004 Datenverarbeitung; Informatik
Refine
Has Fulltext
- yes (285)
Year of publication
Document Type
- Journal article (127)
- Doctoral Thesis (80)
- Working Paper (37)
- Preprint (19)
- Conference Proceeding (9)
- Jahresbericht (5)
- Master Thesis (4)
- Report (3)
- Other (1)
Language
- English (257)
- German (27)
- Multiple languages (1)
Keywords
- virtual reality (16)
- Datennetz (14)
- Leistungsbewertung (13)
- Quran (8)
- Robotik (8)
- Koran (7)
- Mobiler Roboter (7)
- Text Mining (7)
- Autonomer Roboter (6)
- Simulation (6)
Institute
- Institut für Informatik (203)
- Theodor-Boveri-Institut für Biowissenschaften (29)
- Institut Mensch - Computer - Medien (17)
- Institut für deutsche Philologie (17)
- Institut für Klinische Epidemiologie und Biometrie (7)
- Rechenzentrum (7)
- Center for Computational and Theoretical Biology (4)
- Graduate School of Science and Technology (3)
- Medizinische Klinik und Poliklinik II (3)
- Institut für Funktionsmaterialien und Biofabrikation (2)
Schriftenreihe
Sonstige beteiligte Institutionen
- Cologne Game Lab (2)
- Birmingham City University (1)
- DATE Lab, KITE Research Insititute, University Health Network, Toronto, Canada (1)
- EMBL Heidelberg (1)
- INAF Padova, Italy (1)
- Jacobs University Bremen, Germany (1)
- Open University of the Netherlands (1)
- Servicezentrum Medizin-Informatik (Universitätsklinikum) (1)
- Social and Technological Systems (SaTS) lab, School of Art, Media, Performance and Design, York University, Toronto, Canada (1)
- TH Köln (1)
This paper describes the estimation of the body weight of a person in front of an RGB-D camera. A survey of different methods for body weight estimation based on depth sensors is given. First, an estimation of people standing in front of a camera is presented. Second, an approach based on a stream of depth images is used to obtain the body weight of a person walking towards a sensor. The algorithm first extracts features from a point cloud and forwards them to an artificial neural network (ANN) to obtain an estimation of body weight. Besides the algorithm for the estimation, this paper further presents an open-access dataset based on measurements from a trauma room in a hospital as well as data from visitors of a public event. In total, the dataset contains 439 measurements. The article illustrates the efficiency of the approach with experiments with persons lying down in a hospital, standing persons, and walking persons. Applicable scenarios for the presented algorithm are body weight-related dosing of emergency patients.
Interactive system for similarity-based inspection and assessment of the well-being of mHealth users
(2021)
Recent digitization technologies empower mHealth users to conveniently record their Ecological Momentary Assessments (EMA) through web applications, smartphones, and wearable devices. These recordings can help clinicians understand how the users' condition changes, but appropriate learning and visualization mechanisms are required for this purpose. We propose a web-based visual analytics tool, which processes clinical data as well as EMAs that were recorded through a mHealth application. The goals we pursue are (1) to predict the condition of the user in the near and the far future, while also identifying the clinical data that mostly contribute to EMA predictions, (2) to identify users with outlier EMA, and (3) to show to what extent the EMAs of a user are in line with or diverge from those users similar to him/her. We report our findings based on a pilot study on patient empowerment, involving tinnitus patients who recorded EMAs with the mHealth app TinnitusTips. To validate our method, we also derived synthetic data from the same pilot study. Based on this setting, results for different use cases are reported.
A key feature for Internet of Things (IoT) is to control what content is available to each user. To handle this access management, encryption schemes can be used. Due to the diverse usage of encryption schemes, there are various realizations of 1-to-1, 1-to-n, and n-to-n schemes in the literature. This multitude of encryption methods with a wide variety of properties presents developers with the challenge of selecting the optimal method for a particular use case, which is further complicated by the fact that there is no overview of existing encryption schemes. To fill this gap, we envision a cryptography encyclopedia providing such an overview of existing encryption schemes. In this survey paper, we take a first step towards such an encyclopedia by creating a sub-encyclopedia for secure group communication (SGC) schemes, which belong to the n-to-n category. We extensively surveyed the state-of-the-art and classified 47 different schemes. More precisely, we provide (i) a comprehensive overview of the relevant security features, (ii) a set of relevant performance metrics, (iii) a classification for secure group communication schemes, and (iv) workflow descriptions of the 47 schemes. Moreover, we perform a detailed performance and security evaluation of the 47 secure group communication schemes. Based on this evaluation, we create a guideline for the selection of secure group communication schemes.
