Das Einbringen von Dotierstoffen in das Substratmaterial ist einer der wichtigsten Teilprozesse in der Halbleiterindustrie. Größe, Lage und Konzentration elektrisch aktiver Dotiergebiete bestimmen wesentlich die Eigenschaften der mikroelektronischen Basisbauelemente und damit die Funktionalität der Endprodukte. Die kontinuierliche Verkleinerung dieser Bauelemente zieht steigende Anforderungen an die Präzision bei ihrer Herstellung nach sich. Analyseverfahren, mit denen die genannten Kenngrößen gemessen werden können, sind aus diesem Grund von hoher Bedeutung. Elektronenholographie ist eine dafür prinzipiell geeignete Messmethode, da sie eine zweidimensionale Vermessung der durch die Dotanden veränderten Potentialstruktur des Halbleiters in der geforderten Ortsauflösung von wenigen nm erlaubt. Ein Teil dieser Arbeit befasst sich mit der Optimierung der für die holographische Untersuchung wichtigen Parameter. Zu diesem Zweck werden sowohl präparative Aspekte wie geeignete Probendicke und Struktur der Proben als auch messtechnische Aspekte wie kohärente Beleuchtung und TEM-Parameter diskutiert. Während sich der Hauptteil der Arbeit mit den dabei gewonnenen wissenschaftlichen Erkenntnissen befasst, werden im Anhang die bei Präparation und Messung wichtigen Details ausführlich beschrieben. Ein wesentliches Problem bei der elektronenholographischen Messung stellt die Präparation der Objekte für die Untersuchung im TEM dar. Die einzige sinnvolle Möglichkeit für eine industrielle Anwendung ist die Zielpräparation mit dem fokussierten Ionenstrahl („FIB“), da keine andere Methode vergleichbar effizient arbeitet. Leider wird bei dieser Art von Präparation die Probe von der Oberfläche bis in eine gewisse Tiefe sowohl strukturell als auch elektrisch verändert. Diese Artefakte beeinflussen das Ergebnis der hochsensiblen holographischen Messung. Um die gewonnenen Daten dennoch verlässlich quantitativ auswerten zu können, muss klar zwischen ursprünglichen Objekteigenschaften und präparativ induzierten Schädigungen unterschieden werden. Um dieses Ziel zu erreichen, wurden durch die FIB-Präparation hervorgerufene Schädigungen der Probe systematisch analysiert. Mit Hilfe von SIMS konnte die Tiefenverteilung des beim Ionenschneiden eingedrungenen Fremdmaterials gemessen werden. Es wurden Querschnitte von FIB-Proben durch konventionelle, holographische sowie holographisch-tomographische Abbildung im TEM an einer eigens dafür entwickelten nadelförmigen Probengeometrie untersucht. Dabei wurden die entstandenen strukturellen und elektrischen Veränderungen beobachtet und quantitativ charakterisiert. Der Einsatz von Tomographie erlaubte schließlich die Messung der Potentialverteilung im Inneren der Nadeln ohne eine Verfälschung durch Projektionseffekte. Es wurde gezeigt, dass die über die Schädigungen gewonnenen Erkenntnisse für eine Korrektur der holographischen Daten genutzt werden können. Dazu wurden entsprechende Untersuchungen an verschiedenen Bauelementen aus der Halbleiterindustrie durchgeführt. Die korrigierten Ergebnisse wurden dabei stets mit den theoretischen Erwartungen verglichen.

