Publication Server of Constructor University Library
Not a member yet
858 research outputs found
Sort by
Interactive Visualization for Assistance of Needle-Based Interventions
No full textNeedle-based interventions are an important part of tumor ablation therapies. Due to their common technical procedure, low complication rate, and low cost, needle-based ablation therapies have become alternatives to surgical resections in clinical practice. To date, the most well-studied and clinically relevant ablation therapy is radiofrequency ablation (RFA) in the liver. However, achieving successful outcomes is a challenging task which depends on a variety of partially unknown reasons. This uncertainty is mirrored by the high local recurrence rates reported in clinical studies. The aim of this PhD thesis is the investigation of interactive visualization methods to support the physician in planning and assessing needle-based interventions, in particular RFA.
This work proposes a software workflow designed to guide the physician in planning the intervention in clinical practice. To support the physician with spatial information about pathological structures as well as finding trajectories without harming vitally important tissue, 2D slice and volume visualization techniques for interactive planning of needle-based interventions are presented. A further contribution for planning RFA is the fast approximation of the ablation zone, incorporating the heat-sink effects of blood vessels which decrease thermal ablation. In this approach, weighted distance fields are utilized to calculate the ablation zone within interactive frame rates on the graphics processing unit (GPU).
To support the assessment of needle-based interventions such as RFA, an imageprocessing workflow for automatic alignment of CT images is proposed. Based on segmentation masks of tumor and coagulation, the workflow enables an automatic rigid registration algorithm to perform at least as accurately as experienced medical experts, but in significantly less time. A further contribution is a novel overview visualization as well as a navigation tool, the so-called tumor map. The goal of this method is the combined visualization of the tumor and its surface distance to the coagulation necrosis to support the physician in reliable therapy assessment. The tumor map additionally serves as an interactive tool for intuitive navigation within a 3D volume rendering of the tumor vicinity as well as within familiar 2D viewers.
To facilitate the development of highly interactive visualization techniques, a rapid prototyping framework for GPU-based volume rendering is developed as part of this PhD thesis. The framework supports flexible extension and dynamic alteration of the volume rendering configuration. Therefore, a dynamic shader pipeline based on the SuperShader concept allows the user to insert custom shaders at run-time of the prototyping framework.Nadelbasierte Interventionen stellen einen wichtigen Teil von Ablationstherapien zur Behandlung von Tumoren dar. Durch ihre einfache Handhabbarkeit, geringen Komplikationsraten und geringfügigen Kosten gelten sie in der klinischen Routine als alternative Therapien gegenüber der chirurgischen Resektion. Bis dato ist die Radiofrequenz Ablation (RFA) in der Leber die meist studierte und klinisch relevanteste Ablationstherapie. Allerdings ist das Erreichen von erfolgreichen Resultaten eine herausfordernde Aufgabe, welche von teilweise unbekannten Ursachen abhängig ist. Klinische Studien spiegeln diese Unsicherheit in hohen Rezidivraten wieder. Das Ziel dieser Dissertation ist die Erforschung von interaktiven Visualisierungsmethoden um Radiologen bei der Planung und Bewertung von nadelbasierten Interventionen, insbesondere der RFA, zu unterstützen.
In dieser Arbeit wird ein Software-Workflow präsentiert, welcher den Radiologen bei der Planung von Interventionen in der klinischen Routine assistiert. Um den Radiologen mit räumlichen Information von pathologischen Strukturen zu unterstützen, sowie das Finden von Zugangswegen unter Berücksichtigung von Risikostrukturen zu erleichtern, werden 2D Slice- und 3D Volumenrendering Techniken vorgestellt. Ein weiterer Beitrag dieser Arbeit ist die schnelle Berechnung der Ablationszone unter Berücksichtigung der Kühleffekte von nahegelegenen Blutgefäßen, welche die Größe der Ablationszone reduzieren können. In diesem Verfahren werden gewichtete Distanzfelder verwendet, welches es erlauben die Ablationszone in Echtzeit auf der Grafikkarte zu berechnen.
