SEARCH · Search PubMed
Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Prior arts in handwritten word recognition model either discrete features or continuous features, but not both. This paper combines discrete symbols and continuous attributes into structural handwriting features and model, them by transition-emitting and state-emitting hidden Markov models. The models are rigorously defined and experiments have proven their effectiveness.
A novel algorithm for font recognition on a single unknown Chinese character, independent of the identity of the character, is proposed in this paper. We employ a wavelet transform on the character image and extract wavelet features from the transformed image. After a Box-Cox transformation and LDA (Linear Discriminant Analysis) process, the discriminating features for font recognition are extracted and classified through a MQDF (Modified Quadric Distance Function) classifier with only one prototype for each font class. Our experiments show that our algorithm can achieve a recognition rate of 90.28 percent on a single unknown character and 99.01 percent if five characters are used for font recognition. Compared with existing methods, all of which are based on a text block, our method can provide a higher recognition rate and is more flexible and robust, since it is based on a single unknown character. Additionally, our method demonstrates that it is possible to extract subtle yet discriminative signals embedded in a much larger noisy background.
This paper presents the interpretation of digits and commands using a modified neural network and the genetic algorithm. The modified neural network exhibits a node-to-node relationship which enhances its learning and generalization abilities. A digit-and-command interpreter constructed by the modified neural networks is proposed to recognize handwritten digits and commands. A genetic algorithm is employed to train the parameters of the modified neural networks of the digit-and-command interpreter. The proposed digit-and-command interpreter is successfully realized in an electronic book. Simulation and experimental results will be presented to show the applicability and merits of the proposed approach.
This paper aims at automatic understanding of online handwritten mathematical expressions (MEs) written on an electronic tablet. The proposed technique involves two major stages: symbol recognition and structural analysis. Combination of two different classifiers have been used to achieve high accuracy for the recognition of symbols. Several online and offline features are used in the structural analysis phase to identify the spatial relationships among symbols. A context-free grammar has been designed to convert the input expressions into their corresponding T(E)X strings which are subsequently converted into MathML format. Contextual information has been used to correct several structure interpretation errors. A new method for evaluating performance of the proposed system has been formulated. Experiments on a dataset of considerable size strongly support the feasibility of the proposed system.
A common task in computer graphics is the mapping of digital high dynamic range images to low dynamic range display devices such as monitors and printers. This task is similar to the adaptation processes which occur in the human visual system. Physiological evidence suggests that adaptation already occurs in the photoreceptors, leading to a straightforward model that can be easily adapted for tone reproduction. The result is a fast and practical algorithm for general use with intuitive user parameters that control intensity, contrast, and level of chromatic adaptation, respectively.
Categorical data dimensions appear in many real-world data sets, but few visualization methods exist that properly deal with them. Parallel Sets are a new method for the visualization and interactive exploration of categorical data that shows data frequencies instead of the individual data points. The method is based on the axis layout of parallel coordinates, with boxes representing the categories and parallelograms between the axes showing the relations between categories. In addition to the visual representation, we designed a rich set of interactions. Parallel Sets allow the user to interactively remap the data to new categorizations and, thus, to consider more data dimensions during exploration and analysis than usually possible. At the same time, a metalevel, semantic representation of the data is built. Common procedures, like building the cross product of two or more dimensions, can be performed automatically, thus complementing the interactive visualization. We demonstrate Parallel Sets by analyzing a large CRM data set, as well as investigating housing data from two US states.
We present a framework for the holographic representation and display of graphics objects. As opposed to traditional graphics representations, our approach reconstructs the light wave reflected or emitted by the original object directly from the underlying digital hologram. Our novel holographic graphics pipeline consists of several stages including the digital recording of a full-parallax hologram, the reconstruction and propagation of its wavefront, and rendering of the final image onto conventional, framebuffer-based displays. The required view-dependent depth image is computed from the phase information inherently represented in the complex-valued wavefront. Our model also comprises a correct physical modeling of the camera taking into account optical elements, such as lens and aperture. It thus allows for a variety of effects including depth of field, diffraction, interference, and features built-in anti-aliasing. A central feature of our framework is its seamless integration into conventional rendering and display technology which enables us to elegantly combine traditional 3D object or scene representations with holograms. The presented work includes the theoretical foundations and allows for high quality rendering of objects consisting of large numbers of elementary waves while keeping the hologram at a reasonable size.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
A teaching-oriented multimedia database authoring system, MILES ( Multimediales Informations- und Lehrsystem), has been in development since 1980 in our department. The hardware consists of a network of personal computers connected to digital and, until recently, audio/video storage devices. The system provides a database capable of handling all kinds of multimedia data and computer programs. User-friendly software provides input, editing, retrieval, and communication; the authoring system allows these components to be organized into structures of complex menus, combined with free database access. More than 12,000 components have been stored, including approximately 3,500 pictures. The paper reports on an extensive field test, in which the system has been applied as a common source for all kinds of materials used in teaching physiology to students of medicine. ResuIts show that the "new media" are powerful instruments for improving teaching and learning. However, they should not be expected to provide the sole basis for education. Their application still faces many problems regarding concepts, efficiency, and acceptance by students and staff.