Ces dernières années avec les progrès matériels, les dimensions et le contenu des images acquises se sont complexifiés de manière notable. Egalement, le différentiel de performance entre les architectures classiques mono-processeur et parallèles est passé résolument en faveur de ces dernières. Pourtant, les manières de programmer sont restées largement les mêmes, instituant un manque criant de performance même sur ces architectures. Dans cette thèse, nous explorons en détails un algorithme particulier, les flots maximaux continus. Nous explicitons pourquoi cet algorithme est important et utile, et nous proposons plusieurs implémentations sur diverses architectures, du mono-processeur à l'architecture SMP et NUMA, ainsi que sur les architectures massivement parallèles des GPGPU. Nous explorons aussi des applications et nous évaluons ses performances sur des images de grande taille en science des matériaux et en biologie à l'échelle nano / In recent years, with the advance of computing equipment and image acquisition techniques, the sizes, dimensions and content of acquired images have increased considerably. Unfortunately as time passes there is a steadily increasing gap between the classical and parallel programming paradigms and their actual performance on modern computer hardware. In this thesis we consider in depth one particular algorithm, the continuous maximum flow computation. We review in detail why this algorithm is useful and interesting, and we propose efficient and portable implementations on various architectures. We also examine how it performs in the terms of segmentation quality on some recent problems of materials science and nano-scale biology

Les agrégats de suie émis par les processus de combustion dans les turbines aéronautiques contribuent à modifier le bilan radiatif de l'atmosphère et la qualité de l'air. La connaissance de leurs caractéristiques physiques et chimiques est indispensable pour évaluer leur rôle dans les processus physico-chimiques atmosphériques et leur impact sur l'environnement et la santé publique. Dans ce contexte, notre étude vise à déterminer la taille et la morphologie d'agrégats de suie aéronautique à partir de mesures expérimentales menées en microscopie électronique en transmission (MET) et en tomographie électronique.Nous avons réalisé des clichés MET d'agrégats de suie émis par des turboréacteurs aéronautiques. Nous avons établi une méthode pour caractériser la morphologie des agrégats en déterminant leur allongement, leur compacité et la tortuosité de leur contour en analysant leur projection. Nous avons également développé un logiciel de traitement et d'analyse des images MET qui permet de reconstruire en 3D un agrégat à partir de ses projections et l'analyse de ses caractéristiques dimensionnelles et morphologiques à partir de sa reconstruction. Les résultats obtenus nous ont permis d'étudier la validité des relations liant les caractéristiques microphysiques 2D et 3D proposées dans la littérature et d'en proposer de nouvelles pour les agrégats étudiés.Ces résultats constituent la première caractérisation morphologique 3D d'agrégats de suie aéronautique à partir d'analyses par MET et tomographie électronique. Ils montrent que les propriétés morphologiques de ces agrégats ne permettent pas d'utiliser la méthode d'ensemble pour déterminer la dimension fractale massique. / Soot aggregates emitted by aircraft engines' combustion processes are involved in the modification of the global radiative budget and the air quality. The knowledge of their physical and chemical characteristics is a prerequisite to any evaluation of the way they may act in the atmospheric physical and chemical processes and their impact on the environment and public health. In this context, our study aims at determining the size and morphological characteristics of aircraft soot aggregates on the basis of experimental measurements by transmission electron microscopy (TEM) and electron tomography.We have acquired TEM pictures of soot aggregates emitted by aircraft engines. We have established a method to characterize the morphology of these aggregates by determining their elongation, their compacity and the tortuosity of their edge. This method is based on the analysis of their TEM projection. Besides, we have developed a software to process and analyse TEM pictures. It allows to reconstruct aggregates from their projections and to determine their size and morphological characteristics. Our results have lead us to study the validity of the relationships linking the 2D and 3D microphysical characteristics presented in the literature and to suggest new ones for the studied aggregates.These results constitute the first 3D morphological and size characterizations of aircraft soot aggregates using TEM and electron tomography. They highlight the fact that the morphological properties of these aggregates do not fulfil the hypotheses required for the use of the collective method to determine the mass fractal dimension.

