Search

BACKGROUND The average accepted depth for Non-Tunneled Catheters (NTC) insertion does not guarantee its correct position, so controversy exists. The aim is to assess the effect of two NTC placement depths on the number of NTC complication episodes. METHODS We designed a triple blind, parallel-group, randomized controlled trial in a single Hemodialysis Center in Mexico (Registry: ACTRN12619000774123). We included patients in urgent need of hemodialysis via internal right jugular vein NTC. The length of the NTC tip placement depth was randomized to second intercostal space (2ICS) or fourth intercostal space (4ICS), using physical landmarks. The primary outcome is to compare the composite number of NTC dysfunction, repositioning, and relocation episodes during 48 hours post procedure. RESULTS One hundred and sixty five patients were included, 86 and 79 patients to NTC placement in the 2ICS and 4ICS, respectively. All patients underwent intention-to treat analysis. The incidence of the composite outcome was lower in the 2ICS group compared to the 4ICS group, 4 (4.6%) and 50 (63%) combined episodes, respectively (p

From simple averaging to more sophisticated registration and restoration strategies, such as super-resolution (SR), there exist different computational techniques that use a series of images of the same object to generate enhanced images where noise and other distortions have been reduced. In this work, we provide qualitative and quantitative measurements of this enhancement for high-angle annular dark-field scanning transmission electron microscopy imaging. These images are compared in two ways, qualitatively through visual inspection in real and reciprocal space, and quantitatively, through the calculation of objective measurements, such as signal-to-noise ratio and atom column roundness. Results show that these techniques improve the quality of the images. In this paper, we use an SR methodology that allows us to take advantage of the information present in the image frames and to reliably facilitate the analysis of more difficult regions of interest in experimental images, such as surfaces and interfaces. By acquiring a series of cross-sectional experimental images of magnetite (Fe3O4) thin films (111), we have generated interpolated images using averaging and SR, and reconstructed the atomic structure of the very top surface layer that consists of a full monolayer of Fe, with topmost Fe atoms in tetrahedrally coordinated sites.

International audience

The influence of the neighbouring atomic-columns in determining the composition at atomic column scale of quaternary semiconductor compounds, using simulated HAADF-STEM images is evaluated. The InAlAsSb alloy, a promising material in the photovoltaic field, is considered. We find that the so called 'crosstalk' effect plays an important role for the aimed compositional determination. The intensity transfer is larger from neighbouring atomic columns with higher average Z, and towards atomic columns with smaller Z. Our results show that in order to obtain precise information on the column composition, the HAADF-STEM intensities of both columns need to be taken into account simultaneously.

During image acquisition of crystalline materials by high-resolution scanning transmission electron microscopy, the sample drift could lead to distortions and shears that hinder their quantitative analysis and characterization. In order to measure and correct this effect, several authors have proposed different methodologies making use of series of images. In this work, we introduce a methodology to determine the drift angle via Fourier analysis by using a single image based on the measurements between the angles of the second Fourier harmonics in different quadrants. Two different approaches, that are independent of the angle of acquisition of the image, are evaluated. In addition, our results demonstrate that the determination of the drift angle is more accurate by using the measurements of non-consecutive quadrants when the angle of acquisition is an odd multiple of 45°.

The Cuban family medicine model has been developed, for more than three decades, as the third model of medical care of the revolutionary period in Cuba. It is the complete expression of the application of the primary health care strategy. However, it is still possible to improve the organization, quality and efficiency of health services in the first level of health care. The objective of this article is to describe the model, its singularities, distinctive elements, main achievements, and challenges. A documentary review was made from journal articles, books and official documents of the Ministry of Public Health of Cuba on the creation and development of the model; documents were selected and analyzed by experts for the preparation of this report. Conceptual elements of the model are presented, as well as indicators supporting the information. The main characteristics of family medicine in Cuba, its fundamental achievements, and challenges are presented. Based on its results, Cuba's family medicine model is a reference for the Region of the Americas. It is still a model in development, however, perfectible and susceptible to transformations to face new challenges, more importantly to renew the model keeping and strengthening the foundational ideas of the program.

The growth of the Internet has fuelled the availability of e-commerce marketplaces and search engines must face with a huge amount of ambiguity and inconsistencies in the data. Product matching aims at disambiguating descriptions of products belonging to different websites in order to be able to recognize identical products and to merge the content from those identical items. In this work first we evaluate some similarity measures for string matching and then, we apply a complete product matching methodology to the retail market of used cars. We use a reference or master list of items and information about a wide variety of used cars offers. The resulting linkage allows energy efficiency assignment of the model identified.

