The response of cells to their environment is driven by a variety of proteins and messenger molecules. In eukaryotes, their distribution and location in the cell is regulated by the vesicular transport system. The transport of aquaporin 2 between membrane and storage region is a crucial part of the water reabsorption in renal principal cells, and its malfunction can lead to Diabetes insipidus. To understand the regulation of this system, I aggregated pathways and mechanisms from literature and derived models in a hypothesis-driven approach. Furthermore, I combined the models to a single multi-scale model to gain insight into key regulatory mechanisms of aquaporin 2 recycling. To achieve this, I developed a computational framework for the modeling and simulation of cellular signaling systems. The framework integrates reaction and difusion of biochemical entities on a microscopic scale with mobile vesicles, membranes, and compartments on a cellular level. The simulation uses an adaptive step-width approach that e ciently regulates the agent-based simulation of macroscopic components with the numerical integration of mass action kinetics and grid-based nite diference methods. A reaction network generation algorithm was designed, that, in combination with a highly-modular modeling approach, allows for fast model prototyping. The analysis of the aquaporin 2 model system rationalizes that the compartmentalization of cAMP in renal principal cells is a result of the protein kinase A signalosome and can only occur if speci c cellular components are observed in conjunction. Endocytotic and exocytotic processes are inherently connected and can be regulated by the same protein kinase A signal.:Abstract 1. Introduction 1.1. Eukaryotic Signaling 1.2. Modeling and Simulation of Cellular Processes 1.3. Aquaporin 2 recycling 1.4. Motivation and Aims 1.5. Outline I. Background 2. Modeling and Simulation of Complex Signaling Pathways 2.1. Multi-scale Modeling 2.1.1. Approaches to Multi-scale Modeling 2.1.2. Reduction of Computational Complexity 2.2. Models of Chemical Reaction Networks 2.2.1. Reactions and Reaction Rates 2.2.2. Numerical Solutions 2.2.3. Reaction Network Generation 2.3. Models of Intracellular Transport 2.3.1. Undirected Transport 2.3.2. Directed Transport 3. Aquaporin 2 Recycling in Renal Principal Cells 3.1. The Physiology of Water Homeostasis 3.2. Molecular Mechanisms of the Vasopressin Response 3.2.1. The Vasopressin Receptor 3.2.2. cAMP Regulation of Protein Kinase A 3.2.3. Endo- and Exocytosis 3.3. Models of Water Transport in Renal Principal Cells II. Results & Discussion 4. Multi-scale Simulation of Cellular Signaling Pathways 4.1. Scale Separation and Bridging 4.2. Micro-scale Simulation Approach 4.2.1. Difusion and Discretization of the Simulation Space 4.2.2. Reaction Kinetics 4.3. Rule-based Reaction Network Generation 4.3.1. Definition of the Data Model 4.3.2. Design of Rule Based Reactions 4.3.3. Automated Generation of Reaction Networks 4.4. Macro-scale Simulation Approach 4.4.1. Agent-based Simulation of Discrete Entities 4.4.2. Modules for Displacement-based Behavior 4.5. Modularization and Error Estimation 4.5.1. Determination of the Numerical Error 4.5.2. Modularization of Concentration-based Events 4.5.3. Determination of the Displacement-based Error 5. Aquaporin 2 Recycling Model and Simulation 5.1. Model of Allosteric PKA Phosphorylation 5.1.1. Model Design 5.1.2. Simulation Results and Discussion 5.1.3. Conclusions 5.2. cAMP Compartmentalization in the Vesicle Storage Region 5.2.1. Model Design 5.2.2. Simulation Results and Discussion 5.2.3. Conclusions 5.3. Clathrin-mediated Endocytosis 5.3.1. Model Design 5.3.2. Simulation Results and Discussion 5.3.3. Conclusions 5.4. Intracellular Transport and Recycling 5.4.1. Model Design 5.4.2. Simulation Results and Discussion 6. Conclusion 6.1. Modeling and simulation approach 6.2. Insights into the AQP2 recycling model III. Appendix A. Code Availability B. Module Overview Bibliography

