Iterative regularization methods for ill-posed problems

This Ph.D thesis focuses on iterative regularization methods for regularizing linear and nonlinear ill-posed problems. Regarding linear problems, three new stopping rules for the Conjugate Gradient method applied to the normal equations are proposed and tested in many numerical simulations, including some tomographic images reconstruction problems. Regarding nonlinear problems, convergence and convergence rate results are provided for a Newton-type method with a modified version of Landweber iteration as an inner iteration in a Banach space setting.

Coarse-graining methods and polymer dynamics

This thesis studies molecular dynamics simulations on two levels of resolution: the detailed level of atomistic simulations, where the motion of explicit atoms in a many-particle system is considered, and the coarse-grained level, where the motion of superatoms composed of up to 10 atoms is modeled. While atomistic models are capable of describing material specific effects on small scales, the time and length scales they can cover are limited due to their computational costs. Polymer systems are typically characterized by effects on a broad range of length and time scales. Therefore it is often impossible to atomistically simulate processes, which determine macroscopic properties in polymer systems. Coarse-grained (CG) simulations extend the range of accessible time and length scales by three to four orders of magnitude. However, no standardized coarse-graining procedure has been established yet. Following the ideas of structure-based coarse-graining, a coarse-grained model for polystyrene is presented. Structure-based methods parameterize CG models to reproduce static properties of atomistic melts such as radial distribution functions between superatoms or other probability distributions for coarse-grained degrees of freedom. Two enhancements of the coarse-graining methodology are suggested. Correlations between local degrees of freedom are implicitly taken into account by additional potentials acting between neighboring superatoms in the polymer chain. This improves the reproduction of local chain conformations and allows the study of different tacticities of polystyrene. It also gives better control of the chain stiffness, which agrees perfectly with the atomistic model, and leads to a reproduction of experimental results for overall chain dimensions, such as the characteristic ratio, for all different tacticities. The second new aspect is the computationally cheap development of nonbonded CG potentials based on the sampling of pairs of oligomers in vacuum. Static properties of polymer melts are obtained as predictions of the CG model in contrast to other structure-based CG models, which are iteratively refined to reproduce reference melt structures. The dynamics of simulations at the two levels of resolution are compared. The time scales of dynamical processes in atomistic and coarse-grained simulations can be connected by a time scaling factor, which depends on several specific system properties as molecular weight, density, temperature, and other components in mixtures. In this thesis the influence of molecular weight in systems of oligomers and the situation in two-component mixtures is studied. For a system of small additives in a melt of long polymer chains the temperature dependence of the additive diffusion is predicted and compared to experiments.

Rheology of arrested colloids : a parameter study using novel experimental methods

The steadily increasing diversity of colloidal systems demands for new theoretical approaches and a cautious experimental characterization. Here we present a combined rheological and microscopical study of colloids in their arrested state whereas we did not aim for a generalized treatise but rather focused on a few model colloids, liquid crystal based colloidal suspensions and sedimented colloidal films. We laid special emphasis on the understanding of the mutual influence of dominant interaction mechanisms, structural characteristics and the particle properties on the mechanical behavior of the colloid. The application of novel combinations of experimental techniques played an important role in these studies. Beside of piezo-rheometry we employed nanoindentation experiments and associated standardized analysis procedures. These rheometric methods were complemented by real space images using confocal microscopy. The flexibility of the home-made setup allowed for a combination of both techniques and thereby for a simultaneous rheological and three-dimensional structural analysis on a single particle level. Though, the limits of confocal microscopy are not reached by now. We show how hollow and optically anisotropic particles can be utilized to quantify contact forces and rotational motions for individual particles. In future such data can contribute to a better understanding of particle reorganization processes, such as the liquidation of colloidal gels and glasses under shear.