Um die Bewertung von nadelbasierten Interventionen wie die RFA zu unterstützen, wird in dieser Arbeit ein Workflow für die automatische Registrierung von CT-Bildern vorgestellt. Dieser erlaubt es eine rigide Registrierung, welche auf Segmentierungen von Tumor und Koagulation basiert, so genau wie erfahrene Mediziner, aber in kürzerer Zeit zu berechnen. Ein weiterer Beitrag ist eine neue Übersichtvisualisierung, welche auch als Interaktionswerkzeug zur Steuerung von 3D Volumenrendering bzw. von 2D Ansichten dient. Das Ziel dieser Methode ist die kombinierte Visualisierung der Tumoroberfläche sowie deren Distanz zur Koagulation, um den Radiologen bei der Bewertung des Therapieerfolgs zu unterstützen.
Als Teil dieser Dissertation wird außerdem ein Prototyping-Framework zur Erstellung von interaktiven Volumenvisualisierungen entwickelt. Das Framework erlaubt es die Konfiguration des Volumenrenderings flexibel zu erweitern und dynamisch anzupassen. Hierfür ermöglicht eine dynamische Shaderpipeline, welche auf dem SuperShader Konzept basiert, dem Benutzer eigens angepasste Shaderprogramme zur Laufzeit des Systems hinzuzufügen
On the Accuracy of Fiber Tractography
No full textInvasive neurosurgical interventions bear the risk of damaging indispensable fiber pathways. Current fiber reconstruction techniques based on diffusion tensor imaging (DTI) do not determine the spatial extent of a fiber bundle in an accurate and reliable manner, due to errors and imprecisions in both the imaging and the algorithmic pipeline. In this thesis, we start by quantifying the errors of current fiber tracking algorithms by means of novel software phantoms which provide a realistic model of neural fiber bundles. This knowledge is used to locally analyze the quality of a patient's diffusion tensor dataset and to construct individual confidence hulls around the tracked fibers. In the following chapter, these ideas are developed further and we suggest an accurate approach to the segmentation of a patient's diffusion tensor image which combines connectivity and tensor clustering information. We then focus our attention on the sensitivity of streamline tractography to user-defined regions of interest. In order to reduce this sensitivity, we demonstrate the feasibility of mapping a fiber bundle from a fiber atlas onto the DTI dataset of a patient. In the last chapter of this thesis, we consider several alternatives to visualize fiber-tracking related uncertainties by means of color-coded streamlines and confidence hulls
The consequences of ocean acidification on benthic foraminifera: calcification and potential proxy applications
No full textThis thesis describes effects of ocean acidification on foraminiferal calcification and the potential of their calcite composition to reconstruct paleo-seawater carbonate chemistry. The first objective of this work is to develop and improve the mechanistic understanding of foraminiferal biomineralization (Topic 1). The second objective is to assess effects of ocean acidification on foraminiferal calcification and, in particular, to quantify the effects of different parameters of the carbonate system individually (Topic 2). In Topic 3 a new calibration of foraminiferal U/Ca ratio as a carbonate system proxy is described
Investigating environment-specific signatures of marine pelagic microbes : Insights from comparative genomics
No full textPelagic microbes have adapted to a dynamic and challenging natural environment. Genomes of microbes from a wide range of environments are now available, creating new opportunities to investigate environment-specific genomic content. However, the lack of reliable environmental contextualisation severely limits comparative analysis. This thesis summarises efforts to enable,implement, and perpetuate large-scale comparative microbial genomics in order to detect environment-specific genomic signatures of marine pelagic microbes. Development and application of the Environment Ontology provided a controlled classification of genomes by environment type. An exploratory survey using this classification detected genomic features with relevance to the demands of the marine water column. Features of unknown function and those with regulatory function featured heavily among marine-specific genomic signatures. To investigate these, this thesis includes analyses of the Global Ocean Sampling metagenomes. These analyses provided ecogenomic perspectives on genomic features of unknown function and related transcriptional regulator abundance to the environmental stability of the water column. Their findings support the connection of these features to the niche-specific adaptations of pelagic microbes. Lastly, this thesis describes efforts to ensure steady growth of contextual data alongside genomic data to support future environment-enabled analyses. Community efforts in establishing contextual data standards are represented by the Minimum Information about a Marker gene Sequence (MIMARKS) and any Sequence (MIxS) checklist projects. The MetaBar and CDinFusion software tools, which promote contextual data acquisition and submission, were also developed in support of these efforts and are described. This thesis concludes with the description of the architecture and capability of Megx.net, which provided the framework for the integration and dissemination of many of this project’s outcoms
