proposed quasi-ARMAX models have both good approximation ability and some easy-to-use properties. The proposed models have been successfully applied to prediction, fault detection and adaptive control of non-linear systems.     

A Study on the Effect of Air on the Dynamic Motion of a MEMS Device and Its Shape Optimization

The authors propose a new design concept for controlling the deflection of a micro-membrane with the aid of its thickness distribution for realizing a prescribed design in the MEMS. As an example, the authors treat a micro air pump that comprises a micro-membrane. The membrane is actuated by an electrostatic force. The membrane deflects and thus the deflection is influenced by the air pressure and the electrostatic field. This is a highly complicated system. To find out a proper thickness distribution, the authors use the genetic algorithm that is appropriate to reduce the searching space of solution.     

Breaking Ability and Interrupting Phenomena of a Circuit Breaker Equipped with a Parallel Resistor or Capacitor

This paper discusses the breaking ability of an air blast circuit breaker (ABB) or SF6 gas circuit breaker (GCB), and the interrupting phenomena around current zero in the case where a resistor or capacitor is connected in parallel with its interrupter to improve the breaking ability. It has been clarified that the use of a parallel resistor for ABB, and that of a parallel capacitor for GCB are the most effective methods to improve the breaking abilities; the parallel resistor not only serves to reduce the rate of rise of the transient recovery voltage (rrrV), but also shifts the current from the interrupter to itself around the current zero, thus reducing the stress for interruption. The improvement of the breaking ability is achieved by reducing the initial average rrrV between zero and 1 microseconds (inherent value), when the parallel capacitor is used with the GCB.     

A new learning method using prior information of neural networks

In this paper, we present a new learning method using prior information for three-layered neural networks. Usually when neural networks are used for identification of systems, all of their weights are trained independently, without considering their interrelation of weight values. Thus the training results are not usually good. The reason for this is that each parameter has its influence on others during the learning. To overcome this problem, first, we give an exact mathematical equation that describes the relation between weight values given by a set of data conveying prior information. Then we present a new learning method that trains a part of the weights and calculates the others by using these exact mathematical equations. In almost all cases, this method keeps prior information given by a mathematical structure exactly during the learning. In addition, a learning method using prior information expressed by inequality is also presented. In any case, the degree of freedom of networks (the number of     

Characterization of polymer films using a slow positron beam

Polymer structures have been investigated using positron annihilation lifetime spectroscopy (PALS) with a slow positron beam as well as a conventional radioactive source (²²Na). The properties of the free volume holes near the polymer surface were studied as a function of the positron implantation energy. The longest lifetime was associated with ortho-positronium (o-Ps) annihilation in the free volume holes. In polytetrafluoroethylene film, the lifetime of o-Ps was observed to decrease with increasing positron implantation depth, and a significant change in the o-Ps lifetime was found at a short distance (about 10 nm) from the surface, while its intensity increased. This result implies that near the polymer surface the free volume holes become larger that in the bulk. The effect of temperature on the polymer sub-surface layers was also studied. For high molecular weight polystyrene, the glass transition temperature for the sub-surface was lower than that for the bulk and thermal expansion coefficient of the sub-surface layers was found to be larger than the bulk value. Electronic Publication