Numerical and Analytical Methods for Laser-Plasma Acceleration Physics

Theories and numerical modeling are fundamental tools for understanding, optimizing and designing present and future laser-plasma accelerators (LPAs). Laser evolution and plasma wave excitation in a LPA driven by a weakly relativistically intense, short-pulse laser propagating in a preformed parabolic plasma channel, is studied analytically in 3D including the effects of pulse steepening and energy depletion. At higher laser intensities, the process of electron self-injection in the nonlinear bubble wake regime is studied by means of fully self-consistent Particle-in-Cell simulations. Considering a non-evolving laser driver propagating with a prescribed velocity, the geometrical properties of the non-evolving bubble wake are studied. For a range of parameters of interest for laser plasma acceleration, The dependence of the threshold for self-injection in the non-evolving wake on laser intensity and wake velocity is characterized. Due to the nonlinear and complex nature of the Physics involved, computationally challenging numerical simulations are required to model laser-plasma accelerators operating at relativistic laser intensities. The numerical and computational optimizations, that combined in the codes INF&RNO and INF&RNO/quasi-static give the possibility to accurately model multi-GeV laser wakefield acceleration stages with present supercomputing architectures, are discussed. The PIC code jasmine, capable of efficiently running laser-plasma simulations on Graphics Processing Units (GPUs) clusters, is presented. GPUs deliver exceptional performance to PIC codes, but the core algorithms had to be redesigned for satisfying the constraints imposed by the intrinsic parallelism of the architecture. The simulation campaigns, run with the code jasmine for modeling the recent LPA experiments with the INFN-FLAME and CNR-ILIL laser systems, are also presented.

Learning Methods and Algorithms for Semantic Text Classification across Multiple Domains

Information is nowadays a key resource: machine learning and data mining techniques have been developed to extract high-level information from great amounts of data. As most data comes in form of unstructured text in natural languages, research on text mining is currently very active and dealing with practical problems. Among these, text categorization deals with the automatic organization of large quantities of documents in priorly defined taxonomies of topic categories, possibly arranged in large hierarchies. In commonly proposed machine learning approaches, classifiers are automatically trained from pre-labeled documents: they can perform very accurate classification, but often require a consistent training set and notable computational effort. Methods for cross-domain text categorization have been proposed, allowing to leverage a set of labeled documents of one domain to classify those of another one. Most methods use advanced statistical techniques, usually involving tuning of parameters. A first contribution presented here is a method based on nearest centroid classification, where profiles of categories are generated from the known domain and then iteratively adapted to the unknown one. Despite being conceptually simple and having easily tuned parameters, this method achieves state-of-the-art accuracy in most benchmark datasets with fast running times. A second, deeper contribution involves the design of a domain-independent model to distinguish the degree and type of relatedness between arbitrary documents and topics, inferred from the different types of semantic relationships between respective representative words, identified by specific search algorithms. The application of this model is tested on both flat and hierarchical text categorization, where it potentially allows the efficient addition of new categories during classification. Results show that classification accuracy still requires improvements, but models generated from one domain are shown to be effectively able to be reused in a different one.

