Bayesian neural network approach for determining the risk of re-intervention after endovascular aortic aneurysm repair

This article proposes a Bayesian neural network approach to determine the risk of re-intervention after endovascular aortic aneurysm repair surgery. The target of proposed technique is to determine which patients have high chance to re-intervention (high-risk patients) and which are not (low-risk patients) after 5 years of the surgery. Two censored datasets relating to the clinical conditions of aortic aneurysms have been collected from two different vascular centers in the United Kingdom. A Bayesian network was first employed to solve the censoring issue in the datasets. Then, a back propagation neural network model was built using the uncensored data of the first center to predict re-intervention on the second center and classify the patients into high-risk and low-risk groups. Kaplan-Meier curves were plotted for each group of patients separately to show whether there is a significant difference between the two risk groups. Finally, the logrank test was applied to determine whether the neural network model was capable of predicting and distinguishing between the two risk groups. The results show that the Bayesian network used for uncensoring the data has improved the performance of the neural networks that were built for the two centers separately. More importantly, the neural network that was trained with uncensored data of the first center was able to predict and discriminate between groups of low risk and high risk of re-intervention after 5 years of endovascular aortic aneurysm surgery at center 2 (p = 0.0037 in the logrank test).

Industrial fault detection and diagnosis using Bayesian belief network

Rapid development in industry have contributed to more complex systems that are prone to failure. In applications where the presence of faults may lead to premature failure, fault detection and diagnostics tools are often implemented. The goal of this research is to improve the diagnostic ability of existing FDD methods. Kernel Principal Component Analysis has good fault detection capability, however it can only detect the fault and identify few variables that have contribution on occurrence of fault and thus not precise in diagnosing. Hence, KPCA was used to detect abnormal events and the most contributed variables were taken out for more analysis in diagnosis phase. The diagnosis phase was done in both qualitative and quantitative manner. In qualitative mode, a networked-base causality analysis method was developed to show the causal effect between the most contributing variables in occurrence of the fault. In order to have more quantitative diagnosis, a Bayesian network was constructed to analyze the problem in probabilistic perspective.

Rates of antimicrobial resistance in latin america (2004-2007) and in vitro activity of the glycylcycline tigecycline and of other antibiotics

As a part of the Tigecycline Evaluation and Surveillance Trial (T.E.S.T.), Gram-positive and Gram-negative bacterial isolates were collected from 33 centers in Latin America (centers in Argentina, Brazil, Chile, Colombia, Guatemala, Honduras, Jamaica, Mexico, Panama, Puerto Rico, and Venezuela) from January 2004 to September 2007. Argentina and Mexico were the greatest contributors of isolates to this study. Susceptibilities were determined according to Clinical Laboratory Standards Institute guidelines. Resistance levels were high for most key organisms across Latin America: 48.3% of Staphylococcus aureus isolates were methicillin-resistant while 21.4% of Acinetobacter spp. isolates were imipenem-resistant. Extended-spectrum β-lactamase were reported in 36.7% of Klebsiella pneumoniae and 20.8% of E. coli isolates. Tigecycline was the most active agent against Gram-positive isolates. Tigecycline was also highly active against all Gram-negative organisms, with the exception of Pseuodomonas aeruginosa, against which piperacillin-tazobactam was the most active agent tested (79.3% of isolates susceptible). The in vitro activity of tigecycline against both Gram-positive and Gram-negative isolates indicates that it may be an useful tool for the treatment of nosocomial infections, even those caused by organisms that are resistant to other antibacterial agents.

Purification and characterization of an antimicrobial peptide produced by Pseudomonas sp strain 4B

An antimicrobial peptide produced by a bacterium isolated from the effluent pond of a bovine abattoir was purified and characterized. The strain was characterized by biochemical profiling and 16S rDNA sequencing as Pseudomonas sp. The antimicrobial peptide was purified by ammonium sulfate precipitation, gel filtration, and ion exchange chromatography. Direct activity on sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) was observed. A major band on SDS-PAGE suggested that the antimicrobial peptide has a molecular mass of about 30 kDa. The substance was inhibitory to a broad range of indicator strains, including pathogenic and food spoilage bacteria such as Listeria monocytogenes, Bacillus cereus, Staphylococcus aureus, among other. The partially purified antimicrobial substance remained active over a wide temperature range and was resistant to all proteases tested. This substance showed different properties than other antimicrobials from Pseudomonas species, suggesting a novel antimicrobial peptide was characterized.

