Network-based high level data classification

Traditional supervised data classification considers only physical features (e. g., distance or similarity) of the input data. Here, this type of learning is called low level classification. On the other hand, the human (animal) brain performs both low and high orders of learning and it has facility in identifying patterns according to the semantic meaning of the input data. Data classification that considers not only physical attributes but also the pattern formation is, here, referred to as high level classification. In this paper, we propose a hybrid classification technique that combines both types of learning. The low level term can be implemented by any classification technique, while the high level term is realized by the extraction of features of the underlying network constructed from the input data. Thus, the former classifies the test instances by their physical features or class topologies, while the latter measures the compliance of the test instances to the pattern formation of the data. Our study shows that the proposed technique not only can realize classification according to the pattern formation, but also is able to improve the performance of traditional classification techniques. Furthermore, as the class configuration's complexity increases, such as the mixture among different classes, a larger portion of the high level term is required to get correct classification. This feature confirms that the high level classification has a special importance in complex situations of classification. Finally, we show how the proposed technique can be employed in a real-world application, where it is capable of identifying variations and distortions of handwritten digit images. As a result, it supplies an improvement in the overall pattern recognition rate.

Network-based stochastic semisupervised learning

Semisupervised learning is a machine learning approach that is able to employ both labeled and unlabeled samples in the training process. In this paper, we propose a semisupervised data classification model based on a combined random-preferential walk of particles in a network (graph) constructed from the input dataset. The particles of the same class cooperate among themselves, while the particles of different classes compete with each other to propagate class labels to the whole network. A rigorous model definition is provided via a nonlinear stochastic dynamical system and a mathematical analysis of its behavior is carried out. A numerical validation presented in this paper confirms the theoretical predictions. An interesting feature brought by the competitive-cooperative mechanism is that the proposed model can achieve good classification rates while exhibiting low computational complexity order in comparison to other network-based semisupervised algorithms. Computer simulations conducted on synthetic and real-world datasets reveal the effectiveness of the model.

Integrated landmark and outline-based morphometric methods efficiently distinguish species of Euglossa (Hymenoptera, Apidae, Euglossini)

Morphometric methods permit identification of insect species and are an aid for taxonomy. Quantitative wing traits were used to identify male euglossine bees. Landmark- and outline-based methods have been primarily used independently. Here, we combine the two methods using five Euglossa. Landmark-based methods correctly classified 84% and outline-based 77%, but an integrated analysis correctly classified 91% of samples. Some species presented significantly high reclassification percentages when only wing cell contour was considered, and correct identification of specimens with damaged wings was also obtained using this methodology.

Particle Competition and Cooperation in Networks for Semi-Supervised Learning

Semi-supervised learning is one of the important topics in machine learning, concerning with pattern classification where only a small subset of data is labeled. In this paper, a new network-based (or graph-based) semi-supervised classification model is proposed. It employs a combined random-greedy walk of particles, with competition and cooperation mechanisms, to propagate class labels to the whole network. Due to the competition mechanism, the proposed model has a local label spreading fashion, i.e., each particle only visits a portion of nodes potentially belonging to it, while it is not allowed to visit those nodes definitely occupied by particles of other classes. In this way, a "divide-and-conquer" effect is naturally embedded in the model. As a result, the proposed model can achieve a good classification rate while exhibiting low computational complexity order in comparison to other network-based semi-supervised algorithms. Computer simulations carried out for synthetic and real-world data sets provide a numeric quantification of the performance of the method.

Application of discriminant analysis-based model for prediction of risk of low back disorders due to workplace design in industrial jobs

The occupational exposure limits of different risk factors for development of low back disorders (LBDs) have not yet been established. One of the main problems in setting such guidelines is the limited understanding of how different risk factors for LBDs interact in causing injury, since the nature and mechanism of these disorders are relatively unknown phenomena. Industrial ergonomists' role becomes further complicated because the potential risk factors that may contribute towards the onset of LBDs interact in a complex manner, which makes it difficult to discriminate in detail among the jobs that place workers at high or low risk of LBDs. The purpose of this paper was to develop a comparative study between predictions based on the neural network-based model proposed by Zurada, Karwowski & Marras (1997) and a linear discriminant analysis model, for making predictions about industrial jobs according to their potential risk of low back disorders due to workplace design. The results obtained through applying the discriminant analysis-based model proved that it is as effective as the neural network-based model. Moreover, the discriminant analysis-based model proved to be more advantageous regarding cost and time savings for future data gathering.

