Weighted geometric mean of large-scale matrices: numerical analysis and algorithms

Computing the weighted geometric mean of large sparse matrices is an operation that tends to become rapidly intractable, when the size of the matrices involved grows. However, if we are not interested in the computation of the matrix function itself, but just in that of its product times a vector, the problem turns simpler and there is a chance to solve it even when the matrix mean would actually be impossible to compute. Our interest is motivated by the fact that this calculation has some practical applications, related to the preconditioning of some operators arising in domain decomposition of elliptic problems. In this thesis, we explore how such a computation can be efficiently performed. First, we exploit the properties of the weighted geometric mean and find several equivalent ways to express it through real powers of a matrix. Hence, we focus our attention on matrix powers and examine how well-known techniques can be adapted to the solution of the problem at hand. In particular, we consider two broad families of approaches for the computation of f(A) v, namely quadrature formulae and Krylov subspace methods, and generalize them to the pencil case f(A\B) v. Finally, we provide an extensive experimental evaluation of the proposed algorithms and also try to assess how convergence speed and execution time are influenced by some characteristics of the input matrices. Our results suggest that a few elements have some bearing on the performance and that, although there is no best choice in general, knowing the conditioning and the sparsity of the arguments beforehand can considerably help in choosing the best strategy to tackle the problem.

Computer Aided Surgery for Percutaneous Nephrolithotomy: Clinical Requirement Analysis and System Design

Percutaneous nephrolithotomy (PCNL) for the treatment of renal stones and other related renal diseases has proved its efficacy and has stood the test of time compared with open surgical methods and extracorporal shock wave lithotripsy. However, access to the collecting system of the kidney is not easy because the available intra-operative image modalities only provide a two dimensional view of the surgical scenario. With this lack of visual information, several punctures are often necessary which, increases the risk of renal bleeding, splanchnic, vascular or pulmonary injury, or damage to the collecting system which sometimes makes the continuation of the procedure impossible. In order to address this problem, this paper proposes a workflow for introduction of a stereotactic needle guidance system for PCNL procedures. An analysis of the imposed clinical requirements, and a instrument guidance approach to provide the physician with a more intuitive planning and visual guidance to access the collecting system of the kidney are presented.

Virtopsy - the concept of a centralized database in forensic medicine for analysis and comparison of radiological and autopsy data

Recent developments in clinical radiology have resulted in additional developments in the field of forensic radiology. After implementation of cross-sectional radiology and optical surface documentation in forensic medicine, difficulties in the validation and analysis of the acquired data was experienced. To address this problem and for the comparison of autopsy and radiological data a centralized database with internet technology for forensic cases was created. The main goals of the database are (1) creation of a digital and standardized documentation tool for forensic-radiological and pathological findings; (2) establishing a basis for validation of forensic cross-sectional radiology as a non-invasive examination method in forensic medicine that means comparing and evaluating the radiological and autopsy data and analyzing the accuracy of such data; and (3) providing a conduit for continuing research and education in forensic medicine. Considering the infrequent availability of CT or MRI for forensic institutions and the heterogeneous nature of case material in forensic medicine an evaluation of benefits and limitations of cross-sectional imaging concerning certain forensic features by a single institution may be of limited value. A centralized database permitting international forensic and cross disciplinary collaborations may provide important support for forensic-radiological casework and research.

EEG reactivity and EEG activity in never-treated acute schizophrenics, measured with spectral parameters and dimensional complexity

Our approaches to the use of EEG studies for the understanding of the pathogenesis of schizophrenic symptoms are presented. The basic assumptions of a heuristic and multifactorial model of the psychobiological brain mechanisms underlying the organization of normal behavior is described and used in order to formulate and test hypotheses about the pathogenesis of schizophrenic behavior using EEG measures. Results from our studies on EEG activity and EEG reactivity (= EEG components of a memory-driven, adaptive, non-unitary orienting response) as analyzed with spectral parameters and "chaotic" dimensionality (correlation dimension) are summarized. Both analysis procedures showed a deviant brain functional organization in never-treated first-episode schizophrenia which, within the framework of the model, suggests as common denominator for the pathogenesis of the symptoms a deviation of working memory, the nature of which is functional and not structural.