In dieser Dissertation wird die Entwicklung eines selbstorganisierenden Produktionssystems untersucht, um die Herausforderungen in der Produktionsplanung und -steuerung, insbesondere bei Unsicherheiten, besser zu bewältigen. Die Arbeit analysiert zunächst die Grundlagen der Produktionsplanung und -steuerung, gefolgt von einer detaillierten Untersuchung des interdisziplinären Themas der Selbstorganisation. Basierend darauf wird ein Vorgehen zur Identifikation geeigneter Methoden der Selbstorganisation für eine bestimmte Problemstellung entwickelt und auf Produktionsplanung und -steuerung angewendet. Durch die Bewertung bestehender Referenzarchitekturen und deren Integration in ein Framework für selbstorganisierende Produktion, werden selbstorganisierende Verfahren mit etablierten Verfahren der Produktionsplanung und -steuerung verglichen. Die Arbeit zeigt die Überlegenheit der selbstorganisierenden Systeme bei der Planung unter Unsicherheit und liefert nachweislich Verbesserungen hinsichtlich der sekundären und tertiären Ziele der Produktion. Die Dissertation gliedert sich in acht Kapitel, die jeweils verschiedene Aspekte der Selbstorganisation, Produktionsplanung und -steuerung, sowie die Entwicklung und Evaluation eines selbstorganisierenden Produktionssystems behandeln. Die Ergebnisse dieser Arbeit haben das Potenzial, die Produktionsplanung und -steuerung unter Unsicherheit effektiver zu gestalten und somit eine bedeutende Verbesserung für die Industrie zu bieten.:1. Einleitung 18 1.1. Motivation zur Nutzung von Selbstorganisation in der Produktionsplanung und -steuerung 1.2. Forschungsfragen: Wie kann Selbstorganisation bei der Bewältigung der Aufgaben der PPS helfen? 1.3. Aufbau und Vorgehensweise dieser Arbeit 2. Produktionsplanung und -steuerung 2.1. Einführung in die Produktionsplanung und -steuerung 2.1.1. Planung 2.1.2. Produktion 2.1.3. Produktionsplanung und -steuerung 2.2. Ziele der Produktionsplanung und -steuerung 2.2.1. Primäre Ziele 2.2.2. Sekundäre Ziele 2.2.3. Tertiäre Ziele 2.3. Zentral organisierte Produktionsplanung und -steuerung 2.3.1. Aufbau der zentral organisierten Produktionsplanung und -steuerung 2.3.2. Ablauf der zentral organisierten Produktionsplanung 2.3.3. Ablauf der zentral organisierten Produktionssteuerung 2.4. Unsicherheit in der Produktionsplanung und -steuerung 2.5. Arten von Entscheidungsproblemen in der Produktionsplanung und -steuerung 3. Selbstorganisation 3.1. Selbstorganisation in komplexen Systemen 3.2. Sich selbst organisierenden Systemen zugeschriebene Eigenschaften 3.2.1. Offenheit des Systems 3.2.2. Emergenz des Systems 3.2.3. Nichtlinearität des Systems 3.2.4. Attraktoren im Zustandsraum des Systems 3.2.5. Indeterminismus des Systems 3.2.6. Pfadabhängigkeit des Systems 3.2.7. Autonomie der Elemente im System 3.2.8. Autopoiesis neuer Elemente im System 3.2.9. Anpassungsfähigkeit des Systems 3.3. Von statischen zu sich selbst organisierenden Systemen 4. Vorgehensweise zur Entwicklung eines sich selbst organisierenden Systems zur Produktionsplanung und -steuerung 4.1. Vorgehen zur Entwicklung sich selbst organisierender Systeme 4.2. Elemente eines sich selbst organisierenden Systems 4.3. Umwelteinflüsse und Interaktionsmuster sich selbst organisierender Systeme 4.3.1. Taxonomie der Interaktion zur Koordination 4.3.2. Interaktionsmuster zur Koordination sich selbst organisierender Systeme 4.3.3. Zuordnung von Interaktionsmustern zu Entscheidungsproblemen 4.4. Organisationsstrukturen von Systemen 4.5. Ungewollte Phänomene bei der Verwendung von Selbstorganisation 4.6. Anwendungsgebiete der