A hybrid knowledge and ensemble classification approach for prediction of venous thromboembolism

Clinical narratives such as progress summaries, lab reports, surgical reports, and other narrative texts contain key biomarkers about a patient's health. Evidence-based preventive medicine needs accurate semantic and sentiment analysis to extract and classify medical features as the input to appropriate machine learning classifiers. However, the traditional approach of using single classifiers is limited by the need for dimensionality reduction techniques, statistical feature correlation, a faster learning rate, and the lack of consideration of the semantic relations among features. Hence, extracting semantic and sentiment-based features from clinical text and combining multiple classifiers to create an ensemble intelligent system overcomes many limitations and provides a more robust prediction outcome. The selection of an appropriate approach and its interparameter dependency becomes key for the success of the ensemble method. This paper proposes a hybrid knowledge and ensemble learning framework for prediction of venous thromboembolism (VTE) diagnosis consisting of the following components: a VTE ontology, semantic extraction and sentiment assessment of risk factor framework, and an ensemble classifier. Therefore, a component-based analysis approach was adopted for evaluation using a data set of 250 clinical narratives where knowledge and ensemble achieved the following results with and without semantic extraction and sentiment assessment of risk factor, respectively: a precision of 81.8% and 62.9%, a recall of 81.8% and 57.6%, an F measure of 81.8% and 53.8%, and a receiving operating characteristic of 80.1% and 58.5% in identifying cases of VTE.     

Encapsulation of phase change materials with alginate modified by nanostructured sodium carbonate and silicate

The encapsulation of phase change materials (PCMs) as thermal energy storage materials is a big issue. PCM is usually encapsulated to avoid spillage, flammability and its reaction with the surrounding environment to improve its application. In the last decade, various methods have been employed and all kinds of microencapsulated PCM are produced. In this paper, we present a facile route to produce an encapsulated PCM with an organic and inorganic shell. The encapsulated phase change material (PCM) was prepared using a coaxial micro-fluidic system combined with an ionic cross-linking process. The alginate was used as the basic shell and a range of capsules was obtained by modifying the original shell using two inorganic components such as sodium carbonate and sodium silicate. Various samples, each with a different surrounding layer, were prepared by combining alginate calcium (Alg–Ca) as an organic shell with an inorganic component such as alginate calcium carbonate (Alg–CaCO3) and alginate calcium silicate (Alg–CaSiO3). In these experimental works, we have investigated the compatibility and the stability of capsules modified with the inorganic component. The scanning electron microscopy (SEM) technique and optical microscopy were utilized to study the capsule morphology. The chemical composition of the shell was evaluated by Fourier transform infrared spectroscopy (ATR-FTIR), thermogravimetry analysis and SEM coupled with the EDX analysis, and the capsule stability was estimated under an accelerated thermal cycling.     

A new generalised α scale spaces quadrature filters

The α   scale spaces is a recent theory that open new possibilities of phase-based image processing. It is a parameterised class (α∈]0,1])$(\alpha \in ]0,1\left]\right)$ of linear scale space representations, which allows a continuous connection beyond the well-known Gaussian scale space (α=1$\alpha =1$). In this paper, we make use of this unified representation to derive new families of band pass quadrature filters, built from derivatives and difference of the α scale space generating kernels. This construction leads to generalised α kernel filters including the commonly known families derived from the Gaussian and the Poisson kernels. The properties of each family are first studied and then experiments on one- and two-dimensional signals are shown to exemplify how the suggested filters can be used for edge detection. This work is complemented by an experimental evaluation, which demonstrates that the new proposed filters are a good alternative to the commonly used Log-Gabor filter.     

Parallelised production of fine and calibrated emulsions by coupling flow-focusing technique and partial wetting phenomenon

The capacity of microfluidic technology to fabricate monodisperse emulsion droplets is well established. Parallelisation of droplet production is a prerequisite for using such an approach for making high-quality materials for either fundamental or industrial applications where product quantity matters. Here, we investigate the emulsification efficiency of parallelised drop generators based on a flow-focusing geometry when incorporating the role of partial wetting in order to make emulsion droplets with a diameter below 10 μm. Confinement intrinsically encountered in microsystems intensifies the role played by interfaces between liquids and solids. We thus take advantage of partial wetting to enhance the maximum confinement accessible due to liquid flow focusing. We compare the performances brought by partial wetting to more established routes such as step emulsification. We show that the step configuration and the partial wetting regime are both well suited for being parallelised and thus open the way to the production of fine and calibrated emulsions for further applications. Finally, this new route of emulsification that exploits partial wetting between the fluids and the channel walls opens possibilities to the formation of substantially smaller droplets, as required in many fields of application.     

Feature selection on probabilistic symbolic objects

In data analysis tasks, we are often confronted to very high dimensional data. Based on the purpose of a data analysis study, feature selection will find and select the relevant subset of features from the original features. Many feature selection algorithms have been proposed in classical data analysis, but very few in symbolic data analysis （SDA） which is an extension of the classical data analysis, since it uses rich objects instead to simple matrices. A symbolic object, compared to the data used in classical data analysis can describe not only individuals, but also most of the time a cluster of individuals. In this paper we present an unsupervised feature selection algorithm on probabilistic symbolic objects （PSOs）, with the purpose of discrimination. A PSO is a symbolic object that describes a cluster of individuals by modal variables using relative frequency distribution associated with each value. This paper presents new dissimilarity measures between PSOs, which are used as feature selection criteria, and explains how to reduce the complexity of the algorithm by using the discrimination matrix.     

