Neural network techniques for object orientation detection.Solution by optimal feedforward network and learning vector quantizationapproaches

The computer-vision problem of determining object orientation from the consensus of orientations of individual symbols or marks is examined. The problem arises in automatic inspection where orientation can be detected from printed text but there is no knowledge of the content of the text. This is a high-dimensional classification problem, and there is a requirement for highly accurate detection and rapid processing. The typical multilayer threshold networks are seen as unsuitable, and the optimal Bayesian detector is derived and found to have the highly parallel structure of a feedforward network. The learning vector quantization neural network method of T. Kohonen (1988) is also applied. Experimental results, comparisons, and a complete implementation are described     

On Discriminative Bayesian Network Classifiers and Logistic Regression

Discriminative learning of the parameters in the naive Bayes model is known to be equivalent to a logistic regression problem. Here we show that the same fact holds for much more general Bayesian network models, as long as the corresponding network structure satisfies a certain graph-theoretic property. The property holds for naive Bayes but also for more complex structures such as tree-augmented naive Bayes (TAN) as well as for mixed diagnostic-discriminative structures. Our results imply that for networks satisfying our property, the conditional likelihood cannot have local maxima so that the global maximum can be found by simple local optimization methods. We also show that if this property does not hold, then in general the conditional likelihood can have local, non-global maxima. We illustrate our theoretical results by empirical experiments with local optimization in a conditional naive Bayes model. Furthermore, we provide a heuristic strategy for pruning the number of parameters and relevant features in such models. For many data sets, we obtain good results with heavily pruned submodels containing many fewer parameters than the original naive Bayes model.Editors: Pedro Larrañaga, Jose A. Lozano, Jose M. Peña and Iñaki Inza     

In Vitro and In Vivo Pharmaco-Toxicological Characterization of 1-Cyclohexyl-x-methoxybenzene Derivatives in Mice: Comparison with Tramadol and PCP

1-cyclohexyl-x-methoxybenzene is a novel psychoactive substance (NPS), first discovered in Europe in 2012 as unknown racemic mixture of its three stereoisomers: ortho, meta and para. Each of these has structural similarities with the analgesic tramadol and the dissociative anesthetic phencyclidine. In light of these structural analogies, and based on the fact that both tramadol and phencyclidine are substances that cause toxic effects in humans, the aim of this study was to investigate the in vitro and in vivo pharmacodynamic profile of these molecules, and to compare them with those caused by tramadol and phencyclidine. In vitro studies demonstrated that tramadol, ortho, meta and para were inactive at mu, kappa and delta opioid receptors. Systemic administration of the three stereoisomers impairs sensorimotor responses, modulates spontaneous motor activity, induces modest analgesia, and alters thermoregulation and cardiorespiratory responses in the mouse in some cases, with a similar profile to that of tramadol and phencyclidine. Naloxone partially prevents only the visual sensorimotor impairments caused by three stereoisomers, without preventing other effects. The present data show that 1-cyclohexyl-x-methoxybenzene derivatives cause pharmaco-toxicological effects by activating both opioid and non-opioid mechanisms and suggest that their use could potentially lead to abuse and bodily harm.     

ALCHEMIST: A General Purpose Transformation Generator

ALCHEMIST is a general purpose transformation generating environment, which supports specification, generation and execution of data transformations. ALCHEMIST allows an abstract specification of the transformation through a window-based interface and supports the generation and compilation of transformation program code from these specifications. Unlike compiler-compilers, ALCHEMIST is intended to automate building transformations between two complex representation formats and is thus especially suitable for constructing transformations between database tools, CASE tools, graphical editors or text formatters. In this paper we describe the design principles and the structure of ALCHEMIST, and demonstrate its use. We also discuss our experiences with several example transformations and present a real-life case study of using ALCHEMIST for interfacing two software development environments.     

Flame retardant polyurethane nanocomposite: Study of clay dispersion and its synergistic effect with dolomite

Polyurethane–clay nanocomposite adhesives were prepared by different synthetic routes and their microstructures were determined by X‐ray diffraction measurements and from transmission electron microscopy images. The preparation method of the polyurethane nanocomposite adhesives was systematically changed, that is, condensation either in the presence or absence of catalyst, concentration and type of nanoclay, premixing order of nanoclay (nanoclay was either premixed with the polyol or isocyanate part) and by using MDI surface treated nanoclays. Depending on the polymerization conditions cluster, intercalated, and exfoliated clay structures were obtained. The flame retardant properties of the manufactured nanocomposite adhesives and the synergistic effect of clay in combination with dolomite were investigated by cone calorimeter and UL 94 vertical burning tests. The results indicate that addition of nanoclay reduces burning time and the total heat evolved (THE) at flame out, and that the type of assembled clay structure (cluster, intercalated or exfoliated) had a significant effect on the flame retardant property. Nanocomposites with 3 wt % of clay loading gave the shortest burning time, the lowest THE and also UL 94 V‐2 ratings were reached, although the flame retardancy in terms of heat release rate and time to ignition was not improved. © 2012 Wiley Periodicals, Inc. J. Appl. Polym. Sci., 2013     