Vibro-acoustical analysis and design of a multiple-layer constrained viscoelastic damping structure

The goal of this research is to provide a framework for vibro-acoustical analysis and design of a multiple-layer constrained damping structure. The existing research on damping and viscoelastic damping mechanism is limited to the following four mainstream approaches: modeling techniques of damping treatments/materials; control through the electrical-mechanical effect using the piezoelectric layer; optimization by adjusting the parameters of the structure to meet the design requirements; and identification of the damping material’s properties through the response of the structure. This research proposes a systematic design methodology for the multiple-layer constrained damping beam giving consideration to vibro-acoustics. A modeling technique to study the vibro-acoustics of multiple-layered viscoelastic laminated beams using the Biot damping model is presented using a hybrid numerical model. The boundary element method (BEM) is used to model the acoustical cavity whereas the Finite Element Method (FEM) is the basis for vibration analysis of the multiple-layered beam structure. Through the proposed procedure, the analysis can easily be extended to other complex geometry with arbitrary boundary conditions. The nonlinear behavior of viscoelastic damping materials is represented by the Biot damping model taking into account the effects of frequency, temperature and different damping materials for individual layers. A curve-fitting procedure used to obtain the Biot constants for different damping materials for each temperature is explained. The results from structural vibration analysis for selected beams agree with published closed-form results and results for the radiated noise for a sample beam structure obtained using a commercial BEM software is compared with the acoustical results of the same beam with using the Biot damping model. The extension of the Biot damping model is demonstrated to study MDOF (Multiple Degrees of Freedom) dynamics equations of a discrete system in order to introduce different types of viscoelastic damping materials. The mechanical properties of viscoelastic damping materials such as shear modulus and loss factor change with respect to different ambient temperatures and frequencies. The application of multiple-layer treatment increases the damping characteristic of the structure significantly and thus helps to attenuate the vibration and noise for a broad range of frequency and temperature. The main contributions of this dissertation include the following three major tasks: 1) Study of the viscoelastic damping mechanism and the dynamics equation of a multilayer damped system incorporating the Biot damping model. 2) Building the Finite Element Method (FEM) model of the multiple-layer constrained viscoelastic damping beam and conducting the vibration analysis. 3) Extending the vibration problem to the Boundary Element Method (BEM) based acoustical problem and comparing the results with commercial simulation software.

ANALYSIS AND VISUALIZATION OF FLOW FIELDS USING INFORMATION-THEORETIC TECHNIQUES AND GRAPH-BASED REPRESENTATIONS

Three-dimensional flow visualization plays an essential role in many areas of science and engineering, such as aero- and hydro-dynamical systems which dominate various physical and natural phenomena. For popular methods such as the streamline visualization to be effective, they should capture the underlying flow features while facilitating user observation and understanding of the flow field in a clear manner. My research mainly focuses on the analysis and visualization of flow fields using various techniques, e.g. information-theoretic techniques and graph-based representations. Since the streamline visualization is a popular technique in flow field visualization, how to select good streamlines to capture flow patterns and how to pick good viewpoints to observe flow fields become critical. We treat streamline selection and viewpoint selection as symmetric problems and solve them simultaneously using the dual information channel [81]. To the best of my knowledge, this is the first attempt in flow visualization to combine these two selection problems in a unified approach. This work selects streamline in a view-independent manner and the selected streamlines will not change for all viewpoints. My another work [56] uses an information-theoretic approach to evaluate the importance of each streamline under various sample viewpoints and presents a solution for view-dependent streamline selection that guarantees coherent streamline update when the view changes gradually. When projecting 3D streamlines to 2D images for viewing, occlusion and clutter become inevitable. To address this challenge, we design FlowGraph [57, 58], a novel compound graph representation that organizes field line clusters and spatiotemporal regions hierarchically for occlusion-free and controllable visual exploration. We enable observation and exploration of the relationships among field line clusters, spatiotemporal regions and their interconnection in the transformed space. Most viewpoint selection methods only consider the external viewpoints outside of the flow field. This will not convey a clear observation when the flow field is clutter on the boundary side. Therefore, we propose a new way to explore flow fields by selecting several internal viewpoints around the flow features inside of the flow field and then generating a B-Spline curve path traversing these viewpoints to provide users with closeup views of the flow field for detailed observation of hidden or occluded internal flow features [54]. This work is also extended to deal with unsteady flow fields. Besides flow field visualization, some other topics relevant to visualization also attract my attention. In iGraph [31], we leverage a distributed system along with a tiled display wall to provide users with high-resolution visual analytics of big image and text collections in real time. Developing pedagogical visualization tools forms my other research focus. Since most cryptography algorithms use sophisticated mathematics, it is difficult for beginners to understand both what the algorithm does and how the algorithm does that. Therefore, we develop a set of visualization tools to provide users with an intuitive way to learn and understand these algorithms.