Modeling associations between genetic markers using Bayesian networks

Motivation: Understanding the patterns of association between polymorphisms at different loci in a population ( linkage disequilibrium, LD) is of fundamental importance in various genetic studies. Many coefficients were proposed for measuring the degree of LD, but they provide only a static view of the current LD structure. Generative models (GMs) were proposed to go beyond these measures, giving not only a description of the actual LD structure but also a tool to help understanding the process that generated such structure. GMs based in coalescent theory have been the most appealing because they link LD to evolutionary factors. Nevertheless, the inference and parameter estimation of such models is still computationally challenging. Results: We present a more practical method to build GM that describe LD. The method is based on learning weighted Bayesian network structures from haplotype data, extracting equivalence structure classes and using them to model LD. The results obtained in public data from the HapMap database showed that the method is a promising tool for modeling LD. The associations represented by the learned models are correlated with the traditional measure of LD D`. The method was able to represent LD blocks found by standard tools. The granularity of the association blocks and the readability of the models can be controlled in the method. The results suggest that the causality information gained by our method can be useful to tell about the conservability of the genetic markers and to guide the selection of subset of representative markers.

Using Bayesian networks with rule extraction to infer the risk of weed infestation in a corn-crop

This paper describes the modeling of a weed infestation risk inference system that implements a collaborative inference scheme based on rules extracted from two Bayesian network classifiers. The first Bayesian classifier infers a categorical variable value for the weed-crop competitiveness using as input categorical variables for the total density of weeds and corresponding proportions of narrow and broad-leaved weeds. The inferred categorical variable values for the weed-crop competitiveness along with three other categorical variables extracted from estimated maps for the weed seed production and weed coverage are then used as input for a second Bayesian network classifier to infer categorical variables values for the risk of infestation. Weed biomass and yield loss data samples are used to learn the probability relationship among the nodes of the first and second Bayesian classifiers in a supervised fashion, respectively. For comparison purposes, two types of Bayesian network structures are considered, namely an expert-based Bayesian classifier and a naive Bayes classifier. The inference system focused on the knowledge interpretation by translating a Bayesian classifier into a set of classification rules. The results obtained for the risk inference in a corn-crop field are presented and discussed. (C) 2009 Elsevier Ltd. All rights reserved.

Impact of antimicrobial resistance on the treatment and outcome of patients with sepsis

Antimicrobial therapy is one of the main stones of sepsis therapy. A recent study of septic shock patients showed that each hour of delay in antimicrobial administration during the ensuing 6 h after the onset of hypotension was associated with a decrease in survival rates. However, many questions regarding the impact of infection caused by antimicrobial-resistant pathogens on the mortality of patients with sepsis still need to be clarified. There is a lack of fair studies in the literature. Most studies have had inadequate sample size, inadequate adjustment for predictors of adverse outcomes, and inadequate definition of appropriate antibiotic therapy. Despite the fact that appropriate therapy is essential to treat sepsis, it seems that severity of underlying diseases and comorbidities are more important than resistance, although the studies were not well designed to examine the real impact of resistance on outcome. Finally, new technologies such as microarray that can identify different microorganisms, genes of resistance, and virulence in a few hours might have a great impact on the treatment of sepsis due to antimicrobial-resistant pathogens in the future.

PHB-PEO electrospun fiber membranes containing chlorhexidine for drug delivery applications

Fiber meshes of poly(hydroxybutyrate) (PHB) and poly(hydroxybutyrate)/ poly(ethylene oxide) (PHB/PEO) with different concentrations of chlorhexidine (CHX) were prepared by electrospinning, for assessment as a polymer based drug delivery system. The electrospun fibers were characterized at morphological, molecular and mechanical levels. The bactericidal potential of PHB and PHB/PEO electrospun fibers with and without CHX was investigated against Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus) by disk diffusion susceptibility tests. Electrospun fibers containing CHX exhibited bactericidal activity. PHB/PEO-1%CHX displayed higher CHX release levels and equivalent antibacterial activity when compared to PHB/PEO with 5 and 10 wt% CHX. Bactericidal performance of samples with 1 wt% CHX was assessed by Colony Forming Units (CFU), where a reduction of 100 % and 99.69 % against E. coli and S. aureus were achieved, respectively.

Comportamento do Staphilococcus aureus causadores de infecções hospitalares, frente a atividade "in vitro" da fosfomicina

Estudou-se a atividade "in vitro" da fosfomicina em 337 amostras de Staphylococcus aureus coletados de infecções intra-hospitalares, obtendo-se 310 amostras sensíveis (91,9%). Comparando-se com outros antimicrobianos, concluiu-se que a atividade da fosfomicina foi superior a todos.