During cell division, chromosome segregation takes place on bipolar, microtubulebased spindles. Here, C. elegans is used to analyze spindle organization under both mitotic and meiotic conditions. First, the role of SAS-4 in organizing centrosome structure was analyzed. Partial depletion of SAS-4 in early embryos results in structurally defective centrioles. The study of this protein sheds light on the poorly understood role of the centrioles in dictating centrosome size. Second, the ultrastructure of wild-type mitotic spindle components was analyzed by electron tomography. This 3-D analysis reveals morphologically distinct microtubule end morphologies in the mitotic spindle pole. These results have structural implications for models of microtubule interactions with centrosomes Third, spindle assembly was studied in female meiosis. Specifically, the role of the microtubule severing complex katanin in spindle organization was analyzed. Electron tomography reveals fragmentation of spindle microtubules and suggests a novel katanin-dependent mechanism of meiotic spindle assembly. In this model, relatively long microtubules seen near the meiotic chromatin are converted into numerous short fragments, thus increasing the total number of polymers in an acentrosomal environment. Taken together, these results provide novel insights into the three-dimensional organization of microtubules during spindle assembly. / Die Segregation der Chromosomen während der Zellteilung wird duch bipolare, von Microtubuli-aufgebauten Spindlen gewährleistet. In der vorliegenden Arbeit wird C. elegans zur Analyse der Spindelorganisation unter mitotischen und meiotischen Bedingungen herangezogen. Erstens wird die Rolle von SAS-4 in der Organisation von Zentrosomen untersucht. Die partielle Depletierung von SAS-4 in frühen Embryonen führt zu strukturell defekten Zentriolen und wirft somit Licht auf die wenig verstandene Rolle der Zentriolen in der Bestimmung der Zentrosomengröße. Zweitens wird die Ultrastruktur der mitotischen Spindelkomponenten im Wildtyp durch Elektronentomographie untersucht. Diese 3-D-Analyse zeigt, dass im mitotischen Spindlepol unterschiedliche Morphologien der Mikrotubulienden zu finden sind. Diese Ergebnisse haben strukturelle Implikationen für Modelle der Mikrotubuli-Zentrosomen-Interaktionen. Drittens wird der Aufbau der Spindel in der weiblichen Meiose, speziell die Rolle des Mikrotubuli-schneidenden Kataninkomplexes in der Spindelorganisation, untersucht. Die Elektronentomographie zeigt hier eine Fragmentierung der Spindelmikrotubuli. Basierend auf diesem Ergebnis wird ein neues Katanin-abhängiges Modell der Formierung der Meiosespindel entwickelt, in dem relativ lange Microtubuli in Nähe des meiotischen Chromatins in zahlreiche kurze Mikrotubuli “zerschnitten” werden. Dies erhöht die Anzahl der verfügbaren Polymere in dieser azentrosomalen Situation. Zusammenfassend bringen diese Ergebnisse neue Einsichten in die räumliche Organisation der Mikrotubuli während des Spindelaufbaus.

The effect of individual and combined zirconium and manganese additions have been compared for an AA2198 6 mm thick sheet in T351 temper regarding their influence primarily on recrystallisation resistance and secondly on fracture toughness and overageing resistance. A complete characterisation of the dispersoid distributions was carried out for a deeper understanding of the effects of the Al3Zr and Al20Cu2Mn3 particles, involving studying their formation from the as-cast and homogenised stage.The most important finding in this work was the lower recrystallisation resistance in the alloy containing 0.1 wt%Zr + 0.3 wt%Mn compared to that containing only 0.1 wt%Zr. This result was rather unexpected, if one considers the opposite microsegregation patterns of Zr and Mn during casting, which leads to dispersoids occupying the majority of the grains’ volume and minimising dispersoid-free zones that could be potential sites for nucleation of recrystallisation. The other two alloys with dispersoid additions 0.05 wt%Zr + 0.3 wt%Mn and 0.4 wt%Mn, were partially and fully recrystallised respectively in the rolled T351 condition.Equally important in this work, was the observation that the opposite microsegregation trend of Zr and Mn sufficed to restrict grain growth in unrecrystallised areas. The 0.1Zr-0.3Mn alloy exhibited the lowest grain size of all alloys, both in the T351 temper and after annealing at 535oC for up to 144 hours. The reason for this was the combined action of Al20Cu2Mn3 dispersoids and Mn solute in the regions where the Zr concentration was low (i.e. near the grain boundaries), which offered additional pinning pressure to those areas compared to the 0.1Zr alloy.The lower recrystallisation resistance of the 0.1Zr-0.3Mn alloy was explained on the grounds of two main factors. The first was the lower subgrain size and hence stored energy within bands of Al20Cu2Mn3 dispersoids, which increased the driving force for recrystallisation in these regions. The second was the interaction between Zr and Mn that led to a decrease in the Al3Zr number density and pinning pressure. Since Zr was the dominant dispersoid family in terms of inhibiting recrystallisation, inevitably this alloy became more prone to recrystallisation. The Al3Zr pinning pressure was found to be much lower especially within bands of Al20Cu2Mn3 dispersoids. The detrimental effect of the Mn addition on the Al3Zr distribution was proven not to result from the dissolution of Zr within Mn-containing phases, and several other phases, at the grain interior and also in grain boundaries. The observed effect could not be precisely explained at this stage.Concerning mechanical properties, the 0.1Zr alloy exhibited the best combination of properties in the Kahn tear tests for fracture toughness. Further, it had a higher overageing resistance compared to the 0.1Zr-0.3Mn alloy.As an overall conclusion from this work, considering all the studied properties here that are essential for damage tolerant applications, the addition of 0.1 wt%Zr to the AA2198 6 mm thick sheet was found to be superior to that of the combined addition of 0.1 wt%Zr + 0.3 wt%Mn.