Selbstorganisation in der Produktionsplanung und -steuerung 5. Referenzarchitekturen für sich selbst organisierende Systeme und für die Produktionsplanung und -steuerung 85 5.1. Anforderungen an eine sich selbst organisierende Produktionsplanung und -steuerung 5.2. Referenzarchitekturen für sich selbst organisierende Systeme 5.2.1. Autonomic Computing 5.2.2. Organic Computing 5.2.3. Multi-Agenten Systeme 5.2.4. Aktor-Modell 5.2.5. Warteschlangennetzwerke 5.3. Referenzarchitekturen für die Produktionsplanung und -steuerung 5.3.1. Product Ressource Order Straff Architecture 5.3.2. ISA-95 5.3.3. Reference Architecture Model for Industry 4.0 5.4. Herausforderungen der Überführung einer sich selbst organisierenden Produktion in die Realität 6. Architektur für eine sich selbst organisierte Produktion 105 6.1. MATE - Manufacturing on Actor Technology 6.2. Elemente der sich selbst organisierenden Produktionsplanung und -steuerung 108 6.3. Umwelteinflüsse auf die sich selbst organisierende Produktionsplanung und -steuerung 6.4. Organisation der Elemente 6.4.1. Koordination der sich selbst organisierenden Produktionsplanung und -steuerung 6.4.2. Verhalten bei der sich selbst organisierten Planung 6.4.3. Verhalten bei der sich selbst organisierten Produktionssteuerung 7. Evaluierung der sich selbst organisierenden Produktionsplanung und -steuerung 122 7.1. Ablauf der Evaluation der Produktionsplanung und -steuerung 7.2. Datenerhebung für verschiedene Produktionen 7.2.1. Möglichkeiten der Datenerhebung 7.2.2. Grundlegendes Vorgehen zur Testdatengenerierung 7.2.3. Strukturbeschreibende Kennzahlen der Produktion 7.3. Generierung der Daten für die Evaluierung der Produktionsplanung und - steuerung 7.3.1. Validierung der generierten Testdaten 7.3.2. Generierung der zu evaluierenden Produktion 7.4. Beschreibung der Ansätze zur zentralen und dezentralen Produktionsplanung und -steuerung zum Zweck des Vergleichs 7.4.1. Zentral organisiertes Planungssystem 7.4.2. Dezentral organisiertes Planungssystem mittels erschöpfender Zuteilung 7.5. Vergleich der Ansätze hinsichtlich der Ziele der Produktionsplanung und - steuerung 7.5.1. Effizienz 7.5.2. Termintreue der Kundenaufträge 7.5.3. Durchlaufzeit der Materialien 7.5.4. Auslastung der Produktionsressourcen 7.5.5. Lagerbelegung über die Zeit 7.5.6. Stabilität 7.5.7. Robustheit 7.5.8. Flexibilität 7.5.9. Anpassungsfähigkeit 7.5.10. Skalierbarkeit 7.6. Vergleich der Ansätze hinsichtlich verschiedener Eigenschaften der Produktion 8. Einordnung und Bewertung der Ergebnisse, Zusammenfassung und Ausblick 8.1. Einordnung der Eigenschaften der sich selbst organisierenden Produktionsplanung und -steuerung 8.2. Bewertung der Ergebnisse der Evaluation der sich selbst organisierenden Produktionsplanung und -steuerung 8.3. Zusammenfassung 8.4. Ausblick

Diffusion Models have achieved state-of-the-art results in image generating tasks, yet face different challenges when used in different domains. We first give a brief overview of the Diffusion Models architecture. Then, we present a new model and architecture called SVGFusion that applies the principles of Diffusion Models to generate Vector Graphics. Vector Graphics have a complex structure and are vastly different than pixel images, and thus the main challenge when working with Vector Graphics is how to represent their complex structure in a way that a Diffusion Model can effectively process. We will explain this and the further challenges that we encountered during the process and how we successfully addressed some of them. We demonstrate the effectiveness of our approach by training a sample model on a decently sized dataset as well as running valuable experiments. Furthermore, we offer useful insights, recommendations and code to researchers who wish to further explore this topic.