Development and biological evaluation of API-ILs based on anti-bacterial and anti-fungal drugs

In recent years Ionic Liquids (ILs) are being applied in life sciences. ILs are being produce with active pharmaceutical drugs (API) as they can reduce polymorphism and drug solubility problems [1] Also ILs are being applied as a drug delivery device in innovative therapies What is appealing in ILs is the ILs building up platform, the counter-ion can be carefully chosen in order to avoid undesirable side effects or to give innovative therapies in which two active ions are paired. This work shows ILs based on ampicillin (an anti-bacterial agent) and ILs based on Amphotericin B. Also we show studies that indicate that ILs based on Ampicillin could reverse resistance in some bacteria. The ILs produced in this work were synthetized by the neutralization method described in Ferraz et. al. [2] Ampicillin anion was combined with the following organic cations 1-ethyl-3-methylimidazolium, [EMIM]; 1-hydroxy-ethyl-3-methylimidazolium, [C2OHMIM]; choline, [cholin]; tetraethylammonium, [TEA]; cetylpyridinium, [C16pyr] and trihexyltetradecylphosphonium, [P6,6,6,14]. Amphotericin B was combined with [C16pyr], [cholin] and 1-metohyethyl-3-methylimidazolium, [C3OMIM]. The ILs-APIs based on ampicillin[2] were tested against sensitive Gram-negative bacteria Escherichia coli ATCC 25922 and Klebsiella pneumonia (clinical isolated), as well as on Gram positive Staphylococcus Aureus ATCC 25923, Staphylococcus epidermidis and Enterococcus faecalis. The arising resistance developed by bacteria to antibiotics is a serious public health threat and needs new and urgent measures. We study the bacterial activity of these compounds against a panel of resistant bacteria (clinical isolated strains): E. coli CTX M9, E. coli TEM CTX M9, E. coli TEM1, E. coli CTX M2, E. coli AmpC Mox2. In this work we demonstrate that is possible to produce ILs from anti-bacterial and anti-fungal compounds. We show here that the new ILs can reverse the bacteria resistance. With the careful choice of the organic cation, it is possible to create important biological and physic-chemical properties. This work also shows that the ion-pair is fundamental in ampicillin mechanism of action.

Reinforcement Learning Based on the Bayesian Theorem for Electricity Markets Decision Support

This paper presents the applicability of a reinforcement learning algorithm based on the application of the Bayesian theorem of probability. The proposed reinforcement learning algorithm is an advantageous and indispensable tool for ALBidS (Adaptive Learning strategic Bidding System), a multi-agent system that has the purpose of providing decision support to electricity market negotiating players. ALBidS uses a set of different strategies for providing decision support to market players. These strategies are used accordingly to their probability of success for each different context. The approach proposed in this paper uses a Bayesian network for deciding the most probably successful action at each time, depending on past events. The performance of the proposed methodology is tested using electricity market simulations in MASCEM (Multi-Agent Simulator of Competitive Electricity Markets). MASCEM provides the means for simulating a real electricity market environment, based on real data from real electricity market operators.

Epidemiological analysis of bacterial strains involved in hospital infection in a University Hospital from Brazil

Hospital infections cause an increase in morbidity and mortality of hospitalized patients with significant rise in hospital costs. The aim of this work was an epidemiological analysis of hospital infection cases occurred in a public University Hospital in Rio de Janeiro. Hence, 238 strains were isolated from 14 different clinical materials of 166 patients hospitalized in the period between August 1995 and July 1997. The average age of the patients was 33.4 years, 72.9% used antimicrobials before having a positive culture. The most common risk conditions were surgery (19.3%), positive HIV or AIDS (18.1%) and lung disease (16.9%). 24 different bacterial species were identified, S. aureus (21%) and P. aeruginosa (18.5%) were predominant. Among 50 S. aureus isolated strains 36% were classified as MRSA (Methicillin Resistant S. aureus). The Gram negative bacteria presented high resistance to aminoglycosides and cephalosporins. A diarrhea outbreak, detected in high-risk neonatology ward, was caused by Salmonella serovar Infantis strain, with high antimicrobial resistance and a plasmid of high molecular weight (98Mda) containing virulence genes and positive for R factor.