During cell division, chromosome segregation takes place on bipolar, microtubulebased spindles. Here, C. elegans is used to analyze spindle organization under both mitotic and meiotic conditions. First, the role of SAS-4 in organizing centrosome structure was analyzed. Partial depletion of SAS-4 in early embryos results in structurally defective centrioles. The study of this protein sheds light on the poorly understood role of the centrioles in dictating centrosome size. Second, the ultrastructure of wild-type mitotic spindle components was analyzed by electron tomography. This 3-D analysis reveals morphologically distinct microtubule end morphologies in the mitotic spindle pole. These results have structural implications for models of microtubule interactions with centrosomes Third, spindle assembly was studied in female meiosis. Specifically, the role of the microtubule severing complex katanin in spindle organization was analyzed. Electron tomography reveals fragmentation of spindle microtubules and suggests a novel katanin-dependent mechanism of meiotic spindle assembly. In this model, relatively long microtubules seen near the meiotic chromatin are converted into numerous short fragments, thus increasing the total number of polymers in an acentrosomal environment. Taken together, these results provide novel insights into the three-dimensional organization of microtubules during spindle assembly. / Die Segregation der Chromosomen während der Zellteilung wird duch bipolare, von Microtubuli-aufgebauten Spindlen gewährleistet. In der vorliegenden Arbeit wird C. elegans zur Analyse der Spindelorganisation unter mitotischen und meiotischen Bedingungen herangezogen. Erstens wird die Rolle von SAS-4 in der Organisation von Zentrosomen untersucht. Die partielle Depletierung von SAS-4 in frühen Embryonen führt zu strukturell defekten Zentriolen und wirft somit Licht auf die wenig verstandene Rolle der Zentriolen in der Bestimmung der Zentrosomengröße. Zweitens wird die Ultrastruktur der mitotischen Spindelkomponenten im Wildtyp durch Elektronentomographie untersucht. Diese 3-D-Analyse zeigt, dass im mitotischen Spindlepol unterschiedliche Morphologien der Mikrotubulienden zu finden sind. Diese Ergebnisse haben strukturelle Implikationen für Modelle der Mikrotubuli-Zentrosomen-Interaktionen. Drittens wird der Aufbau der Spindel in der weiblichen Meiose, speziell die Rolle des Mikrotubuli-schneidenden Kataninkomplexes in der Spindelorganisation, untersucht. Die Elektronentomographie zeigt hier eine Fragmentierung der Spindelmikrotubuli. Basierend auf diesem Ergebnis wird ein neues Katanin-abhängiges Modell der Formierung der Meiosespindel entwickelt, in dem relativ lange Microtubuli in Nähe des meiotischen Chromatins in zahlreiche kurze Mikrotubuli “zerschnitten” werden. Dies erhöht die Anzahl der verfügbaren Polymere in dieser azentrosomalen Situation. Zusammenfassend bringen diese Ergebnisse neue Einsichten in die räumliche Organisation der Mikrotubuli während des Spindelaufbaus.

Projet I : Les endospores représentent un état de dormance des bactéries leur permettant de résister à des conditions extrêmes et de persister pendant des années. La formation d'endospores a façonné l'évolution puisqu’elle se produit exclusivement chez les Firmicutes. Plusieurs études ont rapporté la formation d'endospores chez des espèces en dehors des Firmicutes, en particulier chez deux espèces de Protéobactéries, Rhodobacter johrii et Serratia marscescens, et une espèce d'Actinobacteries, Mycobacterium marinum. Le fait d’identifier les endospores en dehors des Firmicutes pourrait affecter la forme de l'arbre de vie et aiderait dans notre lutte contre les agents pathogènes humains. Par conséquent, nous avons visé d’étudier l'endosporulation chez ces trois espèces en utilisant des approches avancées d'imagerie et d'analyse, y compris la microscopie corrélative alliant la microscopie optique et électronique (CLEM), la tomographie de cryo- électron (cryo-ET) et la lipidomique. Nous avons utilisé la bactérie sporulante bien caractérisée Bacillus subtilis comme contrôle positif de la sporulation. L'examen de R. johrii, S. marcescens et M. marinum en utilisant CLEM et cryo-ET a montré que les objets à phase brillante ne ressemblaient à aucun stade de l'endosporulation. Les cryo-tomogrammes ont montré que les objets à phase brillante chez S. marcescens étaient des débris cellulaires agrégés de cellules mortes, alors qu'ils présentaient des structures granulaires typiques des cellules bactériennes chez les R. johrii et M. marinum. L'analyse lipidomique chez R. johrii a identifié les structures granulaires comme des granules de stockage potentiels enrichis en triacylglycérides (TAG). Nous pensons que les TAG peuvent fournir une source d'énergie pour résister à l'épuisement des nutriments. Des approches biochimiques et bioinformatiques supplémentaires ont soutenu nos conclusions selon lesquelles R. johrii, S. marcescens et M. marinum sont des bactéries non sporulantes. Projet II : Les plasmides