Future broadband wireless networks should be able to support not only best effort traffic but also real-time traffic with strict Quality of Service (QoS) constraints. In addition, their available resources are scare and limit the number of users. To facilitate QoS guarantees and increase the maximum number of concurrent users, wireless networks require careful planning and optimization. In this monograph, we studied three aspects of performance optimization in wireless networks: resource optimization in WLAN infrastructure networks, quality of experience control in wireless mesh networks, and planning and optimization of wireless mesh networks. An adaptive resource management system is required to effectively utilize the limited resources on the air interface and to guarantee QoS for real-time applications. Thereby, both WLAN infrastructure and WLAN mesh networks have to be considered. An a-priori setting of the access parameters is not meaningful due to the contention-based medium access and the high dynamics of the system. Thus, a management system is required which dynamically adjusts the channel access parameters based on the network load. While this is sufficient for wireless infrastructure networks, interferences on neighboring paths and self-interferences have to be considered for wireless mesh networks. In addition, a careful channel allocation and route assignment is needed. Due to the large parameter space, standard optimization techniques fail for optimizing large wireless mesh networks. In this monograph, we reveal that biology-inspired optimization techniques, namely genetic algorithms, are well-suitable for the planning and optimization of wireless mesh networks. Although genetic algorithms generally do not always find the optimal solution, we show that with a good parameter set for the genetic algorithm, the overall throughput of the wireless mesh network can be significantly improved while still sharing the resources fairly among the users.
Die künstliche Intelligenz (KI) entwickelt sich rasant und hat bereits eindrucksvolle Erfolge zu verzeichnen, darunter übermenschliche Kompetenz in den meisten Spielen und vielen Quizshows, intelligente Suchmaschinen, individualisierte Werbung, Spracherkennung, -ausgabe und -übersetzung auf sehr hohem Niveau und hervorragende Leistungen bei der Bildverarbeitung, u. a. in der Medizin, der optischen Zeichenerkennung, beim autonomen Fahren, aber auch beim Erkennen von Menschen auf Bildern und Videos oder bei Deep Fakes für Fotos und Videos. Es ist zu erwarten, dass die KI auch in der Entscheidungsfindung Menschen übertreffen wird; ein alter Traum der Expertensysteme, der durch Lernverfahren, Big Data und Zugang zu dem gesammelten Wissen im Web in greifbare Nähe rückt. Gegenstand dieses Beitrags sind aber weniger die technischen Entwicklungen, sondern mögliche gesellschaftliche Auswirkungen einer spezialisierten, kompetenten KI für verschiedene Bereiche der autonomen, d. h. nicht nur unterstützenden Entscheidungsfindung: als Fußballschiedsrichter, in der Medizin, für richterliche Entscheidungen und sehr spekulativ auch im politischen Bereich. Dabei werden Vor- und Nachteile dieser Szenarien aus gesellschaftlicher Sicht diskutiert.
The Fifth Generation (5G) communication technology, its infrastructure and architecture, though already deployed in campus and small scale networks, is still undergoing continuous changes and research. Especially, in the light of future large scale deployments and industrial use cases, a detailed analysis of the performance and utilization with regard to latency and service times constraints is crucial. To this end, a fine granular investigation of the Network Function (NF) based core system and the duration for all the tasks performed by these services is necessary. This work presents the first steps towards analyzing the signaling traffic in 5G core networks, and introduces a tool to automatically extract sequence diagrams and service times for NF tasks from traffic traces.
This work proposes a novel approach to disperse dense transmission intervals and reduce bursty traffic patterns without the need for centralized control. Furthermore, by keeping the mechanism as close to the Long Range Wide Area Network (LoRaWAN) standard as possible the suggested mechanism can be deployed within existing networks and can even be co-deployed with other devices.
