Les traitements en radiothérapie consistent à irradier le patient avec desfaisceaux de particules énergétiques (typiquement des photons) ciblant la tumeur. Cesparticules sont transporté à travers le milieu et y dépose de l'énergie. Cette énergiedéposée, appelée la dose, est responsable des effets biologiques des radiations.Ce travail a pour but de développer des méthodes numériques de calcul etd'optimisation de la dose qui sont compétitives en termes de coût de calcul et de précisionpar rapport à des méthodes de référence.Le mouvement des particules est d'abord étudié via un système d'équationscinétiques linéaires. Cependant, résoudre directement ces systèmes est numériquementtrop coûteux pour des applications médicales. Pour palier ce coût de calcul, la méthodemoment, et en particulier les modèles Mn, est utilisée. Ces équations aux moments sontnon linéaires et valides sous une condition appelée réalisabilité.Les schémas numériques standards pour les équations aux moments sontcontraints par des conditions de stabilité qui se trouvent être très restrictives lorsque lemilieu contient des zones sous-denses. Des schémas numériques inconditionnellementstables et adaptés aux équations aux moments (préservant la réalisation) sontdéveloppés. Ces schémas se révèlent compétitifs en termes de coûts de calcul par rapportaux approches de référence. Finalement, ces méthodes sont appliquées dans uneprocédure d'optimisation visant à maximiser la dose dans la tumeur et la minimiser dansles tissus sains. / Radiotherapy treatments consists in irradiating the patient with beams ofenergetic particles (typically photons) targeting the tumor. Such particles are transportedthrough the medium and deposit energy in the medium. This deposited energy is the socalleddose, responsible for the biological effect of the radiations.The present work aim to develop numerical methods for dose computation andoptimization that are competitive in terms of computational cost and accuracy compared toreference method.The motion of particles is first studied through a system of linear transport equationsat the kinetic level. However, solving directly such systems is numerically too costly formedical application. Instead, the moment method is used with a special focus on the Mnmodels. Those moment equations are non-linear and valid under a condition calledrealizability.Standard numerical schemes for moment equations are constrained by stabilityconditions which happen to be very restrictive when the medium contains low densityregions. Inconditionally stable numerical schemes adapted to moment equations(preserving the realizability property) are developped. Those schemes are shown to becompetitive in terms of computational costs compared to reference approaches. Finallythey are applied to in an optimization procedure aiming to maximize the dose in the tumorand to minimize the dose in healthy tissues.

No description available.

Chirality is a nuclear symmetry which is suggested to occur in nuclei when the total angular momentum of the system has an aplanar orientation [Fra97, Fra01]. It can occur for nuclei with triaxial shape, which have valence protons and neutrons with predominantly particle and hole nature. It is expected that the angular momenta of an odd particle and an odd hole (both occupying high-j orbitals) are aligned predominantly along the short and the long axes of the nucleus respectively, whereas the collective rotation occurs predominantly around the intermediate axis of a triaxially deformed nucleus in order to minimize the total energy of the system. Such symmetry is expected to be exhibited by a pair of degenerate DI = 1 rotational bands, i.e. all properties of the partner bands should be identical. The results suggested that spin independence of the energy staggering parameter S(I ) within two-quasiparticle chiral bands (previously suggested a fingerprint of chirality) is found only if the Coriolis interaction can be completely neglected. However, if the configuration is nonrestricted, the Coriolis interaction is often strong enough to create considerable energy staggering. It was also found that staggering in the intra- and inter-band B(M1) reduced transition probabilities (proposed as another fingerprint of chirality) may be a result of effects other than strongly broken chirality. Therefore, the use of the B(M1) staggering as a fingerprint of strongly broken chiral symmetry seems rather risky, in particular if the phase of the staggering is not checked.

Data on elastic scattering of 96 MeV neutrons from 56Fe, 89Y and 208Pb in the angular interval 10-70° are presented. The previously published data on 208Pb have been extended, as a new method has been developed to obtain additional information at the most forward angles. The results are compared with phenomenological and microscopic optical potentials. The theory predictions are in general in good agreement with the experimental data. A study of the deviation of the zero-degree cross section from Wick's limit has been performed. The data on 208Pb are in agreement with Wick's limit, while those on lighter nuclei overshoot the limit significantly. A novel analysis method has been developed to obtain the inelastic neutron emission cross sections from the existing 56Fe data. The method is based on folding a trial spectrum with the response of the detector setup. The data cover the angular interval 26-65° and an excitation energy range of 0-45 MeV, ranges hitherto not studied. The results are compared with nuclear model predictions and found to be in good agreement with the experimental data.