High-quality image reconstruction techniques allow the generation of high pixel density images from a set of low-resolution micrographs. In general, these techniques consist of two main steps, namely, accurate registration, and formulation of an appropriate forward image model via some restoration method. There exist a wide variety of algorithms to cope with both stages and depending on their practical applications, some methods can outperform others, since they can be sensitive to the assumed data model, noise, drift, etc. When dealing with images generated by Z-contrast scanning transmission electron microscopes, a current trend is based on non-rigid approximations in the registration stage. In our work we aimed at reaching similar accuracy but addressing the most complex calculations in the reconstruction stage, instead of in the registration stage (as the non-rigid approaches do), but using a much smaller number of images. We review some of the most significant methods and address their shortcomings when they are applied to the field of microscopy. Simulated images with known targets will be used to evaluate and compare the main approaches in terms of quality enhancement and computing time. In addition, a procedure to determine the reference image will be proposed to minimise the global drift on the series. The best registration and restoration strategies will be applied to experimental images in order to point up the enhanced capability of this high quality image reconstruction methodology in this field.

Product matching aims at disambiguating descriptions of products belonging to different websites in order to be able to recognize identical elements and to merge the content from those identical items. Most approaches face this matter applying various machine learning methods to textual product descriptions. Recently some authors are including information extracted from an image associated to a textual description of a product. Modern machine learning methods, such as content based information retrieval (CBIR) or deep learning, can be applied to this type of images since they can manage very large data sets for finding hidden structure within them, and for making accurate predictions. This information could boost the performance of the traditional textual matching but at the same time increase the computational complexity of the process. In this paper we review some CBIR and deep learning models and analyse the performance of these approaches when they are applied to images for product matching. The results obtained will help to introduce a combined classifier using textual and image information.

Super-resolution (SR) software-based techniques aim at generating a final image by combining several noisy frames with lower resolution from the same scene. A comparative study on high-resolution high-angle annular dark field images of InAs/GaAs QDs has been carried out in order to evaluate the performance of the SR technique. The obtained SR images present enhanced resolution and higher signal-to-noise (SNR) ratio and sharpness regarding the experimental images. In addition, SR is also applied in the field of strain analysis using digital image processing applications such as geometrical phase analysis and peak pairs analysis. The precision of the strain mappings can be improved when SR methodologies are applied to experimental images.

This paper presents a new approach to model selection based on hypothesis testing. We 4rst describe a procedure to generate di5erent scores for any candidate model from a single sample of training data and then discuss how to apply multiple comparison procedures (MCP) to model selection. MCP statistical tests allow us to compare three or more groups of data while controlling the probability of making at least one Type I error. The complete procedure is illustrated on several model selection tasks, including the determination of the number of hidden units for feed-forward neural networks and the number of kernels for RBF networks. c 2002 Elsevier Science B.V. All rights reserved.

Complexity-penalization strategies are one way to decide on the most appropriate network size in order to address the trade-off between overfitted and underfitted models. In this paper we propose a new penalty term derived from the behaviour of candidate models under noisy conditions that seems to be much more robust against catastrophic overfitting errors that standard techniques. This strategy is applied to several regression problems using polynomial functions, univariate autoregressive models and RBF neural networks. The simulation study at the end of the paper will show that the proposed criterion is extremely competitive when compared to state-of-the-art criteria.

The combination of strain measurements1 and the analysis of High Angle Annular Dark Field (HAADF) Scanning Transmission Electron Microscopy (STEM) images constitutes a powerful approach to investigate strained heterostructures on the nanometer scale and has proved to be extremely powerful for characterizing nanomaterials2,3.

In this paper we describe a pattern recognition system implemented to determine thickness and defocus from HRTEM simulated images. A specific task has been designed to quantify the influence of certain operation parameters of a transmission electron microscope in the global recognition error rate. This influence allows us to estimate human recognition confidence when applying pattern matching to the determination of thickness and defocus from HRTEM maps. The images considered in this task correspond to InP with the sphalerite crystalline structure and were simulated using the EMS computer software package.

The goal of combining the outputs of multiple models is to form an improved meta-model with higher generalization capability than the best single model used in isolation. Most popular ensemble methods do specify neither the number of component models nor their complexity. However, these parameters strongly influence the generalization capability of the meta-model. In this paper we propose an ensemble method which generates a meta-model with optimal values for these parameters. The proposed method suggests using resampling techniques to generate multiple estimations of the generalization error and multiple comparison procedures to select the models that will be combined to form the meta-model. Experimental results show the performance of the model on regression and classification tasks using artificial and real databases.