This project concerns studying the early development of living organisms. This period is accompanied by dynamic morphogenetic events. There is an increase in the number of cells, changes in the shape of cells and specification of cell fate during this time. Typically, in order to capture the dynamic morphological changes, one can employ a form of microscopy imaging such as Selective Plane Illumination Microscopy (SPIM) which offers a single-cell resolution across time, and hence allows observing the positions, velocities and trajectories of most cells in a developing embryo. Unfortunately, the dynamic genetic activity which underlies these morphological changes and influences cellular fate decision, is captured only as static snapshots and often requires processing (sequencing or imaging) multiple distinct individuals. In order to set the stage for characterizing the factors which influence cellular fate, one must bring the data arising from the above-mentioned static snapshots of multiple individuals and the data arising from SPIM imaging of other distinct individual(s) which characterizes the changes in morphology, into the same frame of reference. In this project, a computational pipeline is established, which achieves the aforementioned goal of mapping data from these various imaging modalities and specimens to a canonical frame of reference. This pipeline relies on the three core building blocks of Instance Segmentation, Tracking and Registration. In this dissertation work, I introduce EmbedSeg which is my solution to performing instance segmentation of 2D and 3D (volume) image data. Next, I introduce LineageTracer which is my solution to performing tracking of a time-lapse (2d+t, 3d+t) recording. Finally, I introduce PlatyMatch which is my solution to performing registration of volumes. Errors from the application of these building blocks accumulate which produces a noisy observation estimate of gene expression for the digitized cells in the canonical frame of reference. These noisy estimates are processed to infer the underlying hidden state by using a Hidden Markov Model (HMM) formulation. Lastly, for wider dissemination of these methods, one requires an effective visualization strategy. A few details about the employed approach are also discussed in the dissertation work. The pipeline was designed keeping imaging volume data in mind, but can easily be extended to incorporate other data modalities, if available, such as single cell RNA Sequencing (scRNA-Seq) (more details are provided in the Discussion chapter). The methods elucidated in this dissertation would provide a fertile playground for several experiments and analyses in the future. Some of such potential experiments and current weaknesses of the computational pipeline are also discussed additionally in the Discussion Chapter.

Embedding small wireless sensors into the environment allows for monitoring physical processes with high spatio-temporal resolutions. Today, these devices are equipped with a battery to supply them with power. Despite technological advances, the high maintenance cost and environmental impact of batteries prevent the widespread adoption of wireless sensors. Battery-free devices that store energy harvested from light, vibrations, and other ambient sources in a capacitor promise to overcome the drawbacks of (rechargeable) batteries, such as bulkiness, wear-out and toxicity. Because of low energy input and low storage capacity, battery-free devices operate intermittently; they are forced to remain inactive for most of the time charging their capacitor before being able to operate for a short time. While it is known how to deal with intermittency on a single device, the coordination and communication among groups of multiple battery-free devices remain largely unexplored. For the first time, the present thesis addresses this problem by proposing new methods and tools to investigate and overcome several fundamental challenges.