The introduction of new types of frequency spectrum in 6G technology facilitates the convergence of conventional mobile communications and radar functions. Thus, the mobile network itself becomes a versatile sensor system. This enables mobile network operators to offer a sensing service in addition to conventional data and telephony services. The potential benefits are expected to accrue to various stakeholders, including individuals, the environment, and society in general. The paper discusses technological development, possible integration, and use cases, as well as future development areas.
Background
Localization-based super-resolution microscopy resolves macromolecular structures down to a few nanometers by computationally reconstructing fluorescent emitter coordinates from diffraction-limited spots. The most commonly used algorithms are based on fitting parametric models of the point spread function (PSF) to a measured photon distribution. These algorithms make assumptions about the symmetry of the PSF and thus, do not work well with irregular, non-linear PSFs that occur for example in confocal lifetime imaging, where a laser is scanned across the sample. An alternative method for reconstructing sparse emitter sets from noisy, diffraction-limited images is compressed sensing, but due to its high computational cost it has not yet been widely adopted. Deep neural network fitters have recently emerged as a new competitive method for localization microscopy. They can learn to fit arbitrary PSFs, but require extensive simulated training data and do not generalize well. A method to efficiently fit the irregular PSFs from confocal lifetime localization microscopy combining the advantages of deep learning and compressed sensing would greatly improve the acquisition speed and throughput of this method.
Results
Here we introduce ReCSAI, a compressed sensing neural network to reconstruct localizations for confocal dSTORM, together with a simulation tool to generate training data. We implemented and compared different artificial network architectures, aiming to combine the advantages of compressed sensing and deep learning. We found that a U-Net with a recursive structure inspired by iterative compressed sensing showed the best results on realistic simulated datasets with noise, as well as on real experimentally measured confocal lifetime scanning data. Adding a trainable wavelet denoising layer as prior step further improved the reconstruction quality.
Conclusions
Our deep learning approach can reach a similar reconstruction accuracy for confocal dSTORM as frame binning with traditional fitting without requiring the acquisition of multiple frames. In addition, our work offers generic insights on the reconstruction of sparse measurements from noisy experimental data by combining compressed sensing and deep learning. We provide the trained networks, the code for network training and inference as well as the simulation tool as python code and Jupyter notebooks for easy reproducibility.
In the last 40 years, complexity theory has grown to a rich and powerful field in theoretical computer science. The main task of complexity theory is the classification of problems with respect to their consumption of resources (e.g., running time or required memory). To study the computational complexity (i.e., consumption of resources) of problems, similar problems are grouped into so called complexity classes. During the systematic study of numerous problems of practical relevance, no efficient algorithm for a great number of studied problems was found. Moreover, it was unclear whether such algorithms exist. A major breakthrough in this situation was the introduction of the complexity classes P and NP and the identification of hardest problems in NP. These hardest problems of NP are nowadays known as NP-complete problems. One prominent example of an NP-complete problem is the satisfiability problem of propositional formulas (SAT). Here we get a propositional formula as an input and it must be decided whether an assignment for the propositional variables exists, such that this assignment satisfies the given formula. The intensive study of NP led to numerous related classes, e.g., the classes of the polynomial-time hierarchy PH, P, #P, PP, NL, L and #L. During the study of these classes, problems related to propositional formulas were often identified to be complete problems for these classes. Hence some questions arise: Why is SAT so hard to solve? Are there modifications of SAT which are complete for other well-known complexity classes? In the context of these questions a result by E. Post is extremely useful. He identified and characterized all classes of Boolean functions being closed under superposition. It is possible to study problems which are connected to generalized propositional logic by using this result, which was done in this thesis. Hence, many different problems connected to propositional logic were studied and classified with respect to their computational complexity, clearing the borderline between easy and hard problems.