Estimating Prediction Risk is important for providing a way of computing the expected error for predictions made by a model, but it is also an important tool for model selection. This paper addresses an empirical comparison of model selection techniques based on the Prediction Risk estimation, with particular reference to the structure of nonlinear regularized neural networks. To measure the performance of the different model selection criteria a large-scale small-samples simulation is conducted for feedforward neural networks.

In this paper we describe a methodology developed at the University of Cadiz (Spain) in the past few years for the extraction of quantitative information from electron microscopy images at the atomic level. This work is based on a coordinated and synergic activity of several research groups that have been working together over the last decade in two different and complementary fields: Materials Science and Computer Science. The aim of our joint research has been to develop innovative high-performance computing techniques and simulation methods in order to address computationally challenging problems in the analysis, modelling and simulation of materials at the atomic scale, providing significant advances with respect to existing techniques. The methodology involves several fundamental areas of research including the analysis of high resolution electron microscopy images, materials modelling, image simulation and 3D reconstruction using quantitative information from experimental images. These techniques for the analysis, modelling and simulation allow optimizing the control and functionality of devices developed using materials under study, and have been tested using data obtained from experimental samples.

One of the main research concern in neural networks is to find the appropriate network size in order to minimize the trade-off between overfitting and poor approximation. In this paper the choice among different competing models that fit to the same data set is faced when statistical methods for model comparison are applied. The study has been conducted to find a range of models that can work all the same as the cost of complexity varies. If they do not, then the generalization error estimates should be about the same among the set of models. If they do, then the estimates should be different and our job would consist on analyzing pairwise differences between the least generalization error estimate and each one of the range, in order to bound the set of models which might result in an equal performance. This method is illustrated applied to polynomial regression and RBF neural networks.

High-Angle Annular Dark-Field Scanning Transmission Electron Microscopy (HAADF-STEM) in combination with strain mapping techniques provides a powerful tool for quantitative analysis of crystalline semiconductor materials. Due to the complex interaction of a focused probe and a sample in HAADF, the calculation of each pixel in a simulation process requires a complete multislice iteration, making the overall computing process a rather demanding task in time and memory. SICSTEM is a parallel software code recently developed for running on the University of Cadiz Supercomputer (3.75 Tflops) that allows the simulation of images from large nanostructures containing more than one million atoms. The software has been designed to be able to generate not only one dimensional line scans or two dimensional images, but also to perform optical sectioning in the STEM simulation process, providing an easy way to simulate 3D HAADF-STEM images. In this work we consider GaAs capped GaSb nanostructures epitaxially oriented on a GaAs substrate. A methodology has been developed by combining the through-focal series STEM imaging and image analysis to estimate shape and position of buried GaSb nanostructures.

We show in this work that GaAs capping of 2 ML of GaSb grown by molecular beam epitaxy results in the formation of very small (with heights of about 1 nm) GaAsxSb1−x nanostructures surrounded by a GaAs rich layer. This conclusion is obtained by analyzing the morphology of the GaAsxSb1−x nanostructures by high resolution scanning transmission electron microscopy in Z-contrast mode. This result shows that a significant fraction of the Sb atoms must segregate along the growth direction during the GaAs capping process., This work was supported by the Office of Basic Energy Sciences, Division of Materials Sciences and Engineering, U.S. DOE MFC, the SANDiE European Network of Excellence Contract No. NMP4-CT-2004-500101, the Spanish MEC Grant No. TEC2005-05781-C03-01 y 02, TEC2008-06756-C03-02/TEC, NAN2004-09109-C04-01, Consolider-Ingenio 2010 CSD2006-00019, the CAM Grant No. S 0505ESP 0200, and the Junta de Andalucía PAI research groups TEP-120 and TIC-145; projects PAI05-TEP-00383 and TEP-03516.

High angle annular dark field scanning transmission electron microscopy (HAADF-STEM) is a powerful tool to quantify size, shape, position, and composition of nano-objects with the assessment of image simulation. Due to the high computational requirements needed, nowadays it can only be applied to a few unit cells in standard computers. To overpass this limitation, a parallel software (SICSTEM) has been developed. This software can afford HAADF-STEM image simulations of nanostructures composed of several hundred thousand atoms in manageable time. The usefulness of this tool is exemplified by simulating a HAADF-STEM image of an InAs nanowire.