Manyindustrialautomationsolutionsusewirelesscommunicationandrelyontheavail- ability and quality of the wireless channel. At the same time the wireless medium is highly congested and guaranteeing the availability of wireless channels is becoming increasingly difficult. In this work we show, that ad-hoc networking solutions can be used to provide new communication channels and improve the performance of mobile automation systems. These ad-hoc networking solutions describe different communi- cation strategies, but avoid relying on network infrastructure by utilizing the Peer-to- Peer (P2P) channel between communicating entities. This work is a step towards the effective implementation of low-range communication technologies(e.g. VisibleLightCommunication(VLC), radarcommunication, mmWave communication) to the industrial application. Implementing infrastructure networks with these technologies is unrealistic, since the low communication range would neces- sitate a high number of Access Points (APs) to yield full coverage. However, ad-hoc networks do not require any network infrastructure. In this work different ad-hoc net- working solutions for the industrial use case are presented and tools and models for their examination are proposed. The main use case investigated in this work are Automated Guided Vehicles (AGVs) for industrial applications. These mobile devices drive throughout the factory trans- porting crates, goods or tools or assisting workers. In most implementations they must exchange data with a Central Control Unit (CCU) and between one another. Predicting if a certain communication technology is suitable for an application is very challenging since the applications and the resulting requirements are very heterogeneous. The proposed models and simulation tools enable the simulation of the complex inter- action of mobile robotic clients and a wireless communication network. The goal is to predict the characteristics of a networked AGV fleet. Theproposedtoolswereusedtoimplement, testandexaminedifferentad-hocnetwork- ing solutions for industrial applications using AGVs. These communication solutions handle time-critical and delay-tolerant communication. Additionally a control method for the AGVs is proposed, which optimizes the communication and in turn increases the transport performance of the AGV fleet. Therefore, this work provides not only tools for the further research of industrial ad-hoc system, but also first implementations of ad-hoc systems which address many of the most pressing issues in industrial applica- tions. / Viele industrielle Automatisierungslösungen verwenden drahtlose Kommunikations- systeme und sind daher auf die Verfügbarkeit und Qualität des drahtlosen Kanals an- gewiesen. Gleichzeitig ist das drahtlose Medium stark belastet und die Gewährleis- tung der Verfügbarkeit der drahtlosen Kanäle wird zunehmends herrausfordernder. In dieser Arbeit wird gezeigt, dass Ad-hoc-Netzwerklösungen genutzt werden können, um neue Kommunikationskanäle bereitzustellen und die Leistung von mobilen Au- tomatisierungssystemen zu verbessern. Diese Ad-hoc-Netzwerklösungen können un- terschiedliche Kommunikationsstrategien bezeichnen. In all diesen Strategien wird der Peer-to-Peer (P2P)-Kanal zwischen zwei kommunizierenden Systemen verwendet statt Netzwerk-Infrastruktur. Diese Arbeit ist ein Schritt hin zur effektiven Implementierung von Kommunikations- technologien mit geringer Reichweite (z.B. Visible Light Communication (VLC), Radar- kommunikation, mmWave-Kommunikation) in der industriellen Anwendung. Die Im- plementierung von Infrastrukturnetzen mit diesen Technologien ist unrealistisch, da die geringe Kommunikationsreichweite eine hohe Anzahl von Access Points (APs) er- fordern würde um eine flächendeckende Bereitstellung von Kommunikationskanälen zu gewährleisten. Ad-hoc-Netzwerke hingegen benötigen keine Netzwerkinfrastruk- tur. In dieser Arbeit werden verschiedene Ad-hoc-Netzwerklösungen für den industri- ellenAnwendungsfallvorgestelltundWerkzeugeundModellefürderenUntersuchung vorgeschlagen. Der Hauptanwendungsfall, der in dieser Arbeit untersucht wird, sind Fahrerlose Trans- portSysteme (FTS) (fortführend als Automated Guided Vehicles (AGVs)) für industri- elle Anwendungen. Diese FTS fahren durch die Produktionsanlage um Kisten, Waren oder Werkzeuge zu transportieren oder um Mitarbeitern zu assistieren. In den meisten Implementierungen müssen sie Daten mit einer Central Control Unit (CCU) und unter- einander austauschen. Die Vorhersage, ob eine bestimmte Kommunikationstechnologie für eine Anwendung geeignet ist, ist sehr anspruchsvoll, da sowohl Anwendungen als auch Anforderungen sehr heterogen sind. Die präsentierten Modelle und Simulationswerkzeuge ermöglichen die Simulation der komplexen Interaktion von mobilen Robotern und drahtlosen Kommunikationsnetz- werken. Das Ziel ist die Vorhersage der Eigenschaften einer vernetzten FTS-Flotte. Mit den vorgestellten Werkzeugen wurden verschiedene Ad-hoc-Netzwerklösungen für industrielle Anwendungen mit FTS implementiert, getestet und untersucht. Die- se Kommunikationssysteme übertragen zeitkritische und verzögerungstolerante Nach- richten. Zusätzlich wird eine Steuerungsmethode für die FTS vorgeschlagen, die die KommunikationoptimiertunddamiteinhergehenddieTransportleistungderFTS-Flotte erhöht. Dieses Werk führt also nicht nur neue Werkzeuge ein um die Entwicklung in- dustrieller Ad-hoc Systeme zu ermöglichen, sondern schlägt auch einige Systeme für die kritischsten Kommunikationsprobleme industrieller Anwendungen vor.