Causes of pulmonary granulomas: a retrospective study of 500 cases from seven countries

Background The frequencies of various causes of pulmonary granulomas in pathological material are unknown, as is the influence of geographical location on aetiology. The aim of this study was to identify the causes of pulmonary granulomas in pathological specimens, to define their frequencies, and to determine whether these causes vary by geographical location. Methods 500 lung biopsies and resections containing granulomas were reviewed retrospectively by expert pulmonary pathologists from 10 institutions in seven countries. Fifty consecutive cases from each location were assigned a diagnosis based on histological features and available clinical/microbiological data. Results A specific cause was identified in 58% of cases (290/500), most commonly sarcoidosis (136, 27%) and mycobacterial or fungal infections (125, 25%). Mycobacteria were identified in 19% of cases outside the USA versus 8% within the USA. In contrast, fungi accounted for 19% cases in the USA versus 4% in other locations. Fungi were mostly detected by histology, whereas most mycobacteria were identified in cultures. In 42% of cases (210/500) an aetiology could not be determined. Conclusions Across several geographical settings, sarcoidosis and infections are the most common causes of pulmonary granulomas diagnosed in pathological specimens. Fungi are more commonly identified than mycobacteria in the USA, whereas the reverse is true in other countries. A definite aetiology cannot be demonstrated in more than a third of all cases of pulmonary granulomas, even after histological examination. These findings highlight the need to submit material for histology as well as cultures in all cases in which granulomatous disease enters the differential diagnosis.

Extensive cross-talk and global regulators identified from an analysis of the integrated transcriptional and signaling network in Escherichia coli

To understand the regulatory dynamics of transcription factors (TFs) and their interplay with other cellular components we have integrated transcriptional, protein-protein and the allosteric or equivalent interactions which mediate the physiological activity of TFs in Escherichia coli. To study this integrated network we computed a set of network measurements followed by principal component analysis (PCA), investigated the correlations between network structure and dynamics, and carried out a procedure for motif detection. In particular, we show that outliers identified in the integrated network based on their network properties correspond to previously characterized global transcriptional regulators. Furthermore, outliers are highly and widely expressed across conditions, thus supporting their global nature in controlling many genes in the cell. Motifs revealed that TFs not only interact physically with each other but also obtain feedback from signals delivered by signaling proteins supporting the extensive cross-talk between different types of networks. Our analysis can lead to the development of a general framework for detecting and understanding global regulatory factors in regulatory networks and reinforces the importance of integrating multiple types of interactions in underpinning the interrelationships between them.

Dental fluorosis in the primary dentition and intake of manufactured soy-based foods with fluoride

Background & aims: To identify manufactured soy-based products more recommended by pediatricians and nutritionists; to determine fluoride concentrations in these products; to evaluate children concerning fluorosis in primary teeth and its association with the consumption of soy-based products. Methods: Pediatricians and Nutritionists answered a questionnaire about soy-based products they most recommended to children. Fluoride concentrations of the 10 products more cited were analyzed with the ion-specific electrode. Dental fluorosis exams were performed in 315 4e6-year-old children. Dean’s Index was used to assess fluorosis. Among the children examined, 26 had lactose intolerance. Their parents answered a questionnaire about children’s and family’s profile, besides permitting the identification of soy-based products use. Chi-squared and Multivariable Logistic Regression tests were used (p < 0.05). Results: Fluoride content in the analyzed products ranged from 0.03 to 0.50 mg F /mL. Dental fluorosis was detected in 11% of the children, with very mild and mild degrees. Dental fluorosis in primary teeth was associated with lactose intolerance (p < 0.05), but there was no significant association with the use of manufactured soy-based products. Conclusions: Isolated consumption of soy-based products recommended by health professionals to children do not offer risk of dental fluorosis in primary teeth, which had a low prevalence and severity.

Detecting an abnormality in a recovery boiler using dynamic multivariate data analysis with parallel coordinate plots

Data visualization techniques are powerful in the handling and analysis of multivariate systems. One such technique known as parallel coordinates was used to support the diagnosis of an event, detected by a neural network-based monitoring system, in a boiler at a Brazilian Kraft pulp mill. Its attractiveness is the possibility of the visualization of several variables simultaneously. The diagnostic procedure was carried out step-by-step going through exploratory, explanatory, confirmatory, and communicative goals. This tool allowed the visualization of the boiler dynamics in an easier way, compared to commonly used univariate trend plots. In addition it facilitated analysis of other aspects, namely relationships among process variables, distinct modes of operation and discrepant data. The whole analysis revealed firstly that the period involving the detected event was associated with a transition between two distinct normal modes of operation, and secondly the presence of unusual changes in process variables at this time.

A complex networks approach for data clustering

This work proposes a method for data clustering based on complex networks theory. A data set is represented as a network by considering different metrics to establish the connection between each pair of objects. The clusters are obtained by taking into account five community detection algorithms. The network-based clustering approach is applied in two real-world databases and two sets of artificially generated data. The obtained results suggest that the exponential of the Minkowski distance is the most suitable metric to quantify the similarities between pairs of objects. In addition, the community identification method based on the greedy optimization provides the best cluster solution. We compare the network-based clustering approach with some traditional clustering algorithms and verify that it provides the lowest classification error rate. (C) 2012 Elsevier B.V. All rights reserved.