Explore the source record for details and available documents.
We have developed a data-collection system that we believe has wide applicability in dispersed organizations that have different local information systems infrastructures. A questionnaire was developed using commonly available software tools and was implemented several times across a dispersed network of individuals to rapidly collect, organize, and analyze information with a minimum of secretarial time and administrative cost. A common e-mail database was created identifying all of the medical directors and care management directors in the Blue Cross Blue Shield system. Initial polling of these individuals yielded specific questions of interest, and a final questionnaire was developed. The focus of the project centered on the evaluation of disease state management initiatives within each of the plans and on the ongoing use and future potential for various medical management initiatives. A questionnaire was developed using a Microsoft Excel spreadsheet with the ultimate development of a database in mind. All of the questions, whether single or multiple answers, were mapped from question response sections to a hidden specified range created to import the flat file answer block into a database. Individual cells containing answers to questions were each mapped to a hidden area of the spreadsheet arrayed as a series of rows. As each questionnaire was returned to the central site, data was imported from the hidden range name into a Microsoft Access database. The process of collecting extensive information on a questionnaire-by-questionnaire basis took approximately 20 seconds of time per questionnaire. A final report was ultimately created composed of some 24 pages of detailed information on managed care across the participants, representing some 90% of the member organizations. Secretarial costs were needed only for final transcription of the report.
BACKGROUND: Microarray experiments, as well as other genomic analyses, often result in large gene sets containing up to several hundred genes. The biological significance of such sets of genes is, usually, not readily apparent. Identification of the functions of the genes in the set can help highlight features of interest. The Gene Ontology Consortium 1 has annotated genes in several model organisms using a controlled vocabulary of terms and placed the terms on a Gene Ontology (GO), which comprises three disjoint hierarchies for Molecular functions, Biological processes and Cellular locations. The annotations can be used to identify functions that are enriched in the set, but this analysis can be misleading since the underlying distribution of genes among various functions is not uniform. For example, a large number of genes in a set might be kinases just because the genome contains many kinases. RESULTS: We use the Gene Ontology hierarchy and the annotations to pick significant functions and pathways by comparing the distribution of functions in a given gene list against the distribution of all the genes in the genome, using the hypergeometric distribution to assign probabilities. GObar is a web-based visualizer that implements this algorithm. The public website for GObar 2 can analyse gene lists from the yeast (S. cervisiae), fly (D. Melanogaster), mouse (M. musculus) and human (H. sapiens) genomes. It also allows visualization of the GO tree, as well as placement of a single gene on the GO hierarchy. We analyse a gene list from a genomic study of pre-mRNA splicing to demonstrate the utility of GObar. CONCLUSION: GObar is freely available as a web-based tool at http://katahdin.cshl.org:9331/GO2 and can help analyze and visualize gene lists from genomic analyses.
BACKGROUND: BLAST is one of the most common and useful tools for Genetic Research. This paper describes a software application we have termed Windows .NET Distributed Basic Local Alignment Search Toolkit (W.ND-BLAST), which enhances the BLAST utility by improving usability, fault recovery, and scalability in a Windows desktop environment. Our goal was to develop an easy to use, fault tolerant, high-throughput BLAST solution that incorporates a comprehensive BLAST result viewer with curation and annotation functionality. RESULTS: W.ND-BLAST is a comprehensive Windows-based software toolkit that targets researchers, including those with minimal computer skills, and provides the ability increase the performance of BLAST by distributing BLAST queries to any number of Windows based machines across local area networks (LAN). W.ND-BLAST provides intuitive Graphic User Interfaces (GUI) for BLAST database creation, BLAST execution, BLAST output evaluation and BLAST result exportation. This software also provides several layers of fault tolerance and fault recovery to prevent loss of data if nodes or master machines fail. This paper lays out the functionality of W.ND-BLAST. W.ND-BLAST displays close to 100% performance efficiency when distributing tasks to 12 remote computers of the same performance class. A high throughput BLAST job which took 662.68 minutes (11 hours) on one average machine was completed in 44.97 minutes when distributed to 17 nodes, which included lower performance class machines. Finally, there is a comprehensive high-throughput BLAST Output Viewer (BOV) and Annotation Engine components, which provides comprehensive exportation of BLAST hits to text files, annotated fasta files, tables, or association files. CONCLUSION: W.ND-BLAST provides an interactive tool that allows scientists to easily utilizing their available computing resources for high throughput and comprehensive sequence analyses. The install package for W.ND-BLAST is freely downloadable from http://liru.ars.usda.gov/mainbioinformatics.html. With registration the software is free, installation, networking, and usage instructions are provided as well as a support forum.