Optimized implementations of blocking and nonblocking collective operations are most important for scalable high-performance applications. Offloading such collective operations into the communication layer can improve performance and asynchronous progression of the operations. However, it is most important that such offloading schemes remain flexible in order to support user-defined (sparse neighbor) collective communications. In this work we propose a design for a collective offload unit. Our hardware design is able to execute dependency graph based representations of collective functions. To cope with the scarcity of memory resources we designed a new point to point messaging protocol which does not need to store information about unexpected messages. The offload unit proposed in this thesis could be integrated into high performance networks such as EXTOLL. Our design achieves a clock frequency of 212 MHz on a Xilinx Virtex6 FPGA, while using less than 10% of the available logic slices and less than 30% of the available memory blocks. Due to the specialization of our design we can accelerate important tasks of the message passing framework, such as message matching by a factor of two, compared to a software implementation running on a CPU with a ten times higher clock speed.:1. Task Description 1.1. Theses 2. Introduction 2.1. Motivation 2.2. Outline of this Thesis 2.3. Related Work 2.3.1. NIC Based Packet Forwarding 2.3.2. Hardware Barrier Implementations 2.3.3. ConnectX2 CORE-Direct Collective Offload Support 2.3.4. Collective Offload Support in the Portals 4 API 2.4. Group Operation Assembly Language 2.4.1. GOAL API 2.4.2. Scratchpad Buffer 2.4.3. Schedule Execution 2.5. The EXTOLL Network 2.6. Field Programmable Gate Arrays 3. Dealing with Constrained Resources 3.1. Hardware Limitations 3.2. Common Collective Functions in GOAL 3.3. Schedule Representation for the Hardware GOAL Interpreter 3.4. Executing Large Schedules using a small amount of Memory 3.4.1. Limits of Previously Suggested Approaches 3.4.2. Testing for Deadlocks in Schedules 3.4.3. Transforming Process Local Schedules into Global Schedules 3.4.4. Predetermined Buffer Locations 3.5. Queueing Active Operations in Hardware 3.6. Designing a Low-Memory-Footprint Point to Point Protocol 3.6.1. Arrival Times 3.6.2. Eager Protocol 3.6.3. Rendezvous Protocol 3.6.4. A Protocol without an Unexpected Queue 3.7. Protocol Verification 3.7.1. Capabilities of the Model Checker SPIN 3.7.2. Modeling the Protocol 3.7.3. Limitations of the Basic Protocol 4. The Matching Problem 4.1. Matching on the Host CPU 4.2. Implementation Methodology 4.3. Matching Unit Interface 4.4. Matching Unit Implementation 4.4.1. Slot Management Unit 4.4.2. The Input Consumer 4.4.3. The Output Generator 4.4.4. The Matching Unit 4.5. Slot Management Unit for Non-synchronous Transfers 5. The GOAL Interpreter 5.1. Schedule Interpreter Design 5.1.1. The Active Queue 5.1.2. The Dependency Resolver 5.2. Transceiver Interface 5.3. The Starter 5.3.1. Starting Operations 5.3.2. Processing Incoming Packets 5.3.3. Incoming Non-synchronous Packets 5.3.4. Presorting the Active Queue 5.3.5. Arbitration Units 5.3.6. IN-Filter 5.3.7. Outcommand Manager 5.3.8. Non-synchronous Protocol 5.3.9. Send Protocol 5.3.10. Receive Protocol 5.3.11. Local Operations on FPGA 6 Evaluation 6.1. Performance Analysis 6.2. Future Work 6.3. Conclusions Bibliography

No description available.