Large-scale linked data integration using probabilistic reasoning and crowdsourcing

We tackle the problems of semiautomatically matching linked data sets and of linking large collections of Web pages to linked data. Our system, ZenCrowd, (1) uses a three-stage blocking technique in order to obtain the best possible instance matches while minimizing both computational complexity and latency, and (2) identifies entities from natural language text using state-of-the-art techniques and automatically connects them to the linked open data cloud. First, we use structured inverted indices to quickly find potential candidate results from entities that have been indexed in our system. Our system then analyzes the candidate matches and refines them whenever deemed necessary using computationally more expensive queries on a graph database. Finally, we resort to human computation by dynamically generating crowdsourcing tasks in case the algorithmic components fail to come up with convincing results. We integrate all results from the inverted indices, from the graph database and from the crowd using a probabilistic framework in order to make sensible decisions about candidate matches and to identify unreliable human workers. In the following, we give an overview of the architecture of our system and describe in detail our novel three-stage blocking technique and our probabilistic decision framework. We also report on a series of experimental results on a standard data set, showing that our system can achieve a 95 % average accuracy on instance matching (as compared to the initial 88 % average accuracy of the purely automatic baseline) while drastically limiting the amount of work performed by the crowd. The experimental evaluation of our system on the entity linking task shows an average relative improvement of 14 % over our best automatic approach.     

Global stabilization of a class of delay discrete-time nonlinear systems via state and output feedback

This paper deals with stabilization of a class of delay discrete-time nonlinear systems through state and output feedback. We provide an explicit bounded state feedback law as an extension of the Jurdjevic-Quinn method, from nonlinear theory, to this class of systems. Next, we present a useful and systematic approach to design an observer for the same class of systems. Then, we show how the global stabilization problem via dynamic output feedback can be solved by using the two previous results. Finally, numerical examples are given to illustrate the effectiveness of the proposed design method.     

Hybrid metaheuristics for scheduling of machines and transport robots in job shop environment

In real manufacturing environments, the control of some elements in systems based on robotic cells, such as transport robots has some difficulties when planning operations dynamically. The Job Shop scheduling Problem with Transportation times and Many Robots (JSPT-MR) is a generalization of the classical Job Shop scheduling Problem (JSP) where a set of jobs additionally have to be transported between machines by several transport robots. Hence, the JSPT-MR is more computationally difficult than the JSP presenting two NP-hard problems simultaneously: the job shop scheduling problem and the robot routing problem. This paper proposes a hybrid metaheuristic approach based on clustered holonic multiagent model for the JSPT-MR. Firstly, a scheduler agent applies a Neighborhood-based Genetic Algorithm (NGA) for a global exploration of the search space. Secondly, a set of cluster agents uses a tabu search technique to guide the research in promising regions. Computational results are presented using two sets of benchmark literature instances. New upper bounds are found, showing the effectiveness of the presented approach.     

Detection of pathogenic Escherichia coli strains in groundwater in the Yaoundé region (Cameroon,Central Africa)

This study was aimed at assessing the prevalence of pathogenic Escherichia coli strains, relative to the total count of E. coli, faecal coliforms and other heterotrophic mesophilic aerobic bacteria (HMAB) isolated in groundwater in the equatorial region of Cameroon (Central Africa). Bacteria were isolated using standard methods. Pathogenic E. coli strains were then identified using haemagglutination and antisera tests. The maximum abundance of HMAB, faecal coliforms and E. coli strains were 4.9 × 10⁶, 5.6 × 10³ and 1 × 10³ colony‐forming units (CFU)/100 mL, respectively. The count of pathogenic E. coli strains reached 3 CFU/100 mL. The counts of commensal and pathogenic E. coli strains underwent temporal and spatial fluctuations. In 21% of sampling sites, the abundance of faecal coliforms was significantly correlated to that of E. coli (P < 0.05). However, the isolated bacterial count was not significantly correlated to that of the pathogenic E. coli strains (P > 0.05). The bacteria abundance dynamics may be impacted by many interacting factors.     

Combined microwave and enzymatic treatments for ��-lactoglobulin and bovine whey proteins and their effect on the IgE immunoreactivity

Effect of combined microwave (MW) and enzymatic hydrolysis on the human immunoglobulin E (IgE)-binding properties of ??-lactoglobulin (??-lg) and other whey proteins (WP) was investigated. Separated ??-lg and full whey protein isolate (WPI) were hydrolyzed with trypsin, chymotrypsin, mixture of trypsin/chymotrypsin, and pepsin at three microwave power levels: 50?W during 1 and 5?min, 100 and 200?W during 1 and 3?min. The immunoreactivity of the obtained hydrolysates resulting from combined microwave protease treatment was assessed using sera of young patients allergic to bovine whey proteins. The application of microwave treatment at 200?W enhances the hydrolysis of ??-lg by pepsin in 3?min and decreases significantly its immunoreactivity. The extensive hydrolysis of the microwave-treated ??-lg and WPI with trypsin, chymotrypsin, and the mixture of trypsin with chymotrypsin did not have an impact on the IgE binding of the products obtained in all the studied conditions.