Practical optimization problems often comprise several incomparable and conflicting objectives. When booking a trip using several means of transport, for instance, it should be fast and at the same time not too expensive. The first part of this thesis is concerned with the algorithmic solvability of such multiobjective optimization problems. Several solution notions are discussed and compared with respect to their difficulty. Interestingly, these solution notions are always equally difficulty for a single-objective problem and they differ considerably already for two objectives (unless P = NP). In this context, the difference between search and decision problems is also investigated in general. Furthermore, new and improved approximation algorithms for several variants of the traveling salesperson problem are presented. Using tools from discrepancy theory, a general technique is developed that helps to avoid an obstacle that is often hindering in multiobjective approximation: The problem of combining two solutions such that the new solution is balanced in all objectives and also mostly retains the structure of the original solutions. The second part of this thesis is dedicated to several aspects of systems of equations for (formal) languages. Firstly, conjunctive and Boolean grammars are studied, which are extensions of context-free grammars by explicit intersection and complementation operations, respectively. Among other results, it is shown that one can considerably restrict the union operation on conjunctive grammars without changing the generated language. Secondly, certain circuits are investigated whose gates do not compute Boolean values but sets of natural numbers. For these circuits, the equivalence problem is studied, i.\,e.\ the problem of deciding whether two given circuits compute the same set or not. It is shown that, depending on the allowed types of gates, this problem is complete for several different complexity classes and can thus be seen as a parametrized) representative for all those classes.
Learning is a central component of human life and essential for personal development. Therefore, utilizing new technologies in the learning context and exploring their combined potential are considered essential to support self-directed learning in a digital age. A learning environment can be expanded by various technical and content-related aspects. Gamification in the form of elements from video games offers a potential concept to support the learning process. This can be supplemented by technology-supported learning. While the use of tablets is already widespread in the learning context, the integration of a social robot can provide new perspectives on the learning process. However, simply adding new technologies such as social robots or gamification to existing systems may not automatically result in a better learning environment. In the present study, game elements as well as a social robot were integrated separately and conjointly into a learning environment for basic Spanish skills, with a follow-up on retained knowledge. This allowed us to investigate the respective and combined effects of both expansions on motivation, engagement and learning effect. This approach should provide insights into the integration of both additions in an adult learning context. We found that the additions of game elements and the robot did not significantly improve learning, engagement or motivation. Based on these results and a literature review, we outline relevant factors for meaningful integration of gamification and social robots in learning environments in adult learning.
Lightning has fascinated humanity since the beginning of our existence. Different types of lightning like sprites and blue jets were discovered, and many more are theorized. However, it is very likely that these phenomena are not exclusive to our home planet. Venus’s dense and active atmosphere is a place where lightning is to be expected. Missions like Venera, Pioneer, and Galileo have carried instruments to measure electromagnetic activity. These measurements have indeed delivered results. However, these results are not clear. They could be explained by other effects like cosmic rays, plasma noise, or spacecraft noise. Furthermore, these lightning seem different from those we know from our home planet. In order to tackle these issues, a different approach to measurement is proposed. When multiple devices in different spacecraft or locations can measure the same atmospheric discharge, most other explanations become increasingly less likely. Thus, the suggested instrument and method of VELEX incorporates multiple spacecraft. With this approach, the question about the existence of lightning on Venus could be settled.
The first step towards aerial planetary exploration has been made. Ingenuity shows extremely promising results, and new missions are already underway. Rotorcraft are capable of flight. This capability could be utilized to support the last stages of Entry, Descent, and Landing. Thus, mass and complexity could be scaled down.
Autorotation is one method of descent. It describes unpowered descent and landing, typically performed by helicopters in case of an engine failure. MAPLE is suggested to test these procedures and understand autorotation on other planets. In this series of experiments, the Ingenuity helicopter is utilized. Ingenuity would autorotate a ”mid-air-landing” before continuing with normal flight. Ultimately, the collected data shall help to understand autorotation on Mars and its utilization for interplanetary exploration.
Background: Since the replication crisis, standardization has become even more important in psychological science and neuroscience. As a result, many methods are being reconsidered, and researchers’ degrees of freedom in these methods are being discussed as a potential source of inconsistencies across studies.
New Method: With the aim of addressing these subjectivity issues, we have been working on a tutorial-like EEG (pre-)processing pipeline to achieve an automated method based on the semi-automated analysis proposed by Delorme and Makeig.
Results: Two scripts are presented and explained step-by-step to perform basic, informed ERP and frequency-domain analyses, including data export to statistical programs and visual representations of the data. The open-source software EEGlab in MATLAB is used as the data handling platform, but scripts based on code provided by Mike Cohen (2014) are also included.