jouent un rôle vital dans la propagation des gènes de résistance au sein et entre les espèces bactériennes. Par conséquent, il est essentiel de comprendre les systèmes bactériens impliqués dans le transfert et la maintenance des plasmides pour mieux aider dans notre lutte contre la propagation de la résistance aux antibiotiques. Dans cette thèse de doctorat, nous avons cherché à caractériser l'opéron alp7ARC, en utilisant l'homologue de l'actine bactérienne Alp7A pour séparer le plasmide pLS20 codant pour la résistance à la tétracycline dans B. subtilis. La stabilité du plasmide s'est avérée dépendante de l'opéron alp7ARC, indiquant un rôle essentiel dans la ségrégation plasmidique. Nos résultats préliminaires sur Alp7A ont montré qu'il s'assemble dans une nouvelle nanostructure tubulaire plutôt que des filaments, suggérant un nouveau mécanisme de ségrégation de l'ADN par Alp7A. Nous avons également étudié la structure d'Alp7A in vivo en utilisant une combinaison d'approches, notamment la biologie moléculaire, la Cryo-ET et la fLM. Nous avons également utilisé la CLEM pour localiser Alp7A dans des cellules entières à une résolution macromoléculaire. En outre, nous avons étudié la structure et la fonction d'Alp7A in vitro en transfectant B. subtilis et E. coli avec diverses constructions plasmidiques incorporant des mutations dans le gène d’Alp7A. Nous avons déployé différentes méthodes pour la purification de la protéine Alp7A, y compris la séparation par chromatographie, et le fractionnement au sulfate d'ammonium. J'ai discuté des divers défis que nous avons rencontrés dans ces expériences, tels que la contamination, l'instabilité de la protéine Alp7A et l'épaisseur bactérienne. Enfin, j'ai proposé des approches expérimentales alternatives qui aideraient à étudier le mécanisme de ségrégation des plasmides par Alp7ARC. / Project I: Endospores represent a dormant state of bacteria that allows them to withstand extreme conditions and persist for years. Endospore formation has shaped evolution, whereby it exclusively occurs in Firmicutes. Several studies have reported endospore formation in species outside of Firmicutes, particularly in two species of Proteobacteria, Rhodobacter johrii and Serratia marcescens, and one species of Actinobacteria, Mycobacterium marinum. Identifying endospores outside of Firmicutes would affect the shape of the tree of life and aid in our fight against human pathogens. Therefore, we aimed to investigate endosporulation in these three species using advanced imaging and analytical approaches, including correlative light and electron microscopy (CLEM), cryo-electron tomography (cryo-ET), and lipidomics. We used the well-characterized sporulating bacterium Bacillus subtilis as a positive control of sporulation. Examination of R. johrii, S. marcescens, and M. marinum using CLEM and cryo-ET showed that phase-bright objects did not resemble any stages of endosporulation. Cryo-tomograms revealed that the phase-bright objects in S. marcescens were aggregated cellular debris of dead cells, whereas they displayed granular structures typical of bacterial cells in R. johrii and M. marinum. Lipidomic analysis in R. johrii identified the granular structures as potential storage granules enriched with triacyl-glycerides (TAGs). We speculate that TAGs may provide an energy source to withstand the nutrient depletion. Additional biochemical and bioinformatics approaches supported our conclusions that R. johrii, S. marcescens, and M. marinum are non-sporulating bacteria. Project II: Plasmids play a vital role in the spread of resistance genes within and across bacterial species. Therefore, it is essential to understand the bacterial systems involved in the transfer and maintenance of plasmids to better aid in our fight against the spread of antibiotic resistance. In this doctorate, we aimed to characterize the alp7ARC operon, employing the bacterial actin homolog Alp7A to segregate the tetracycline resistance-encoding plasmid pLS20 in B. subtilis. The stability of the plasmid was shown to be dependent on the alp7ARC operon, indicating an essential role in plasmid segregation. Preliminary results on Alp7A showed that it assembles into a novel tubular nanostructure rather than filaments, suggesting a novel mechanism for DNA segregation by Alp7A. We further studied the structure of Alp7A in vivo using combination of approaches, including molecular biology, cryo-ET, and fLM. We also used CLEM to localize Alp7A in whole cells to a macromolecular resolution. Besides, we investigated the structure and function of Alp7A in vitro by transfecting E. coli with various plasmid constructs and purification by several methods, including affinity chromatography and ammonium sulfate precipitation. I discussed the diverse challenges we encountered in these experiments, such as bacterial thickness, contamination, and Alp7A protein instability. Finally, I proposed alternative experimental approaches for investigating the mechanism of plasmid segregation by Alp7ARC.