Multiscale Simulation Studies of Interactions of Carbon Nanotubes with Biopolymers and Lipid Bilayer
No full textEver since their discovery, carbon nanotubes (CNTs) have grabbed attention of the researchers from different fields of science and industry. In fact, their excellent physicochemical properties together with an ability to cross biological membranes provide innumerable possibilities for diverse applications, including biomedical and pharmacological applications. Surface modification of CNTs by biological molecules will create a new series of bioactive nanotubes which will lead to the direction of specific cell targeting. The main principles of CNT interactions with biological interfaces are still under active investigation. Experimental data suggest various ways of CNT surface modification, as well as, their pathways of internalization through the cellular membrane. Nowadays, the availability of both large computational resources and powerful computational methods, as Molecular Dynamics (MD) simulation, provides an opportunity to study these interactions at atomic level with high order of accuracy.
The goal of the research work for this thesis is the multi-scale simulation study of CNTs with various biological interfaces. Initially, the study of the CNT coating mechanism by surfactants at both atomistic and coarse-grained levels has been carried out. Linear polymeric ether – based surfactants are known to coat the surface of the nanotubes and, thus, make them soluble in water. In this simulation study the distribution of the polymers of different length around CNT, as well as, the numbers of aggregated chains are found to be compatible with experimental data results. Further, investigation of the orientation and aspect ratio dependent interaction of CNT bundles with the DPPC lipid bilayer has been performed by a recently developed MD-SCF approach. The simulations have provided a molecular model of the perturbation in the structure of the lipids bilayer induced by the CNT bundles insertion process. The results have shown that strong perturbations occur only when the CNT bundles are oriented perpendicularly respect to the bilayer plane. The pore formation has also been observed with the longest CNT bundle.
Finally, the atomistic MD simulations have been performed to study the structural behavior of antimicrobial hybrid peptide CA-MA in aqueous solutions under different physiological conditions. This specific hybrid peptide is of great interest to be explored, because of its bactericidal and tumoricidal abilities. The results of this study have shown that the peptide receives random coil conformation in water, which is in accordance with experimental results. Partial stabilization of the α-helix structure was found in the case of simulation in salty environment and the irreversible loss of the α-helical content was observed at the physiological temperature of 310 K. The peptide concentration plays a stabilizing role on secondary structure of the peptide. However, at these concentrations, peptide aggregation takes place. The studies of the CA-MA peptide have been further extended for the case of interactions with CNT and graphene sheet. In both cases the rapid loss of α-helical content has been observed at the time of the peptide’s contact with the solid carbon – based nanosurfaces
Characterization of phytoplankton in the North Sea with optical and molecular methods
No full textPhytoplankton is the basis of the marine food web and an important element in global carbon cycling. To understand processes, interactions and also shifts within its community, a thorough observation is required, especially in the light of a changing climate and increased utilization of marine ecosystems by humanity.
A major hindrance in observation is the associated effort: Traditionally, phytoplankton biomass is mostly determined via measurement of chlorophyll-a (chl-a), while community composition is primarily investigated by microscopy. This is time consuming, which limits the number of samples that can be analyzed and prevents thereby a high spatial and temporal resolution of the investigations. Furthermore, only insufficiently detailed information is gained about species which are not readily visible under the microscope and which have no morphological features for reliable identification. To overcome these limitations and to make phytoplankton investigation more convenient, a variety of methods have evolved in the last decades, mainly based on the measurement of optical proxy values or molecular biology. Some of them provide the opportunity for continuous measurement of parameters like chl-a, other facilitate a rapid analysis of discrete samples, even on a taxonomical level.