The heaviest fragments, recoils, have been studied in proton and deuteron induced 28Si reactions and proton-20Ne reactions at 100-300 MeV per nucleon. Inclusive charge and angular distributions and coincidences between He nuclei and recoils have been compared to two theoretical models, Dubna Cascade Model and JAERI Quantum Molecular Dynamics. The overall agreement was good for the reactions with 28Si, however the angular distributions of He fragments could not be reproduced. For the 20Ne reactions the recoil angular distributions were only reproduced for large angles. There was a significant underestimation at small angles and low recoil charge. α-clustering in the bombarding nucleus is a possible explanation for the deviations. In the 100 MeV per nucleon reactions all assumptions of the models may not be valid and the agreement was worst for these reactions. In proton-natXe reactions intermediate mass fragments have been studied from 200 to 1400 MeV. Slow ramping was used to scan the energy. Charge distributions and a caloric curve have been compared to Cascade Fragmentation Evaporation Model. Charge distributions showed good agreement for fragments with Z<8 but the heavier fragments were underestimated.

Despite the immense advances of science and medicine in recent years, several aspects regarding the physiology and the anatomy of the human brain are yet to be discovered and understood. A particularly challenging area in the study of human brain anatomy is that of brain connectivity, which describes the intricate means by which different regions of the brain interact with each other. The study of brain connectivity is deeply dependent on understanding the organization of white matter. The latter is predominantly comprised of bundles of myelinated axons, which serve as connecting pathways between approximately 10¹¹ neurons in the brain. Consequently, the delineation of fine anatomical details of white matter represents a highly challenging objective, and it is still an active area of research in the fields of neuroimaging and neuroscience, in general. Recent advances in medical imaging have resulted in a quantum leap in our understanding of brain anatomy and functionality. In particular, the advent of diffusion magnetic resonance imaging (dMRI) has provided researchers with a non-invasive means to infer information about the connectivity of the human brain. In a nutshell, dMRI is a set of imaging tools which aim at quantifying the process of water diffusion within the human brain to delineate the complex structural configurations of the white matter. Among the existing tools of dMRI high angular resolution diffusion imaging (HARDI) offers a desirable trade-off between its reconstruction accuracy and practical feasibility. In particular, HARDI excels in its ability to delineate complex directional patterns of the neural pathways throughout the brain, while remaining feasible for many clinical applications. Unfortunately, HARDI presents a fundamental trade-off between its ability to discriminate crossings of neural fiber tracts (i.e., its angular resolution) and the signal-to-noise ratio (SNR) of its associated images. Consequently, given that the angular resolution is of fundamental importance in the context of dMRI reconstruction, there is a need for effective algorithms for de-noising HARDI data. In this regard, the most effective de-noising approaches have been observed to be those which exploit both the angular and the spatial-domain regularity of HARDI signals. Accordingly, in this thesis, we propose a formulation of the problem of reconstruction of HARDI signals which incorporates regularization assumptions on both their angular and their spatial domains, while leading to a particularly simple numerical implementation. Experimental evidence suggests that the resulting cross-domain regularization procedure outperforms many other state of the art HARDI de-noising methods. Moreover, the proposed implementation of the algorithm supersedes the original reconstruction problem by a sequence of efficient filters which can be executed in parallel, suggesting its computational advantages over alternative implementations.

The interactions, with emulsion nuclei, of O-16 ions accelerated at 2.1 GeV/nucleon in the Berkeley accelerator, are analyzed. A phenomenological study of the events, particularly reaction cross sections and multiplicities is exposed to nuclear theories with nucleons acting individually and collectively An study of the angular distributions of the reaction products and their correlations, particularly in the transversal space, is shown. Finally, a discussion of the present parton model of the nucleon structure is done, comparing it with the experimental results.