IMEX finite volume methods for the shallow water equations

Die Flachwassergleichungen (SWE) sind ein hyperbolisches System von Bilanzgleichungen, die adäquate Approximationen an groß-skalige Strömungen der Ozeane, Flüsse und der Atmosphäre liefern. Dabei werden Masse und Impuls erhalten. Wir unterscheiden zwei charakteristische Geschwindigkeiten: die Advektionsgeschwindigkeit, d.h. die Geschwindigkeit des Massentransports, und die Geschwindigkeit von Schwerewellen, d.h. die Geschwindigkeit der Oberflächenwellen, die Energie und Impuls tragen. Die Froude-Zahl ist eine Kennzahl und ist durch das Verhältnis der Referenzadvektionsgeschwindigkeit zu der Referenzgeschwindigkeit der Schwerewellen gegeben. Für die oben genannten Anwendungen ist sie typischerweise sehr klein, z.B. 0.01. Zeit-explizite Finite-Volume-Verfahren werden am öftersten zur numerischen Berechnung hyperbolischer Bilanzgleichungen benutzt. Daher muss die CFL-Stabilitätsbedingung eingehalten werden und das Zeitinkrement ist ungefähr proportional zu der Froude-Zahl. Deswegen entsteht bei kleinen Froude-Zahlen, etwa kleiner als 0.2, ein hoher Rechenaufwand. Ferner sind die numerischen Lösungen dissipativ. Es ist allgemein bekannt, dass die Lösungen der SWE gegen die Lösungen der Seegleichungen/ Froude-Zahl Null SWE für Froude-Zahl gegen Null konvergieren, falls adäquate Bedingungen erfüllt sind. In diesem Grenzwertprozess ändern die Gleichungen ihren Typ von hyperbolisch zu hyperbolisch.-elliptisch. Ferner kann bei kleinen Froude-Zahlen die Konvergenzordnung sinken oder das numerische Verfahren zusammenbrechen. Insbesondere wurde bei zeit-expliziten Verfahren falsches asymptotisches Verhalten (bzgl. der Froude-Zahl) beobachtet, das diese Effekte verursachen könnte.Ozeanographische und atmosphärische Strömungen sind typischerweise kleine Störungen eines unterliegenden Equilibriumzustandes. Wir möchten, dass numerische Verfahren für Bilanzgleichungen gewisse Equilibriumzustände exakt erhalten, sonst können künstliche Strömungen vom Verfahren erzeugt werden. Daher ist die Quelltermapproximation essentiell. Numerische Verfahren die Equilibriumzustände erhalten heißen ausbalanciert.rnrnIn der vorliegenden Arbeit spalten wir die SWE in einen steifen, linearen und einen nicht-steifen Teil, um die starke Einschränkung der Zeitschritte durch die CFL-Bedingung zu umgehen. Der steife Teil wird implizit und der nicht-steife explizit approximiert. Dazu verwenden wir IMEX (implicit-explicit) Runge-Kutta und IMEX Mehrschritt-Zeitdiskretisierungen. Die Raumdiskretisierung erfolgt mittels der Finite-Volumen-Methode. Der steife Teil wird mit Hilfe von finiter Differenzen oder au eine acht mehrdimensional Art und Weise approximniert. Zur mehrdimensionalen Approximation verwenden wir approximative Evolutionsoperatoren, die alle unendlich viele Informationsausbreitungsrichtungen berücksichtigen. Die expliziten Terme werden mit gewöhnlichen numerischen Flüssen approximiert. Daher erhalten wir eine Stabilitätsbedingung analog zu einer rein advektiven Strömung, d.h. das Zeitinkrement vergrößert um den Faktor Kehrwert der Froude-Zahl. Die in dieser Arbeit hergeleiteten Verfahren sind asymptotisch erhaltend und ausbalanciert. Die asymptotischer Erhaltung stellt sicher, dass numerische Lösung das "korrekte" asymptotische Verhalten bezüglich kleiner Froude-Zahlen besitzt. Wir präsentieren Verfahren erster und zweiter Ordnung. Numerische Resultate bestätigen die Konvergenzordnung, so wie Stabilität, Ausbalanciertheit und die asymptotische Erhaltung. Insbesondere beobachten wir bei machen Verfahren, dass die Konvergenzordnung fast unabhängig von der Froude-Zahl ist.

Regularization methods for the solution of a nonlinear least-squares problem in tomography

In this work we study a polyenergetic and multimaterial model for the breast image reconstruction in Digital Tomosynthesis, taking into consideration the variety of the materials forming the object and the polyenergetic nature of the X-rays beam. The modelling of the problem leads to the resolution of a high-dimensional nonlinear least-squares problem that, due to its nature of inverse ill-posed problem, needs some kind of regularization. We test two main classes of methods: the Levenberg-Marquardt method (together with the Conjugate Gradient method for the computation of the descent direction) and two limited-memory BFGS-like methods (L-BFGS). We perform some experiments for different values of the regularization parameter (constant or varying at each iteration), tolerances and stop conditions. Finally, we analyse the performance of the several methods comparing relative errors, iterations number, times and the qualities of the reconstructed images.

A Hybrid Multimodal Non-Rigid Registration of MR Images Based on Diffeomorphic Demons

n this paper we present a novel hybrid approach for multimodal medical image registration based on diffeomorphic demons. Diffeomorphic demons have proven to be a robust and efficient way for intensity-based image registration. A very recent extension even allows to use mutual information (MI) as a similarity measure to registration multimodal images. However, due to the intensity correspondence uncertainty existing in some anatomical parts, it is difficult for a purely intensity-based algorithm to solve the registration problem. Therefore, we propose to combine the resulting transformations from both intensity-based and landmark-based methods for multimodal non-rigid registration based on diffeomorphic demons. Several experiments on different types of MR images were conducted, for which we show that a better anatomical correspondence between the images can be obtained using the hybrid approach than using either intensity information or landmarks alone.

Automated Intervertebral Disc Detection from Low Resolution, Sparse MRI Images for the Planning of Scan Geometries

Robust and accurate identification of intervertebral discs from low resolution, sparse MRI scans is essential for the automated scan planning of the MRI spine scan. This paper presents a graphical model based solution for the detection of both the positions and orientations of intervertebral discs from low resolution, sparse MRI scans. Compared with the existing graphical model based methods, the proposed method does not need a training process using training data and it also has the capability to automatically determine the number of vertebrae visible in the image. Experiments on 25 low resolution, sparse spine MRI data sets verified its performance.