Variability of the clinical picture and the classification of progressive systemic scleroderma

Fiftythree patients with progressive systemic sclerosis were studied. Four of them (3 males) had the diffuse form of the disease. The skin manifestation of this clinical picture is characterized by diffuse progression of the cutaneous sclerosis over almost the whole body surface, except for the hands where it eventually may appear late. The prognosis for these patients it especially poor. Fortyfive patients (44 females) had acrosclerosis in the widest sence. Twentyseven of these ("acrosclerosis stricto sensu") had cutaneous sclerosis of the hands, face, and often other parts of the body, but not on the abdomen, arms or thighs. The remaining 18 patients had sclerotic alterations on these surfaces also. In this syndrome (which the authors call "the intermediary syndrome"), i.e. where the abdomen, arms, and thighs also are affected, certain internal organs and the joints are more involved than in "acrosclerosis stricto sensu". With rare exceptions, a symptomatic tetrade (REST-syndrome) occurred in acrosclerosis and all the intermediary syndromes. This consisted of Raynaud's syndrome (R), esophagopathy (E), cutaneous sclerosis (S), and telangectasia (T). Fifty % of the patients in addition had calcinosis (C), either subcutaneous or para-articular. The tetrade "REST syndrome" becomes in these cases the pentade "CREST syndrome". The addition of calcinosis to the other four phenomena of the REST syndrome does not alter the frequency of internal organ involvement or the prognosis of the disease. The term "REST syndrome" and its variant "CREST syndrome" should replace the conservative term "acrosclerosis" because they add to the purely cutaneous phenomena other characteristic manifestations of the disease. Two patients could neither be classified under the REST syndrome nor the progressive diffuse syndromes. Two other patients had no cutaneous phenomena ("scleroderma sine scleroderma").     

Studies on the hydrogen belts of membranes: I. Diester, diether, and dialkyl phosphatidycholines and polyoxyethylene glycerides in monolayers with cholesterol

The hydrogen belts of membranes are defined as the regions consisting of hydrogen bond acceptors, i.e., the C=O groups of glycero- and sphingolipids, and hydrogen bond donors, i.e., cholesterol-OH, sphingolipid-OH, proteins, and water. Lipid-lipid hydrogen bonding in these belts has been suggested. The connection of such hypothetical bonding with the condensation effect, i.e., the apparent reduction of surface area occupied by phospholipids in mixed monolayers with cholesterol, has been tested with lipids possissing and lacking C=O groups: diester, diether, and dialkyl phosphatidylcholine, and analogous polyoxyethylene diglycerides. Condensation by cholesterol was observed for all lipids. Consequently, the hypothetical lipid-C=O-cholesterol hydrogen bonding is not a prerequisite for the condensation effect.     

The effect of cyclophosphamide on the experimental inflammation induced by the toxic mushroom Cortinarius speciosissimus in the rat kidney

A single intraperitoneal dose of cyclophosphamide (150 mg/kg) given at the same time as an oral dose of Cortinarius speciosissimus prevented the renal inflammation induced by this toxic mushroom in the male rat. Furthermore, a scar formation around dilated collecting ducts was clearly reduced by cyclophosphamide treatment. In general the only lesions observed in the cyclophosphamide treated animals were dilated collecting ducts in the outer medullary zone, the epithelia of which were either in regenerative mitosis or were atrophic. Apparently the primary sites of action of Cortinarius toxins in male rats are the collecting ducts of the outer medullary zone. When inflammation and the subsequent scar formation is prevented by cyclophosphamide, the damaged tubules can regenerate by mitotic activity and perhaps restore normal function.     

Implementing expert system rule conditions by neural networks

The relation of subsymbolic (neural computing) and symbolic computing has been a topic of intense discussion. We address some of the drawbacks of current expert system technology and study the possibility of using neural computing principles to improve their competence. In this paper we focus on the problem of using neural networks to implement expert system rule conditions. Our approach allows symbolic inference engines to make direct use of complex sensory input via so called detector predicates. We also discuss the use of self organizing Kohonen networks as a means to determine those attributes (properties) of data that reflect meaningful statistical relationships in the expert system input space. This mechanism can be used to address the defficult problem of conceptual clustering of information. The concepts introduced are illustrated by two application examples: an automatic inspection system for circuit packs and an expert system for respiratory and anesthesia monitoring. The adopted approach differs from the earlier research on the use of neural networks as expert systems, where the only method to obtain knowledge is learning from training data. In our approach the synergy of rules and detector predicates combines the advantages of both worlds: it maintains the clarity of the rule-based knowledge representation at the higher reasoning levels without sacrificing the power of noise-tolerant pattern association offered by neural computing methods. This research is supported by Technology Development Center (TEKES) in Software Technology Programme (FINSOFT). Part of this work was done while the author was visiting AT & T Bell Laboratories.