La forta exigència de qualitat a que són sotmesos els productes actuals fa que els processos d'inspecció i control industrial ocupin, cada vegada més, un lloc destacat en la seva fabricació. La indústria relacionada amb els materials de superfície texturada i periòdica (malles metàl·liques, plàstics, paper, làmines, roba, etc.) no és aliena a aquest problema i, en l'actualitat, està esmerçant molts esforços en aquest camp. Un cas representatiu dels materials anteriors, que presenta una ampla problemàtica en quan a inspecció i control així com una elevada exigència de qualitat en l'acabat, és el teixit tèxtil.La majoria de defectes que es produeixen en el procés productiu d'un material tèxtil, com en tants altres materials i productes industrials, són detectats encara per inspectors humans. La seva feina és, en general, molt repetitiva i rutinària i, moltes vegades, el consum de temps que hi dediquen és molt elevat. A més cal fixar l'atenció en petits detalls situats en un camp extens, que sovint es troba en moviment, el que acaba produint fatiga visual en l'operari.En les últimes dècades, gràcies al gran desenvolupament tecnològic que hi ha hagut, estem assistint a una sèrie d'avenços, molt ràpids i de gran abast, en el món dels processos d'automatització i control. A conseqüència d'això, cada vegada el procés de producció és més ràpid, el que requereix que el control de qualitat també ho sigui.L'interès industrial en automatitzar les operacions d'inspecció és cada vegada més gran. A tall d'exemple la fira tèxtil internacional que reuneix més interès i expectació a nivell mundial, ITMA-99, celebrada aquest any a París, mostrava en els seus stands diferents productes dedicats a l'automatització dels processos d'inspecció i control de qualitat. Entre les ofertes d'aquesta fira cal citar una gran varietat d'instruments i dispositius: sensors òptics per inspeccionar la qualitat del fil en temps real en processos de filatura, espectrecolorímetres acoblats a l'estampadora i a la màquina de tintar per determinar amb més precisió el color, cameres lineals i cameres 2-D ultraràpides per detectar fils trencats en ordidors. En el camp del tissatge s'ha començat a veure en aquesta fira un sistema d'inspecció format per un conjunt (10 o 12) de cameres 2-D, situades damunt del teler, que detecta defectes locals que es poden produir en la roba durant el procés de producció. Des de la penúltima fira (ITMA-94) ha sigut novetat una màquina de repassar automàtica, on un sistema digital de processat format per 4 cameres i un potent ordinador, permet detectar una més ampla varietat de defectes locals i amb una velocitat nominal que pot arribar fins a 300 metres de roba per minut. Malgrat tot, les novetats aportades per la fira presenten grans limitacions: preu elevat, infrastructura complexa en la captació i processat de la imatge i, els resultats que s'obtenen només han estat validats per casos limitats (teixit de plana i d'un sol color). A més, si es té en compte que molts processos de producció són de tirada curta, canviant constantment el tipus d'article del teler, es necessita un sistema que sigui robust i versàtil, capaç de treballar amb diferents lligaments i colors i que el seu cost sigui més baix que el dels productes actuals.Revisant la última dècada, la més rellevant en el desenvolupament d'aquestes tecnologies, s'observa un clar increment en l'aplicació de tècniques de processat digital de la imatge a l'anàlisi de mostres texturades [Brzakovic-1992]. En el cas de la inspecció de materials tèxtils Chetverikov [Chetverikov-1988] desenvolupà un procés d'inspecció automàtica en teixits tèxtils que combina l'anàlisi de textures i la segmentació d'imatges. Neubauer [Neubauer-1992] segmentà defectes en mostres reals amb l'ajut d'una camera lineal i una xarxa neuronal. Siew et al. [Siew-1988] mesuraren les característiques texturals en catifes a partir d'estadístiques de primer i segon ordre aplicades als nivells de grisos de la imatge. Zhang i Bresse [Zhang-1995] detectaren i classificaren nusos, carreres i trencades mitjançant l'autocorrelació i operacions de morfologia matemàtica. Tsai et al. [Tsai-1995] detectaren i classificaren exactament quatre classes de defectes en teixits tèxtils utilitzant matrius de co-ocurrència i xarxes neuronals. Xu [Xu-1997] mesurà la rugositat en les superfícies de catifes a partir de la dimensió fractal de la imatge. Mitjançant tècniques combinades de filtrat Gaussià, umbralització, equalització de l'histograma i autocorrelació Kang et al. [Kang-1999] determinaren densitats lineals de fils en els direccions de trama i ordit per teixits de plana.La inspecció de materials texturats quasiperiòdics i, en particular els tèxtils, mitjançant tècniques de processat òptic-digital de la imatge és un camp d'investigació relativament poc treballat i que presenta un interès renovat. En aquest apartat es troben poques contribucions entre les que cal destacar Kang et al. [Kang-1999] que trobaven directament, a partir de la llum transmesa pel teixit, el punt de creuament dels fils en teixits de plana i Ciamberlini [Ciamberlini-1996] que operava amb la TF òptica de la llum transmesa pel teixit.Les eines que actualment estan al mercat són insuficients i es centren principalment en la resolució d'un aspecte: la detecció de defectes locals. En la literatura no és troben plantejaments generals que abastin des de models i procediments per caracteritzar les estructures fins a mecanismes que determinin la seva evolució o resistència davant d'agressions, la detecció de defectes globals en l'estructura i la detecció de defectes locals. Creiem que aquest plantejament general, que pot ser vàlid per a una gran varietat de materials, és també necessari. En aquest treball s'han aplicat, principalment a materials tèxtils, un conjunt d'eines inspirades en aquest plantejament.