Youth criminality and urban social conflict in the city of Rosario, Argentina. Analysis and proposals for conflict transformation

The present article describes and analyses youth criminality in the city of Rosario, Argentina between the years 2003-2006. Key actors’ understandings of and responses to the conflict were investigated by means of semi-structured interviews, observations, discourse analysis of policy documents, analysis of secondary data, and draw heavily on the experience of the author, a citizen and youth worker of Rosario. The actors examined were the police, the local government, young delinquents and youth organisations. Youth criminality is analysed from a conflict transformation approach using conflict analysis tools. Whereas, the provincial police understand the issue as a delinquency problem, other actors perceive it as an expression of a wider urban social conflict between those that are “included” and those that are “excluded” and as one of the negative effects of globalisation processes. The results suggest that police responses addressing only direct violence are ineffective, even contributing to increased tensions and polarisation, whereas strategies addressing cultural and structural violence are more suitable for this type of social urban conflict. Finally, recommendations for local youth policy are proposed to facilitate participation and inclusion of youth and as a tool for peaceful conflict transformation.

Cancer, meta-analysis and reporting biases: the case of erythropoiesis-stimulating agents.

Reporting and publication bias is a well-known problem in meta-analysis and healthcare research. In 2002 we conducted a meta-analysis on the effects of erythropoiesis-stimulating agents (ESAs) on overall survival in cancer patients, which suggested some evidence for improved survival in patients receiving ESAs compared with controls. However, a meta-analysis of individual patient data conducted several years later showed the opposite of our first meta-analysis, that is, evidence for increased on-study mortality and reduced overall survival in cancer patients receiving ESAs. We aimed to determine whether the results of our first meta-analysis could have been affected by publication and reporting biases and, if so, whether timely access to clinical study reports and individual patient data could have prevented this. We conducted a hypothetical meta-analysis for overall survival including all studies and study data that could have been available in 2002, at the time when we conducted our first meta-analysis. Compared with our original meta-analysis, which suggested an overall survival benefit for cancer patients receiving ESAs [hazard ratio (HR) 0.81, 95% confidence interval (CI) 0.67‒0.99], our hypothetical meta-analysis based on the results of all studies conducted at the time of the first analysis did not show evidence for a beneficial effect of ESAs on overall survival (HR 0.97, 95% CI 0.83‒1.12). Thus we have to conclude that our first meta-analysis showed misleading overall survival benefits due to publication and reporting biases, which could have been prevented by timely access to clinical study reports and individual patient data. Unrestricted access to clinical study protocols including amendments, clinical study reports and individual patient data is needed to ensure timely detection of both beneficial and harmful effects of healthcare interventions.

Growth Codes: Intermediate Performance Analysis and Application to Video

Growth codes are a subclass of Rateless codes that have found interesting applications in data dissemination problems. Compared to other Rateless and conventional channel codes, Growth codes show improved intermediate performance which is particularly useful in applications where partial data presents some utility. In this paper, we investigate the asymptotic performance of Growth codes using the Wormald method, which was proposed for studying the Peeling Decoder of LDPC and LDGM codes. Compared to previous works, the Wormald differential equations are set on nodes' perspective which enables a numerical solution to the computation of the expected asymptotic decoding performance of Growth codes. Our framework is appropriate for any class of Rateless codes that does not include a precoding step. We further study the performance of Growth codes with moderate and large size codeblocks through simulations and we use the generalized logistic function to model the decoding probability. We then exploit the decoding probability model in an illustrative application of Growth codes to error resilient video transmission. The video transmission problem is cast as a joint source and channel rate allocation problem that is shown to be convex with respect to the channel rate. This illustrative application permits to highlight the main advantage of Growth codes, namely improved performance in the intermediate loss region.