During laparoscopic liver resection, the limited access to the organ, the small field of view and lack of palpation can obstruct a surgeon’s workflow. Automatic navigation systems could use the images from preoperative volumetric organ scans to help the surgeons find their target (tumors) and risk-structures (vessels) more efficiently. This requires the preoperative data to be fused (or registered) with the intraoperative scene in order to display information at the correct intraoperative position. One key challenge in this setting is the automatic estimation of the organ’s current intra-operative deformation, which is required in order to predict the position of internal structures. Parameterizing the many patient-specific unknowns (tissue properties, boundary conditions, interactions with other tissues, direction of gravity) is very difficult. Instead, this work explores how to employ deep neural networks to solve the registration problem in a data-driven manner. To this end, convolutional neural networks are trained on synthetic data to estimate an organ’s intraoperative displacement field and thus its current deformation. To drive this estimation, visible surface cues from the intraoperative camera view must be supplied to the networks. Since reliable surface features are very difficult to find, the networks are adapted to also find correspondences between the pre- and intraoperative liver geometry automatically. This combines the search for correspondences with the biomechanical behavior estimation and allows the networks to tackle the full non-rigid registration problem in one single step. The result is a model which can quickly predict the volume deformation of a liver, given only sparse surface information. The model combines the advantages of a physically accurate biomechanical simulation with the speed and powerful feature extraction capabilities of deep neural networks. To test the method intraoperatively, a registration pipeline is developed which constructs a map of the liver and its surroundings from the laparoscopic video and then uses the neural networks to fuse the preoperative volume data into this map. The deformed organ volume can then be rendered as an overlay directly onto the laparoscopic video stream. The focus of this pipeline is to be applicable to real surgery, where everything should be quick and non-intrusive. To meet these requirements, a SLAM system is used to localize the laparoscopic camera (avoiding setup of an external tracking system), various neural networks are used to quickly interpret the scene and semi-automatic tools let the surgeons guide the system. Beyond the concrete advantages of the data-driven approach for intraoperative registration, this work also demonstrates general benefits of training a registration system preoperatively on synthetic data. The method lets the engineer decide which values need to be known explicitly and which should be estimated implicitly by the networks, which opens the door to many new possibilities.:1 Introduction 1.1 Motivation 1.1.1 Navigated Liver Surgery 1.1.2 Laparoscopic Liver Registration 1.2 Challenges in Laparoscopic Liver Registration 1.2.1 Preoperative Model 1.2.2 Intraoperative Data 1.2.3 Fusion/Registration 1.2.4 Data 1.3 Scope and Goals of this Work 1.3.1 Data-Driven, Biomechanical Model 1.3.2 Data-Driven Non-Rigid Registration 1.3.3 Building a Working Prototype 2 State of the Art 2.1 Rigid Registration 2.2 Non-Rigid Liver Registration 2.3 Neural Networks for Simulation and Registration 3 Theoretical Background 3.1 Liver 3.2 Laparoscopic Liver Resection 3.2.1 Staging Procedure 3.3 Biomechanical Simulation 3.3.1 Physical Balance Principles 