In the present study, a selection of these convenient methods (fluorescence and absorption measurements, flow cytometry, DNA/RNA-microarrays, and ARISA fingerprinting) was used in combination with the traditional approaches for characterizing the phytoplankton community in the whole German Bight over two years and different seasons. Furthermore, a newly designed device for the continuous measurement of water constituent absorption (ft-PSICAM) was also applied in the course of this investigation. Focus of the study was an evaluation of the different methods, especially with respect to a potential usage in routine monitoring. Where possible, this was carried out by a direct comparison of the results obtained by the different methods.
Absorption coefficients measurements were found to be a more stable alternative for the determination of chl-a and total suspended matter (TSM) in the water than traditional approaches based on chl-a fluorescence and turbidity, respectively. Furthermore, with the ft-PSICAM, it was possible to measure these absorption coefficients continuously.
The various methods for the assessment of the community composition turned out to be widely complementary to each other. While microscopy provided detailed quantitative and qualitative information, but was mainly restricted to microphytoplankton observation, primarily qualitative data were obtained by the molecular methods. This made them valuable to detect the presence or absence of taxa in the nano- and picophytoplankton, where reliable taxonomical information was otherwise not acquirable, as well as general changes in the community. Quantitative information about the size distribution within the small phytoplankton and also limited information about taxonomical composition was provided by flow cytometry. By analyzing the phytoplankton community in the German Bight with this complementary set of information, it could be found that the microphytoplankton community shows seasonal variation, with a dinoflagellate-dominated community in summer and a diatom-dominated one during spring and autumn. Moreover, also parameters like e.g. species number and biodiversity index exhibited seasonality. In contrast, such variations were not found for the nano- and picophytoplankton. Instead, an abrupt shift in this community was found from the end of 2010 to the beginning of 2011. It was characterized by an increasing contribution of cryptophytes, accompanied with a reduction in cryptophyte and prasinophyte diversity. The reason for this shift was not explainable in the course of this study.
In summary, the convenient methods were found to be a valuable addition to traditional methods by expanding the measurement spectrum. Furthermore, depending on the desired level of detail, they can also be considered suitable to reduce measurement effort in routinely observation. Nevertheless, in order to develop the full potential of these methods, further improvement is required especially with respect to obtaining quantitative taxonomical information
Investigation of the specificity of protein lysine methyltransferases
No full textWe followed the specificity profile based method to identify the best substrate and/or to find novel non-histone targets of the MLL1, NSD2 and SET2 protein lysine methyltransferases. Specificity profile is the specific amino acid sequence motif recognized as substrate by the respective methyltransferase's active site. The MLL1 methyltransferase plays its role in development and leukemia. It exists in vivo, in a multisubunit complex. We showed that MLL1 recognizes a five amino acid residue sequence in which the target lysine is in the center. We investigated the specificity profile to find possible non-histone targets for MLL1, first at the peptide level and then at the protein level in vitro. We found one novel non-histone target at the protein level, the MLL1 protein itself. This auto-methylation of MLL1 could have functional consequences. Of the two MLL1 constructs used in this study, the shorter MLL13812 was significantly more catalytically active than the longer MLL13745. The NSD2 methyltransferase plays a role in some types of cancer. We showed that NSD2 methylates the peptides H3 (1-20), H3 (15-35), H3 (28-48), H4 (38-52) and H1.5 (161-175) which harbour the potential methylation sites H3K4, H3K9, H3K18, H3K27, H3K36, H4K44 and H1.5K168, but it did not methylate the peptide H4 (10-30) which harbours the H4K20 site. Since, the NSD2 methyltransferase was not specific for any of the histone tail peptides, we devised an unbiased random peptide array approach to investigate its substrate specificity and/or non-histone targets. We found that NSD2 methylated those substrates in which a hydrophobic residue preceded the target lysine. This led us to find a novel and stronger histone site, K168 of histone H1.5 than the