Fully Automatic Segmentation of Brain Tumor Images using Support Vector Machine Classiffication in Combination with Hierarchical Conditional Random Field Regularization

Delineating brain tumor boundaries from magnetic resonance images is an essential task for the analysis of brain cancer. We propose a fully automatic method for brain tissue segmentation, which combines Support Vector Machine classification using multispectral intensities and textures with subsequent hierarchical regularization based on Conditional Random Fields. The CRF regularization introduces spatial constraints to the powerful SVM classification, which assumes voxels to be independent from their neighbors. The approach first separates healthy and tumor tissue before both regions are subclassified into cerebrospinal fluid, white matter, gray matter and necrotic, active, edema region respectively in a novel hierarchical way. The hierarchical approach adds robustness and speed by allowing to apply different levels of regularization at different stages. The method is fast and tailored to standard clinical acquisition protocols. It was assessed on 10 multispectral patient datasets with results outperforming previous methods in terms of segmentation detail and computation times.

Computational Study of the Hydration of Sulfuric Acid Dimers: Implications for Acid Dissociation and Aerosol Formation

We have investigated the thermodynamics of sulfuric acid dimer hydration using ab initio quantum mechanical methods. For (H2SO4)2(H2O)n where n = 0−6, we employed high-level ab initio calculations to locate the most stable minima for each cluster size. The results presented herein yield a detailed understanding of the first deprotonation of sulfuric acid as a function of temperature for a system consisting of two sulfuric acid molecules and up to six waters. At 0 K, a cluster of two sulfuric acid molecules and one water remains undissociated. Addition of a second water begins the deprotonation of the first sulfuric acid leading to the di-ionic species (the bisulfate anion HSO4−, the hydronium cation H3O+, an undissociated sulfuric acid molecule, and a water). Upon the addition of a third water molecule, the second sulfuric acid molecule begins to dissociate. For the (H2SO4)2(H2O)3 cluster, the di-ionic cluster is a few kcal mol−1 more stable than the neutral cluster, which is just slightly more stable than the tetra-ionic cluster (two bisulfate anions, two hydronium cations, and one water). With four water molecules, the tetra-ionic cluster, (HSO4−)2(H3O+)2(H2O)2, becomes as favorable as the di-ionic cluster H2SO4(HSO4−)(H3O+)(H2O)3 at 0 K. Increasing the temperature favors the undissociated clusters, and at room temperature we predict that the di-ionic species is slightly more favorable than the neutral cluster once three waters have been added to the cluster. The tetra-ionic species competes with the di-ionic species once five waters have been added to the cluster. The thermodynamics of stepwise hydration of sulfuric acid dimer is similar to that of the monomer; it is favorable up to n = 4−5 at 298 K. A much more thermodynamically favorable pathway forming sulfuric acid dimer hydrates is through the combination of sulfuric acid monomer hydrates, but the low concentration of sulfuric acid relative to water vapor at ambient conditions limits that process.

Au@Hg nanoalloy formation through direct amalgamation: Structural, spectroscopic, and computational evidence for slow nanoscale diffusion

Dynamic core-shell nanoparticles have received increasing attention in recent years. This paper presents a detailed study of Au-Hg nanoalloys, whose composing elements show a large difference in cohesive energy. A simple method to prepare Au@Hg particles with precise control over the composition up to 15 atom% mercury is introduced, based on reacting a citrate stabilized gold sol with elemental mercury. Transmission electron microscopy shows an increase of particle size with increasing mercury content and, together with X-ray powder diffraction, points towards the presence of a core-shell structure with a gold core surrounded by an Au-Hg solid solution layer. The amalgamation process is described by pseudo-zero-order reaction kinetics, which indicates slow dissolution of mercury in water as the rate determining step, followed by fast scavenging by nanoparticles in solution. Once adsorbed at the surface, slow diffusion of Hg into the particle lattice occurs, to a depth of ca. 3 nm, independent of Hg concentration. Discrete dipole approximation calculations relate the UV-vis spectra to the microscopic details of the nanoalloy structure. Segregation energies and metal distribution in the nanoalloys were modeled by density functional theory calculations. The results indicate slow metal interdiffusion at the nanoscale, which has important implications for synthetic methods aimed at core-shell particles.