Studying the internal characteristics of a network using measurements obtained from endhosts is known as network tomography. The foremost challenge in measurement-based approaches is the large size of a network, where only a subset of measurements can be obtained because of the inaccessibility of the entire network. As the network becomes larger, a question arises as to how rapidly the monitoring resources (number of measurements or number of samples) must grow to obtain a desired monitoring accuracy. Our work studies the scalability of the measurements with respect to the size of the network. We investigate the issues of scalability and performance evaluation in IP networks, specifically focusing on fault and congestion diagnosis. We formulate network monitoring as a machine learning problem using probabilistic graphical models that infer network states using path-based measurements. We consider the theoretical and practical management resources needed to reliably diagnose congested/faulty network elements and provide fundamental limits on the relationships between the number of probe packets, the size of the network, and the ability to accurately diagnose such network elements. We derive lower bounds on the average number of probes per edge using the variational inference technique proposed in the context of graphical models under noisy probe measurements, and then propose an entropy lower (EL) bound by drawing similarities between the coding problem over a binary symmetric channel and the diagnosis problem. Our investigation is supported by simulation results. For the congestion diagnosis case, we propose a solution based on decoding linear error control codes on a binary symmetric channel for various probing experiments. To identify the congested nodes, we construct a graphical model, and infer congestion using the belief propagation algorithm. In the second part of the work, we focus on the development of methods to automatically analyze the information contained in electron tomograms, which is a major challenge since tomograms are extremely noisy. Advances in automated data acquisition in electron tomography have led to an explosion in the amount of data that can be obtained about the spatial architecture of a variety of biologically and medically relevant objects with sizes in the range of 10-1000 nm A fundamental step in the statistical inference of large amounts of data is to segment relevant 3D features in cellular tomograms. Procedures for segmentation must work robustly and rapidly in spite of the low signal-to-noise ratios inherent in biological electron microscopy. This work evaluates various denoising techniques and then extracts relevant features of biological interest in tomograms of HIV-1 in infected human macrophages and Bdellovibrio bacterial tomograms recorded at room and cryogenic temperatures. Our approach represents an important step in automating the efficient extraction of useful information from large datasets in biological tomography and in speeding up the process of reducing gigabyte-sized tomograms to relevant byte-sized data. Next, we investigate automatic techniques for segmentation and quantitative analysis of mitochondria in MNT-1 cells imaged using ion-abrasion scanning electron microscope, and tomograms of Liposomal Doxorubicin formulations (Doxil), an anticancer nanodrug, imaged at cryogenic temperatures. A machine learning approach is formulated that exploits texture features, and joint image block-wise classification and segmentation is performed by histogram matching using a nearest neighbor classifier and chi-squared statistic as a distance measure.

Cryo-electron microscopy (cryo-EM) has been established as the leading imaging technique for structural studies from small proteins to whole cells at a molecular level. The great advances in cryo-EM have led to the ability to provide unique insights into a wide variety of biological processes in a close to native, hydrated state at near-atomic resolutions. The developments of computational approaches have significantly contributed to the exciting achievements of cryo-EM. This dissertation emphasizes new approaches to address image processing problems in cryo-EM, including tilt series alignment evaluation, simultaneous determination of sample thickness, tilt, and electron mean free path based on Beer-Lambert law, Model-Based Iterative Reconstruction (MBIR) on tomographic data, minimization of objective lens astigmatism in instrument alignment and defocus and magnification dependent astigmatism of TEM images. The final goal of these methodological developments is to improve the 3D reconstruction of cryo-EM and visualize more detailed characterization.