Importance of hand germ contamination in health-care workers as possible carriers of nosocomial infections

The importance of hands in the transmission of nosocomial infection has been world wide admitted. However, it is difficult to induce this behavior in health-care workers. The aim of the present work was to point out the importance of hand bacteria colonization, the influence of hand washing and of patient physical examination. One hundred health-care workers were randomly divided in two groups: Group A without hand washing previous to patient physical examination or handling (PPE); group B with hand washing previous to PPE. Direct fingerprint samples in Columbia agar before and after PPE were obtained. The colonies were counted and identified by conventional techniques, and antibiograms according to NCCLS were performed. Before PPE group A participants showed a high number of bacteria regarding group B participants (73.9 Vs 20.7; p < 0.001); 44 out of 50 participants were carriers of potentially pathogen bacteria. No group B participants were carriers of potential pathogen bacteria before PPE. The latter group showed an increase in number of bacteria after PPE (20.7 CFU (before) Vs 115.9 CFU (after); p < 0.001). Sixteen group B participants were contaminated after PPE with potential pathogens such as S. aureus (50% of them meticillin resistant); Escherichia coli, Pseudomonas aeruginosa and Enterococcus faecalis, half of them multiresistant. We can conclude on the importance of these results to implement educational programs and to provide the health-care workers with the proper commodities to fulfill this practice.

Análise Retrospectiva do Protocolo de Tratamento Utilizado na Consulta Multidisciplinar de Úlcera de Perna do Hospital dos Capuchos no Período entre 2002 e 1º Semestre 2006

As úlceras de perna constituem uma importante patologia causando uma diminuição da qualidade de vida, hospitalizações frequentes e aumento da mortalidade e morbilidade. Têm uma incidência de 1% na população adulta, sendo que esta incidência atinge níveis de 10% nos escalões etários superiores a 70 anos. Cerca de 95% das úlceras são venosas, arteriais, mistas ou diabéticas, sendo as mais frequentes as úlceras venosas (70 a 80%). Com o objectivo de optimizar o tratamento e acompanhamento dos doentes com esta patologia, foi criada em 2002 uma Consulta de Referência Multidisciplinar de Úlcera de Perna, no Hospital dos Capuchos. Simultaneamente foi estabelecido um protocolo de referenciação/ tratamento com os Centros de Saúde da Unidade B da Sub-região de Saúde de Lisboa. Neste protocolo o doente é observado no contexto de uma equipa multidisciplinar. Os autores fizeram um estudo retrospectivo dos doentes observados nesta consulta no período entre 2002 e 1º semestre de 2006. Foram observados e acompanhados 294 novos doentes, tendo 80% idade superior a 60 anos. Em relação à etiologia das úlceras, 51,3% (n=151) eram venosas, 35,4% (n=104) eram diabéticas e 6,8% (n=20) eram arteriais. A área média das úlceras foi 23,9cm2 e o número médio de úlceras foi 1,6. A duração das úlceras tinha em 42,3% dos casos um período superior a 6 meses. Das 199 culturas positivas, 40,2% apresentavam Staphylococcus aureus, sendo 21,2% destes MRSA. Com o protocolo instituído, foi obtida uma taxa de cicatrização de 72,2%. 45,9% dos doentes tiveram uma cicatrização total da úlcera em menos de 2 meses, resultados estes que são muito positivos face às taxas de cicatrização de 6 meses referidas na literatura.

EVALUATION OF ANTIMICROBIAL AND CYTOTOXIC ACTIVITIES OF PLANT EXTRACTS FROM SOUTHERN MINAS GERAIS CERRADO

The antimicrobial activity of plant hidroethanolic extracts on bacteria Gram positive, Gram negative, yeasts, Mycobacterium tuberculosis H37 and Mycobacterium bovis was evaluated by using the technique of Agar diffusion and microdilution in broth. Among the extracts evaluated by Agar diffusion, the extract of Bidens pilosa leaf presented the most expressive average of haloes of growth inhibition to the microorganisms, followed by the extract of B. pilosa flower, of Eugenia pyriformis' leaf and seed, of Plinia cauliflora leaf which statistically presented the same average of haloes inhibitory formation on bacteria Gram positive, Gram negative and yeasts. The extracts of Heliconia rostrata did not present activity. Mycobacterium tuberculosis H37 and Mycobacterium bovis(BCG) appeared resistant to all the extracts. The susceptibility profile of Candida albicans and Saccharomyces cerevisiae fungi were compared to one another and to the Gram positive Bacillus subtilis, Enterococcus faecalis and the Gram negative Salmonella typhimurium bacteria (p > 0.05). The evaluation of cytotoxicity was carried out on C6-36 larvae cells of the Aedes albopictus mosquito. The extracts of stem and flower of Heliconia rostrata, leaf and stem of Plinia cauliflora, seed of Anonna crassiflora and stem, flower and root of B. pilosa did not present toxicity in the analyzed concentrations. The highest rates of selectivity appeared in the extracts of stem of A. crassiflora and flower of B. pilosa to Staphylococcus aureus, presenting potential for future studies about a new drug development.