Squared Extrapolation Methods (SQUAREM): A New Class of Simple and Efficient Numerical Schemes for Accelerating the Convergence of the EM Algorithm

We derive a new class of iterative schemes for accelerating the convergence of the EM algorithm, by exploiting the connection between fixed point iterations and extrapolation methods. First, we present a general formulation of one-step iterative schemes, which are obtained by cycling with the extrapolation methods. We, then square the one-step schemes to obtain the new class of methods, which we call SQUAREM. Squaring a one-step iterative scheme is simply applying it twice within each cycle of the extrapolation method. Here we focus on the first order or rank-one extrapolation methods for two reasons, (1) simplicity, and (2) computational efficiency. In particular, we study two first order extrapolation methods, the reduced rank extrapolation (RRE1) and minimal polynomial extrapolation (MPE1). The convergence of the new schemes, both one-step and squared, is non-monotonic with respect to the residual norm. The first order one-step and SQUAREM schemes are linearly convergent, like the EM algorithm but they have a faster rate of convergence. We demonstrate, through five different examples, the effectiveness of the first order SQUAREM schemes, SqRRE1 and SqMPE1, in accelerating the EM algorithm. The SQUAREM schemes are also shown to be vastly superior to their one-step counterparts, RRE1 and MPE1, in terms of computational efficiency. The proposed extrapolation schemes can fail due to the numerical problems of stagnation and near breakdown. We have developed a new hybrid iterative scheme that combines the RRE1 and MPE1 schemes in such a manner that it overcomes both stagnation and near breakdown. The squared first order hybrid scheme, SqHyb1, emerges as the iterative scheme of choice based on our numerical experiments. It combines the fast convergence of the SqMPE1, while avoiding near breakdowns, with the stability of SqRRE1, while avoiding stagnations. The SQUAREM methods can be incorporated very easily into an existing EM algorithm. They only require the basic EM step for their implementation and do not require any other auxiliary quantities such as the complete data log likelihood, and its gradient or hessian. They are an attractive option in problems with a very large number of parameters, and in problems where the statistical model is complex, the EM algorithm is slow and each EM step is computationally demanding.

Performance of four dentine excavation methods in deciduous teeth

This in vitro study aimed to assess the speed and caries removal effectiveness of four different new and conventional dentine excavation methods. Eighty deciduous molars were assigned to four groups. Teeth were sectioned longitudinally through the lesion centre. Images of one half per tooth were captured by light microscope and confocal laser scanning microscopy (CLSM) to assess the caries extension. The halves were then reassembled and caries removed using round carbide bur (group 1), Er:YAG laser (group 2), hand excavator (group 3) and a polymer bur (group 4). The time needed for the whole excavation in each tooth was registered. After excavation, the halves were photographed by light microscope. Caries extension obtained from CLSM images were superimposed on the post-excavation images, allowing comparison between caries extension and removal. The regions where caries and preparation limits coincided, as well as the areas of over- and underpreparation, were measured. Steel bur was the fastest method, followed by the polymer bur, hand excavator and laser. Steel bur exhibited also the largest overpreparation area, followed by laser, hand excavator and polymer bur. The largest underpreparation area was found using polymer bur, followed by laser, hand excavator and steel bur. Hand excavator presented the longest coincidence line, followed by polymer and steel burs and laser. Overall, hand excavator seemed to be the most suitable method for carious dentine excavation in deciduous teeth, combining good excavation time with effective caries removal.

Computational Techniques for Spatial Logistic Regression with Large Datasets

In epidemiological work, outcomes are frequently non-normal, sample sizes may be large, and effects are often small. To relate health outcomes to geographic risk factors, fast and powerful methods for fitting spatial models, particularly for non-normal data, are required. We focus on binary outcomes, with the risk surface a smooth function of space. We compare penalized likelihood models, including the penalized quasi-likelihood (PQL) approach, and Bayesian models based on fit, speed, and ease of implementation. A Bayesian model using a spectral basis representation of the spatial surface provides the best tradeoff of sensitivity and specificity in simulations, detecting real spatial features while limiting overfitting and being more efficient computationally than other Bayesian approaches. One of the contributions of this work is further development of this underused representation. The spectral basis model outperforms the penalized likelihood methods, which are prone to overfitting, but is slower to fit and not as easily implemented. Conclusions based on a real dataset of cancer cases in Taiwan are similar albeit less conclusive with respect to comparing the approaches. The success of the spectral basis with binary data and similar results with count data suggest that it may be generally useful in spatial models and more complicated hierarchical models.