Magnetic resonance imaging (MRI) has long become one of the most powerful and accurate tools of medical diagnostic imaging. Central to the diagnostic capabilities of MRI is the notion of contrast, which is determined by the biochemical composition of examined tissue as well as by its morphology. Despite the importance of the prevalent T₁, T₂, and proton density contrast mechanisms to clinical diagnosis, none of them has demonstrated effectiveness in delineating the morphological structure of the white matter - the information which is known to be related to a wide spectrum of brain-related disorders. It is only with the recent advent of diffusion-weighted MRI that scientists have been able to perform quantitative measurements of the diffusivity of white matter, making possible the structural delineation of neural fibre tracts in the human brain. One diffusion imaging technique in particular, namely high angular resolution diffusion imaging (HARDI), has inspired a substantial number of processing methods capable of obtaining the orientational information of multiple fibres within a single voxel while boasting minimal acquisition requirements. HARDI characterization of fibre morphology can be enhanced by increasing spatial and angular resolutions. However, doing so drastically reduces the signal-to-noise ratio. Since pronounced measurement noise tends to obscure and distort diagnostically relevant details of diffusion-weighted MR signals, increasing spatial or angular resolution necessitates application of the efficient and reliable tools of image denoising. The aim of this work is to develop an effective framework for the filtering of HARDI measurement noise which takes into account both the manifold to which the HARDI signal belongs and the statistical nature of MRI noise. These goals are accomplished using an approach rooted in non-local means (NLM) weighted averaging. The average includes samples, and therefore dependencies, from the entire manifold and the result of the average is used to deduce an estimate of the original signal value in accordance with MRI statistics. NLM averaging weights are determined adaptively based on a neighbourhood similarity measure. The novel neighbourhood comparison proposed in this thesis is one of spherical neighbourhoods, which assigns large weights to samples with similar local orientational diffusion characteristics. Moreover, the weights are designed to be invariant to both spatial rotations as well as to the particular sampling scheme in use. This thesis provides a detailed description of the proposed filtering procedure as well as experimental results with synthetic and real-life data. It is demonstrated that the proposed filter has substantially better denoising capabilities as compared to a number of alternative methods.

In this work, ultrafine-grained (UFG) Al-5wt%Zn alloy was produced by equal channel angular extrusion (ECAE). The microstructure evolution during ECAE and the mechanical properties of the UFG Al-Zn alloy were investigated. In order to identify the effect of Zn in the Al-Zn alloy, pure aluminum (4N, 99.99%) was also studied for comparison. The grains of the Al-Zn alloy could be refined effectively by increasing the ECAE passes. However, as the ECAE passes increased, the microhardness increased initially but maintained constant after 4 ECAE passes. The dislocation density within grain interior was decreased gradually with increasing ECAE passes. After being processed to twelve ECAE passes, the UFG Al-Zn alloy exhibited 53.7% of the grain boundaries being high angle grain boundaries (HAGBs). The UFG Al-5wt%Zn alloy exhibits superior tensile strength and elongation as compared with pure aluminum fabricated by the same ECAE process. Experimental results indicated that adding Zn in aluminum alloy could provide solid-solution strengthening and considerable enhancement in tensile ductility which might be related to an improved post-uniform elongation (PUE). The strain rate sensitivity (SRS) of the UFG Al-Zn alloy also increased with increasing the ECAE passes, which might be related to the fine grain size and the contribution of grain boundary sliding. The activation volume of the UFG Al-Zn alloy was in the range of 32b3~76b3, and the pure aluminum was in the range of 57b3~122b3. Because of the small value of the activation volume, it is suggested that the controlling mechanism for dislocation glide in the UFG Al-Zn alloy might be related to the generation and absorption of dislocations in grain boundary, as well as the interaction between dislocations and solute Zn atoms in the grain boundary.