Results of bulk sediment X-ray diffraction analysis and quantification of mineral phases based on the RockJock and on the QUAX quantitative analysis

We have performed quantitative X-ray diffraction (qXRD) analysis of 157 grab or core-top samples from the western Nordic Seas between (WNS) ~57°-75°N and 5° to 45° W. The RockJock Vs6 analysis includes non-clay (20) and clay (10) mineral species in the <2 mm size fraction that sum to 100 weight %. The data matrix was reduced to 9 and 6 variables respectively by excluding minerals with low weight% and by grouping into larger groups, such as the alkali and plagioclase feldspars. Because of its potential dual origins calcite was placed outside of the sum. We initially hypothesized that a combination of regional bedrock outcrops and transport associated with drift-ice, meltwater plumes, and bottom currents would result in 6 clusters defined by "similar" mineral compositions. The hypothesis was tested by use of a fuzzy k-mean clustering algorithm and key minerals were identified by step-wise Discriminant Function Analysis. Key minerals in defining the clusters include quartz, pyroxene, muscovite, and amphibole. With 5 clusters, 87.5% of the observations are correctly classified. The geographic distributions of the five k-mean clusters compares reasonably well with the original hypothesis. The close spatial relationship between bedrock geology and discrete cluster membership stresses the importance of this variable at both the WNS-scale and at a more local scale in NE Greenland.

Development of Efficient Techniques for the Analysis and Design of Antennas in Dual-Reflectarray Configuration

This thesis contributes to the analysis and design of printed reflectarray antennas. The main part of the work is focused on the analysis of dual offset antennas comprising two reflectarray surfaces, one of them acts as sub-reflector and the second one acts as mainreflector. These configurations introduce additional complexity in several aspects respect to conventional dual offset reflectors, however they present a lot of degrees of freedom that can be used to improve the electrical performance of the antenna. The thesis is organized in four parts: the development of an analysis technique for dualreflectarray antennas, a preliminary validation of such methodology using equivalent reflector systems as reference antennas, a more rigorous validation of the software tool by manufacturing and testing a dual-reflectarray antenna demonstrator and the practical design of dual-reflectarray systems for some applications that show the potential of these kind of configurations to scan the beam and to generate contoured beams. In the first part, a general tool has been implemented to analyze high gain antennas which are constructed of two flat reflectarray structures. The classic reflectarray analysis based on MoM under local periodicity assumption is used for both sub and main reflectarrays, taking into account the incident angle on each reflectarray element. The incident field on the main reflectarray is computed taking into account the field radiated by all the elements on the sub-reflectarray.. Two approaches have been developed, one which employs a simple approximation to reduce the computer run time, and the other which does not, but offers in many cases, improved accuracy. The approximation is based on computing the reflected field on each element on the main reflectarray only once for all the fields radiated by the sub-reflectarray elements, assuming that the response will be the same because the only difference is a small variation on the angle of incidence. This approximation is very accurate when the reflectarray elements on the main reflectarray show a relatively small sensitivity to the angle of incidence. An extension of the analysis technique has been implemented to study dual-reflectarray antennas comprising a main reflectarray printed on a parabolic surface, or in general in a curved surface. In many applications of dual-reflectarray configurations, the reflectarray elements are in the near field of the feed-horn. To consider the near field radiated by the horn, the incident field on each reflectarray element is computed using a spherical mode expansion. In this region, the angles of incidence are moderately wide, and they are considered in the analysis of the reflectarray to better calculate the actual incident field on the sub-reflectarray elements. This technique increases the accuracy for the prediction of co- and cross-polar patterns and antenna gain respect to the case of using ideal feed models. In the second part, as a preliminary validation, the proposed analysis method has been used to design a dual-reflectarray antenna that emulates previous dual-reflector antennas in Ku and W-bands including a reflectarray as subreflector. The results for the dualreflectarray antenna compare very well with those of the parabolic reflector and reflectarray subreflector; radiation patterns, antenna gain and efficiency are practically the same when the main parabolic reflector is substituted by a flat reflectarray. The results show that the gain is only reduced by a few tenths of a dB as a result of the ohmic losses in the reflectarray. The phase adjustment on two surfaces provided by the dual-reflectarray configuration can be used to improve the antenna performance in some applications requiring multiple beams, beam scanning or shaped beams. Third, a very challenging dual-reflectarray antenna demonstrator has been designed, manufactured and tested for a more rigorous validation of the analysis technique presented. The proposed antenna configuration has the feed, the sub-reflectarray and the main-reflectarray in the near field one to each other, so that the conventional far field approximations are not suitable for the analysis of such antenna. This geometry is used as benchmarking for the proposed analysis tool in very stringent conditions. Some aspects of the proposed analysis technique that allow improving the accuracy of the analysis are also discussed. These improvements include a novel method to reduce the inherent cross polarization which is introduced mainly from grounded patch arrays. It has been checked that cross polarization in offset reflectarrays can be significantly reduced by properly adjusting the patch dimensions in the reflectarray in order to produce an overall cancellation of the cross-polarization. The dimensions of the patches are adjusted in order not only to provide the required phase-distribution to shape the beam, but also to exploit the crosses by zero of the cross-polarization components. The last part of the thesis deals with direct applications of the technique described. The technique presented is directly applicable to the design of contoured beam antennas for DBS applications, where the requirements of cross-polarisation are very stringent. The beam shaping is achieved by synthesithing the phase distribution on the main reflectarray while the sub-reflectarray emulates an equivalent hyperbolic subreflector. Dual-reflectarray antennas present also the ability to scan the beam over small angles about boresight. Two possible architectures for a Ku-band antenna are also described based on a dual planar reflectarray configuration that provides electronic beam scanning in a limited angular range. In the first architecture, the beam scanning is achieved by introducing a phase-control in the elements of the sub-reflectarray and the mainreflectarray is passive. A second alternative is also studied, in which the beam scanning is produced using 1-bit control on the main reflectarray, while a passive subreflectarray is designed to provide a large focal distance within a compact configuration. The system aims to develop a solution for bi-directional satellite links for emergency communications. In both proposed architectures, the objective is to provide a compact optics and simplicity to be folded and deployed.