Architectures, stability and optimization for clock distribution networks

Synchronous telecommunication networks, distributed control systems and integrated circuits have its accuracy of operation dependent on the existence of a reliable time basis signal extracted from the line data stream and acquirable to each node. In this sense, the existence of a sub-network (inside the main network) dedicated to the distribution of the clock signals is crucially important. There are different solutions for the architecture of the time distribution sub-network and choosing one of them depends on cost, precision, reliability and operational security. In this work we expose: (i) the possible time distribution networks and their usual topologies and arrangements. (ii) How parameters of the network nodes can affect the reachability and stability of the synchronous state of a network. (iii) Optimizations methods for synchronous networks which can provide low cost architectures with operational precision, reliability and security. (C) 2011 Elsevier B. V. All rights reserved.

COMPARISON OF DIFFERENT PROTOCOLS FOR THE EXTRACTION OF MICROBIAL DNA FROM REEF CORALS

This study aimed to test different protocols for the extraction of microbial DNA from the coral Mussismilia harttii. Four different commercial kits were tested, three of them based on methods for DNA extraction from soil (FastDNA SPIN Kit for soil, MP Bio, PowerSoil DNA Isolation Kit, MoBio, and ZR Soil Microbe DNA Kit, Zymo Research) and one kit for DNA extraction from plants (UltraClean Plant DNA Isolation Kit, MoBio). Five polyps of the same colony of M. harttii were macerated and aliquots were submitted to DNA extraction by the different kits. After extraction, the DNA was quantified and PCR-DGGE was used to study the molecular fingerprint of Bacteria and Eukarya. Among the four kits tested, the ZR Soil Microbe DNA Kit was the most efficient with respect to the amount of DNA extracted, yielding about three times more DNA than the other kits. Also, we observed a higher number and intensities of DGGE bands for both Bacteria and Eukarya with the same kit. Considering these results, we suggested that the ZR Soil Microbe DNA Kit is the best adapted for the study of the microbial communities of corals.

The Impact of Tacrolimus as Rescue Therapy in Children Using a Double Immunosuppressive Regimen After Heart Transplantation

Background. Organ transplant recipients with refractory rejection or intolerance to the prescribed immunosuppressant may respond to rescue therapy with tacrolimus. We sought to evaluate the clinical outcomes of children undergoing heart transplantation who required conversion from a cyclosporine-based, steroid-free therapy to a tacrolimus-based regimen. Methods. We performed a prospective, observational, cohort study of 28 children who underwent conversion from cyclosporine-based, steroid-free therapy to a tacrolimus-based therapy for refractory or late rejection or intolerance to cyclosporine. Results. There was complete resolution of refractory rejection episodes and adverse side effects in all patients. The incidence rate (X100) of rejection episodes before and after conversion was 7.98 and 2.11, respectively (P <= .0001). There was a 25% mortality rate in patients using tacrolimus after a mean period of 60 months after conversion. Conclusion. Tacrolimus is effective as rescue therapy for refractory rejection and is a therapeutic option for pediatric patients.

Influence of combined oral contraceptives on the periodontal condition

Most studies investigating the impact of oral contraceptives have been performed some years ago, when the level of sexual hormones was greater than the actual formulations. Objective: The aim of this study was to evaluate the effects of current combined oral contraceptives (COC) on periodontal tissues, correlating the clinical parameters examined with the total duration of continuous oral contraceptive intake. Material and methods: Twenty-five women (19-35 years old) taking combined oral contraceptives for at least 1 year were included in the test group. The control group was composed by 25 patients at the same age range reporting no use of hormone-based contraceptive methods. Clinical parameters investigated included pocket probing depth (PD), clinical attachment level (CAL), sulcular bleeding index (SBI) and plaque index (PI.I). Data were statistically evaluated by unpaired t test, Pearson's correlation test and Spearman's correlation test. Results: The test group showed increased PD (2.228+/-0.011 x 2.154+/-0.012; p<0.0001) and SBI (0.229+/-0.006 x 0.148+/-0.005, p<0.0001) than controls. No significant differences between groups were found in CAL (0.435+/-0.01 x 0.412+/-0.01; p=0.11). The control group showed greater PI.I than the test group (0.206+/-0.007 x 0.303+/-0.008; p<0.0001). No correlation between the duration of oral contraceptive intake, age and periodontal parameters was observed. Conclusions: These findings suggest that the use of currently available combined oral contraceptives can influence the periodontal conditions of the patients, independently of the level of plaque accumulation or total duration of medication intake, resulting in increased gingival inflammation.