Comparison with existing methods: This accompanying tutorial-like article explains and shows how the processing of our automated pipeline affects the data and addresses, especially beginners in EEG-analysis, as other (pre)-processing chains are mostly targeting rather informed users in specialized areas or only parts of a complete procedure. In this context, we compared our pipeline with a selection of existing approaches.
Conclusion: The need for standardization and replication is evident, yet it is equally important to control the plausibility of the suggested solution by data exploration. Here, we provide the community with a tool to enhance the understanding and capability of EEG-analysis. We aim to contribute to comprehensive and reliable analyses for neuro-scientific research.
Inside 2003: IT-Sicherheit
(2003)
The DAEDALUS mission concept aims at exploring and characterising the entrance and initial part of Lunar lava tubes within a compact, tightly integrated spherical robotic device, with a complementary payload set and autonomous capabilities.
The mission concept addresses specifically the identification and characterisation of potential resources for future ESA exploration, the local environment of the subsurface and its geologic and compositional structure.
A sphere is ideally suited to protect sensors and scientific equipment in rough, uneven environments.
It will house laser scanners, cameras and ancillary payloads.
The sphere will be lowered into the skylight and will explore the entrance shaft, associated caverns and conduits. Lidar (light detection and ranging) systems produce 3D models with high spatial accuracy independent of lighting conditions and visible features.
Hence this will be the primary exploration toolset within the sphere.
The additional payload that can be accommodated in the robotic sphere consists of camera systems with panoramic lenses and scanners such as multi-wavelength or single-photon scanners.
A moving mass will trigger movements.
The tether for lowering the sphere will be used for data communication and powering the equipment during the descending phase.
Furthermore, the connector tether-sphere will host a WIFI access point, such that data of the conduit can be transferred to the surface relay station. During the exploration phase, the robot will be disconnected from the cable, and will use wireless communication.
Emergency autonomy software will ensure that in case of loss of communication, the robot will continue the nominal mission.
Modern immersive multimodal technologies enable the learners to completely get immersed in various learning situations in a way that feels like experiencing an authentic learning environment. These environments also allow the collection of multimodal data, which can be used with artificial intelligence to further improve the immersion and learning outcomes. The use of artificial intelligence has been widely explored for the interpretation of multimodal data collected from multiple sensors, thus giving insights to support learners’ performance by providing personalised feedback. In this paper, we present a conceptual approach for creating immersive learning environments, integrated with multi-sensor setup to help learners improve their psychomotor skills in a remote setting.
In this thesis, we present novel approaches for formation driving of nonholonomic robots and optimal trajectory planning to reach a target region. The methods consider a static known map of the environment as well as unknown and dynamic obstacles detected by sensors of the formation. The algorithms are based on leader following techniques, where the formation of car-like robots is maintained in a shape determined by curvilinear coordinates. Beyond this, the general methods of formation driving are specialized and extended for an application of airport snow shoveling. Detailed descriptions of the algorithms complemented by relevant stability and convergence studies will be provided in the following chapters. Furthermore, discussions of the applicability will be verified by various simulations in existing robotic environments and also by a hardware experiment.
With the progress in robotics research the human machine interfaces reach more and more the status of being the major limiting factor for the overall system performance of a system for remote navigation and coordination of robots. In this monograph it is elaborated how mixed reality technologies can be applied for the user interfaces in order to increase the overall system performance. Concepts, technologies, and frameworks are developed and evaluated in user studies which enable for novel user-centered approaches to the design of mixed-reality user interfaces for remote robot operation. Both the technological requirements and the human factors are considered to achieve a consistent system design. Novel technologies like 3D time-of-flight cameras are investigated for the application in the navigation tasks and for the application in the developed concept of a generic mixed reality user interface. In addition it is shown how the network traffic of a video stream can be shaped on application layer in order to reach a stable frame rate in dynamic networks. The elaborated generic mixed reality framework enables an integrated 3D graphical user interface. The realized spatial integration and visualization of available information reduces the demand for mental transformations for the human operator and supports the use of immersive stereo devices. The developed concepts make also use of the fact that local robust autonomy components can be realized and thus can be incorporated as assistance systems for the human operators. A sliding autonomy concept is introduced combining force and visual augmented reality feedback. The force feedback component allows rendering the robot's current navigation intention to the human operator, such that a real sliding autonomy with seamless transitions is achieved. The user-studies prove the significant increase in navigation performance by application of this concept. The generic mixed reality user interface together with robust local autonomy enables a further extension of the teleoperation system to a short-term predictive mixed reality user interface. With the presented concept of operation, it is possible to significantly reduce the visibility of system delays for the human operator. In addition, both advantageous characteristics of a 3D graphical user interface for robot teleoperation- an exocentric view and an augmented reality view – can be combined.