BACKGROUND: This study concerns the development of a high performance workflow that, using grid technology, correlates different kinds of Bioinformatics data, starting from the base pairs of the nucleotide sequence to the exposed residues of the protein surface. The implementation of this workflow is based on the Italian Grid.it project infrastructure, that is a network of several computational resources and storage facilities distributed at different grid sites. METHODS: Workflows are very common in Bioinformatics because they allow to process large quantities of data by delegating the management of resources to the information streaming. Grid technology optimizes the computational load during the different workflow steps, dividing the more expensive tasks into a set of small jobs. RESULTS: Grid technology allows efficient database management, a crucial problem for obtaining good results in Bioinformatics applications. The proposed workflow is implemented to integrate huge amounts of data and the results themselves must be stored into a relational database, which results as the added value to the global knowledge. CONCLUSION: A web interface has been developed to make this technology accessible to grid users. Once the workflow has started, by means of the simplified interface, it is possible to follow all the different steps throughout the data processing. Eventually, when the workflow has been terminated, the different features of the protein, like the amino acids exposed on the protein surface, can be compared with the data present in the output database.
BACKGROUND: Accurate methods for extraction of meaningful patterns in high dimensional data have become increasingly important with the recent generation of data types containing measurements across thousands of variables. Principal components analysis (PCA) is a linear dimensionality reduction (DR) method that is unsupervised in that it relies only on the data; projections are calculated in Euclidean or a similar linear space and do not use tuning parameters for optimizing the fit to the data. However, relationships within sets of nonlinear data types, such as biological networks or images, are frequently mis-rendered into a low dimensional space by linear methods. Nonlinear methods, in contrast, attempt to model important aspects of the underlying data structure, often requiring parameter(s) fitting to the data type of interest. In many cases, the optimal parameter values vary when different classification algorithms are applied on the same rendered subspace, making the results of such methods highly dependent upon the type of classifier implemented. RESULTS: We present the results of applying the spectral method of Lafon, a nonlinear DR method based on the weighted graph Laplacian, that minimizes the requirements for such parameter optimization for two biological data types. We demonstrate that it is successful in determining implicit ordering of brain slice image data and in classifying separate species in microarray data, as compared to two conventional linear methods and three nonlinear methods (one of which is an alternative spectral method). This spectral implementation is shown to provide more meaningful information, by preserving important relationships, than the methods of DR presented for comparison. Tuning parameter fitting is simple and is a general, rather than data type or experiment specific approach, for the two datasets analyzed here. Tuning parameter optimization is minimized in the DR step to each subsequent classification method, enabling the possibility of valid cross-experiment comparisons. CONCLUSION: Results from the spectral method presented here exhibit the desirable properties of preserving meaningful nonlinear relationships in lower dimensional space and requiring minimal parameter fitting, providing a useful algorithm for purposes of visualization and classification across diverse datasets, a common challenge in systems biology.
We consider the reproduction of color subject to material and neighborhood constraints. By "material constraints," we mean any constraints that are applied to the amount of ink, lights, voltages, and currents that are used in the generation of color. In the first instance we consider the problem of reproducing a target color constrained by maximum additive color signals, such as in the phosphorescence process in a cathode ray tube. In the second instance we consider the more difficult problem of reproducing color subject to constraints on the maximum primary color variations in a (spatial) neighborhood. We introduce the idea of adjacent color variance (ACV) and then attempt to reproduce colors subject to an upper bound on the ACV. An algorithm that is suitable for this task is the method of vector space projections (VSP). In order to use VSP for constrained color reproduction, we use a novel approach to linearize nonlinear CIE-Lab space constraints. Experimental results are furnished that demonstrate that using the ACV as a bound helps to reduce reproduction artifacts in a color image.
We determine a set of experimental parameters through the application of an evolutionary inversion procedure. The input to the algorithm is experimental apertureless scanning near-field optical microscope data. The performance of our inversion procedure is assessed by means of a comparison with a nonevolutionary technique.