original K36 of histone H3 in vitro at the peptide level. The methylation of H1.5K168 could have roles in the regulation of gene expression. It may have connections with nucleosome compaction. The SET2 methyltransferase has tumor suppressive functions. We derived the substrate specificity profile for SET2 using peptide arrays. We showed that SET2 recognizes a six amino acid residue sequence in which the target lysine is at position 4. We report that SET2 is a robust H3K36 mono- and di-methyltransferase at the peptide level in vitro. We found three non-histone targets for SET2 at the peptide level in vitro. We report that lysine 119 of Chloride intracellular channel 1 (CLIC1, NCC27), lysine 2245 of A kinase anchoring protein 13 and lysine 683 of ASHIL (another H3K36 methyltransferase) are methylated by SET2 at the peptide level in vitro
Branching laws for tensor modules over classical locally finite Lie algebras
No full textGiven an embedding of a Lie algebra g' into a Lie algebra g and an irreducible g-module M, the branching problem is to determine the structure of M as a module over g'. In this thesis, we consider the case when both g' and g are classical locally finite Lie algebras (embeddings of such Lie algebras have been described by I. Dimitrov and I. Penkov) and M is a simple tensor g-module (the class of tensor modules has been introduced and studied in a series of papers by E. Dan-Cohen, I. Penkov, V. Serganova, and K. Styrkas). The goal of the thesis is to solve the branching problem for such triples g', g, and M. Since M is in general a not completely reducible g'-module, we determine the socle filtration of M over g'. Due to the description of embeddings of classical locally finite Lie algebras given by I. Dimitrov and I. Penkov, when g' is simple our result holds for all possible embeddings of g' into g
Charge Transport Investigation of poly(3-hexylthiophene) by Electroreflectance Spectroscopy
No full textOrganic semiconductors reveal charge transport properties, which can not be explained with standard models of inorganic semiconductors. In this PhD thesis, electroreflectance (ER) spectroscopy is used to investigate optically charge transport processes in organic semiconductor thin films, in order to achieve detailed information of the accumulation layer close to the insulator-organic semiconductor interface. This technique probes the reflectance changes due to an altered charge density within the organic semiconductor. The reflectance geometry offers the possibility to investigate samples with optically opaque substrates. Optical simulations of layer stacks were used to determine quantitatively changes of the dielectric function due to charge injection into the accumulation layer of a poly(3-hexylthiophene) (P3HT) thin film. The spin-coated layers of P3HT revealed an anisotropic dielectric function, as found with spectroscopic ellipsometry. Furthermore, ER data of three samples with altered insulator-organic semiconductor interfaces are analyzed with the developed theoretical approach. Characteristic spectral changes were found in laterally resolved measurements depending on the insulator-organic semiconductor interface. Significant distance-dependent spectral changes indicate a time-dependent relaxation process after charges have been injected from a Au electrode
Physicochemical aggregation of bioparticles and downstream processing of bioactive sulfated microalgal polysaccharides : Bioprocessing of microalgae and their associated metabolites
No full textThe research objectives investigated in this project include physicochemical harvesting of Nannochloropsis salina cells from dilute culture suspensions, extraction of lipids from N. salina concentrates and their conversion to fatty acid methyl esters (FAMEs) and determination of the content of the flocculating chemicals in lipids and FAMEs. Moreover, the effect of recycling the filtrate derived from chemical flocculation of N. salina on the growth of N. salina was analyzed. Part of the work also focused on isolation and purification of polysaccharides from Porphyridium purpureum and determination of the polysaccharide’s Z-average diameters and zeta potentials and polysaccharide-matrix binding thermodynamics on selected anion exchangers. Following extraction of bioactive polysaccharide, the assay for screening antiviral efficacy of sulfated polysaccharides from the microalgae was developed after which the antiviral efficacy of partially purified metabolites was tested against dairy bacteriophages. Additionally, the extracts from P. purpureum and Isochrysis galbana were screened for antibacterial and antifungal properties. Lastly, the bacterial communities in symbiotic interaction with P. purpureum cultures were determined and studied to determine whether or not they affect polysaccharide biosynthesis by the microalgae