Fault Tolerance Analysis and Self-Healing Strategy of Autonomous, Evolvable Hardware Systems

This paper presents an analysis of the fault tolerance achieved by an autonomous, fully embedded evolvable hardware system, which uses a combination of partial dynamic reconfiguration and an evolutionary algorithm (EA). It demonstrates that the system may self-recover from both transient and cumulative permanent faults. This self-adaptive system, based on a 2D array of 16 (4×4) Processing Elements (PEs), is tested with an image filtering application. Results show that it may properly recover from faults in up to 3 PEs, that is, more than 18% cumulative permanent faults. Two fault models are used for testing purposes, at PE and CLB levels. Two self-healing strategies are also introduced, depending on whether fault diagnosis is available or not. They are based on scrubbing, fitness evaluation, dynamic partial reconfiguration and in-system evolutionary adaptation. Since most of these adaptability features are already available on the system for its normal operation, resource cost for self-healing is very low (only some code additions in the internal microprocessor core)

Signal analysis and feature generation for pattern identification of partial discharges in high-voltage equipment

This paper proposes a method for the identification of different partial discharges (PDs) sources through the analysis of a collection of PD signals acquired with a PD measurement system. This method, robust and sensitive enough to cope with noisy data and external interferences, combines the characterization of each signal from the collection, with a clustering procedure, the CLARA algorithm. Several features are proposed for the characterization of the signals, being the wavelet variances, the frequency estimated with the Prony method, and the energy, the most relevant for the performance of the clustering procedure. The result of the unsupervised classification is a set of clusters each containing those signals which are more similar to each other than to those in other clusters. The analysis of the classification results permits both the identification of different PD sources and the discrimination between original PD signals, reflections, noise and external interferences. The methods and graphical tools detailed in this paper have been coded and published as a contributed package of the R environment under a GNU/GPL license.

