Self-organizing map network as an interactive clustering tool — An application to group technology

The Self-Organizing Map (SOM) network, a variation of neural computing networks, is a categorization network developed by Kohonen. The theory of the SOM network is motivated by the observation of the operation of the brain. This paper presents the technique of SOM and shows how it may be applied as a clustering tool to group technology. A computer program for implementing the SOM neural networks is developed and the results are compared with other clustering approaches used in group technology. The study demonstrates the potential of using the Self-Organizing Map as the clustering tool for part family formation in group technology.     

Electronic structure and magnetic properties of the α and γ phases of iron, their solutions with carbon, and cementite

The full-potential LMTO method was used to comparatively analyze the band structure, magnetic and cohesive characteristics, energies of formation of α and γ iron, their carbon solid solutions, and cementite. It has been found that the solid solution γ-Fe + C, similarly to γ-Fe, is characterized by an antiinvar transition and that cementite demonstrates invar behavior. The similarity of curves of the density of states, cohesive energies, and magnetic moments for the supersaturated solid solution of carbon in the fcc iron and cementite suggests the preferable formation of cementite in regions of concentration inhomogeneity of carbon distribution in the solid solution of γ iron.     

Delta virus infection in cirrhotics in a north Indian hospital

Forty patients of HBV related cirrhosis of the liver were studied for the presence of delta infection and subsequently the clinical features, course and outcome of these delta infected cases was compared with those of delta uninfected cases. Out of 40 patients studied, only four patients (10%) were reactive for anti-delta antibodies. The incidence of delta infection was common in the younger age group (31.25 +/- 4.78 years) as compared to the older age group (48 +/- 7.25 years) at p < 0.001. Hepatic Encephalopathy was seen in only 13.88% of delta negative patients as compared to 75% of delta infected patients (p < 0.05). During follow up period of 1 year, 75% of patients died in the delta virus infected group as compared to 22.22% in the delta negative group. The mortality was significantly higher in Child's class C (P < 0.01). So delta virus infection is associated with a more severe course of illness and a poorer prognosis in HBV related cirrhotics.     

LDV MEASUREMENTS IN GAS-SOLID FLOWS—A REVIEW

This paper describes the history of LDV measurements employed in research involving gas-solid flows. Much of the discussion deals with measurement techniques employed to obtain separate and simultaneous solid and gas phase velocities. Also, LDV based measurement techniques that are used to obtain particle size and concentration information in conjunction with the velocity measurements are presented.     

Endogenous vasopressin increases acute endotoxin shock-provoked gastrointestinal mucosal injury in the rat

Administration of a low dose of endotoxin (from Escherichia coli, 3 mg kg(-1), i.v.), which does not affect vascular permeability or blood pressure over 1 h, leads to the release of endogenous vasopressin and damage to the mucosal microvasculature. Thus, endogenous vasopressin could be involved in septic shock. In the present study, we investigated the role of endogenous vasopressin in gastrointestinal mucosal injury induced by acute endotoxin shock, which was generated in rats by administering a high dose of E. coli endotoxin (50 mg kg(-1), i.v.). Tissues were removed 15 min after endotoxin. The vasopressin V1 receptor antagonist, d[CH2]5Tyr[Me]arginine-vasopressin (0.2-1 microg kg(-1), i.v.), was injected 10 min before endotoxin. Monastral blue (30 mg kg(-1), i.v.), which stains damaged vasculature, was injected 10 min before autopsy. Endotoxin reduced systemic arterial blood pressure (from 115+/-5 to 42+/-4 mmHg), generated macroscopic and microvascular injury, and elevated plasma vasopressin levels (from 3.4+/-0.2 to 178+/-16 pg ml(-1)). The vasopressin V1 receptor antagonist reduced this macroscopic injury, and in the vasopressin-deficient Brattleboro rat a similar reduction of gastrointestinal mucosal damage was found. Substantial decreases in endotoxin-induced microvascular damage were observed in each tissue, e.g., the gastric Monastral blue staining was reduced by 47+/-3% and 96+/-3% (P < 0.01) after vasopressin V1 receptor antagonist treatment and in Brattleboro rats, respectively. Vasopressin, acting through its V1 receptors, thus appears to be involved in acute endotoxin shock-provoked gastrointestinal injury.     

Changes in the structural localization of glycogen during glucose resorption

Quantitative changes in the glycogen granula contents of the dog's small intestinal mucosa were investigated with light microscopy using Best-carmin dying. In fasting animals the glycogen could be detected in a patchy arrangement in the supranuclear cytoplasmic region of the enterocytes. During glucose abasorption glycogen appeared in continually increasing quantities also in the lamina propria. Injecting Glucagon (50-100 microgram/15') into the artery of the intestinal loop during glucose absorption, glycogen totally disappeared from the enterocytes, while it remained unchanged in the cells of the lamina propria.     

End-to-end fair rate optimization in wired-cum-wireless networks

In this paper, we address the end-to-end rate optimization problem in a wired-cum-wireless network, where CSMA/CA based wireless LANs extend a wired backbone and provide access to mobile users. The objective is to achieve proportional fairness amongst the end-to-end sessions in the network. Since the network contains wireless links whose attainable throughput is a (non-convex and non-separable) function of MAC protocol parameters, the problem requires joint optimization at both the transport and the link layers. A dual-based algorithm is proposed in this paper to solve this cross-layer rate optimization problem. It is implemented in the distributed manner, and works at the link layer to adjust scheduling rates for the wireless links in the basic service sets, and at the transport layer to adjust end-to-end session rates. We prove rigorously that the proposed algorithm converges to the globally optimal rates. Simulation results are provided to support our conclusions.