3.3.2 Material Models 3.3.3 Numerical Solver: The Finite Element Method (FEM) 3.3.4 The Lagrangian Specification 3.4 Variables and Data in Liver Registration 3.4.1 Observable 3.4.2 Unknowns 4 Generating Simulations of Deforming Organs 4.1 Organ Volume 4.2 Forces and Boundary Conditions 4.2.1 Surface Forces 4.2.2 Zero-Displacement Boundary Conditions 4.2.3 Surrounding Tissues and Ligaments 4.2.4 Gravity 4.2.5 Pressure 4.3 Simulation 4.3.1 Static Simulation 4.3.2 Dynamic Simulation 4.4 Surface Extraction 4.4.1 Partial Surface Extraction 4.4.2 Surface Noise 4.4.3 Partial Surface Displacement 4.5 Voxelization 4.5.1 Voxelizing the Liver Geometry 4.5.2 Voxelizing the Displacement Field 4.5.3 Voxelizing Boundary Conditions 4.6 Pruning Dataset - Removing Unwanted Results 4.7 Data Augmentation 5 Deep Neural Networks for Biomechanical Simulation 5.1 Training Data 5.2 Network Architecture 5.3 Loss Functions and Training 6 Deep Neural Networks for Non-Rigid Registration 6.1 Training Data 6.2 Architecture 6.3 Loss 6.4 Training 6.5 Mesh Deformation 6.6 Example Application 7 Intraoperative Prototype 7.1 Image Acquisition 7.2 Stereo Calibration 7.3 Image Rectification, Disparity- and Depth- estimation 7.4 Liver Segmentation 7.4.1 Synthetic Image Generation 7.4.2 Automatic Segmentation 7.4.3 Manual Segmentation Modifier 7.5 SLAM 7.6 Dense Reconstruction 7.7 Rigid Registration 7.8 Non-Rigid Registration 7.9 Rendering 7.10 Robotic Operating System 8 Evaluation 8.1 Evaluation Datasets 8.1.1 In-Silico 8.1.2 Phantom Torso and Liver 8.1.3 In-Vivo, Human, Breathing Motion 8.1.4 In-Vivo, Human, Laparoscopy 8.2 Metrics 8.2.1 Mean Displacement Error 8.2.2 Target Registration Error (TRE) 8.2.3 Champfer Distance 8.2.4 Volumetric Change 8.3 Evaluation of the Synthetic Training Data 8.4 Data-Driven Biomechanical Model (DDBM) 8.4.1 Amount of Intraoperative Surface 8.4.2 Dynamic Simulation 8.5 Volume to Surface Registration Network (V2S-Net) 8.5.1 Amount of Intraoperative Surface 8.5.2 Dependency on Initial Rigid Alignment 8.5.3 Registration Accuracy in Comparison to Surface Noise 8.5.4 Registration Accuracy in Comparison to Material Stiffness 8.5.5 Champfer-Distance vs. Mean Displacement Error 8.5.6 In-vivo, Human Breathing Motion 8.6 Full Intraoperative Pipeline 8.6.1 Intraoperative Reconstruction: SLAM and Intraoperative Map 8.6.2 Full Pipeline on Laparoscopic Human Data 8.7 Timing 9 Discussion 9.1 Intraoperative Model 9.2 Physical Accuracy 9.3 Limitations in Training Data 9.4 Limitations Caused by Difference in Pre- and Intraoperative Modalities 9.5 Ambiguity 9.6 Intraoperative Prototype 10 Conclusion 11 List of Publications List of Figures Bibliography

Finding bottlenecks and eliminating them to increase the overall flow of a network often appears in real world applications, such as production planning, factory layout, flow related physical approaches, and even cyber security. In many cases, several edges can form a bottleneck (cascaded bottlenecks). This work presents a visual analytics methodology to analyze these cascaded bottlenecks. The methodology consists of multiple steps: identification of bottlenecks, identification of potential improvements, communication of bottlenecks, interactive adaption of bottlenecks, and a feedback loop that allows users to adapt flow networks and their resulting bottlenecks until they are satisfied with the flow network configuration. To achieve this, the definition of a minimal cut is extended to identify network edges that form a (cascaded) bottleneck. To show the effectiveness of the presented approach, we applied the methodology to two flow network setups and show how the overall flow of these networks can be improved.