A generic framework for the analysis and specialization of logic programs

The relationship between abstract interpretation and partial deduction has received considerable attention and (partial) integrations have been proposed starting from both the partial deduction and abstract interpretation perspectives. In this work we present what we argüe is the first fully described generic algorithm for efñcient and precise integration of abstract interpretation and partial deduction. Taking as starting point state-of-the-art algorithms for context-sensitive, polyvariant abstract interpretation and (abstract) partial deduction, we present an algorithm which combines the best of both worlds. Key ingredients include the accurate success propagation inherent to abstract interpretation and the powerful program transformations achievable by partial deduction. In our algorithm, the calis which appear in the analysis graph are not analyzed w.r.t. the original definition of the procedure but w.r.t. specialized definitions of these procedures. Such specialized definitions are obtained by applying both unfolding and abstract executability. Our framework is parametric w.r.t. different control strategies and abstract domains. Different combinations of such parameters correspond to existing algorithms for program analysis and specialization. Simultaneously, our approach opens the door to the efñcient computation of strictly more precise results than those achievable by each of the individual techniques. The algorithm is now one of the key components of the CiaoPP analysis and specialization system.

Regularization for sparsity in statistical analysis and machine learning

Pragmatism is the leading motivation of regularization. We can understand regularization as a modification of the maximum-likelihood estimator so that a reasonable answer could be given in an unstable or ill-posed situation. To mention some typical examples, this happens when fitting parametric or non-parametric models with more parameters than data or when estimating large covariance matrices. Regularization is usually used, in addition, to improve the bias-variance tradeoff of an estimation. Then, the definition of regularization is quite general, and, although the introduction of a penalty is probably the most popular type, it is just one out of multiple forms of regularization. In this dissertation, we focus on the applications of regularization for obtaining sparse or parsimonious representations, where only a subset of the inputs is used. A particular form of regularization, L1-regularization, plays a key role for reaching sparsity. Most of the contributions presented here revolve around L1-regularization, although other forms of regularization are explored (also pursuing sparsity in some sense). In addition to present a compact review of L1-regularization and its applications in statistical and machine learning, we devise methodology for regression, supervised classification and structure induction of graphical models. Within the regression paradigm, we focus on kernel smoothing learning, proposing techniques for kernel design that are suitable for high dimensional settings and sparse regression functions. We also present an application of regularized regression techniques for modeling the response of biological neurons. Supervised classification advances deal, on the one hand, with the application of regularization for obtaining a na¨ıve Bayes classifier and, on the other hand, with a novel algorithm for brain-computer interface design that uses group regularization in an efficient manner. Finally, we present a heuristic for inducing structures of Gaussian Bayesian networks using L1-regularization as a filter. El pragmatismo es la principal motivación de la regularización. Podemos entender la regularización como una modificación del estimador de máxima verosimilitud, de tal manera que se pueda dar una respuesta cuando la configuración del problema es inestable. A modo de ejemplo, podemos mencionar el ajuste de modelos paramétricos o no paramétricos cuando hay más parámetros que casos en el conjunto de datos, o la estimación de grandes matrices de covarianzas. Se suele recurrir a la regularización, además, para mejorar el compromiso sesgo-varianza en una estimación. Por tanto, la definición de regularización es muy general y, aunque la introducción de una función de penalización es probablemente el método más popular, éste es sólo uno de entre varias posibilidades. En esta tesis se ha trabajado en aplicaciones de regularización para obtener representaciones dispersas, donde sólo se usa un subconjunto de las entradas. En particular, la regularización L1 juega un papel clave en la búsqueda de dicha dispersión. La mayor parte de las contribuciones presentadas en la tesis giran alrededor de la regularización L1, aunque también se exploran otras formas de regularización (que igualmente persiguen un modelo disperso). Además de presentar una revisión de la regularización L1 y sus aplicaciones en estadística y aprendizaje de máquina, se ha desarrollado metodología para regresión, clasificación supervisada y aprendizaje de estructura en modelos gráficos. Dentro de la regresión, se ha trabajado principalmente en métodos de regresión local, proponiendo técnicas de diseño del kernel que sean adecuadas a configuraciones de alta dimensionalidad y funciones de regresión dispersas. También se presenta una aplicación de las técnicas de regresión regularizada para modelar la respuesta de neuronas reales. Los avances en clasificación supervisada tratan, por una parte, con el uso de regularización para obtener un clasificador naive Bayes y, por otra parte, con el desarrollo de un algoritmo que usa regularización por grupos de una manera eficiente y que se ha aplicado al diseño de interfaces cerebromáquina. Finalmente, se presenta una heurística para inducir la estructura de redes Bayesianas Gaussianas usando regularización L1 a modo de filtro.