The increased occurrence of Software-Defined-Networking (SDN) not only improves the dynamics and maintenance of network architectures, but also opens up new use cases and application possibilities. Based on these observations, we propose a new network topology consisting of a star and a ring topology. This hybrid topology will be called wheel topology in this paper. We have considered the static characteristics of the wheel topology and compare them with known other topologies.
To enable a sustainable supply of chemicals, novel biotechnological solutions are required that replace the reliance on fossil resources. One potential solution is to utilize tailored biosynthetic modules for the metabolic conversion of CO2 or organic waste to chemicals and fuel by microorganisms. Currently, it is challenging to commercialize biotechnological processes for renewable chemical biomanufacturing because of a lack of highly active and specific biocatalysts. As experimental methods to engineer biocatalysts are time- and cost-intensive, it is important to establish efficient and reliable computational tools that can speed up the identification or optimization of selective, highly active, and stable enzyme variants for utilization in the biotechnological industry. Here, we review and suggest combinations of effective state-of-the-art software and online tools available for computational enzyme engineering pipelines to optimize metabolic pathways for the biosynthesis of renewable chemicals. Using examples relevant for biotechnology, we explain the underlying principles of enzyme engineering and design and illuminate future directions for automated optimization of biocatalysts for the assembly of synthetic metabolic pathways.
Currently, we observe a strong growth of services and applications, which use the Internet for data transport. However, the network requirements of these applications differ significantly. This makes network management difficult, since it complicated to separate network flows into application classes without inspecting application layer data. Network virtualization is a promising solution to this problem. It enables running different virtual network on the same physical substrate. Separating networks based on the service supported within allows controlling each network according to the specific needs of the application. The aim of such a network control is to optimize the user perceived quality as well as the cost efficiency of the data transport. Furthermore, network virtualization abstracts the network functionality from the underlying implementation and facilitates the split of the currently tightly integrated roles of Internet Service Provider and network owner. Additionally, network virtualization guarantees that different virtual networks run on the same physical substrate do not interfere with each other. This thesis discusses different aspects of the network virtualization topic. It is focused on how to manage and control a virtual network to guarantee the best Quality of Experience for the user. Therefore, a top-down approach is chosen. Starting with use cases of virtual networks, a possible architecture is derived and current implementation options based on hardware virtualization are explored. In the following, this thesis focuses on assessing the Quality of Experience perceived by the user and how it can be optimized on application layer. Furthermore, options for measuring and monitoring significant network parameters of virtual networks are considered.
In today's Internet, services are very different in their requirements on the underlying transport network. In the future, this diversity will increase and it will be more difficult to accommodate all services in a single network. A possible approach to cope with this diversity within future networks is the introduction of support for running isolated networks for different services on top of a single shared physical substrate. This would also enable easy network management and ensure an economically sound operation. End-customers will readily adopt this approach as it enables new and innovative services without being expensive. In order to arrive at a concept that enables this kind of network, it needs to be designed around and constantly checked against realistic use cases. In this contribution, we present three use cases for future networks. We describe functional blocks of a virtual network architecture, which are necessary to support these use cases within the network. Furthermore, we discuss the interfaces needed between the functional blocks and consider standardization issues that arise in order to achieve a global consistent control and management structure of virtual networks.
Neural networks have to capture mathematical relationships in order to learn various tasks. They approximate these relations implicitly and therefore often do not generalize well. The recently proposed Neural Arithmetic Logic Unit (NALU) is a novel neural architecture which is able to explicitly represent the mathematical relationships by the units of the network to learn operations such as summation, subtraction or multiplication. Although NALUs have been shown to perform well on various downstream tasks, an in-depth analysis reveals practical shortcomings by design, such as the inability to multiply or divide negative input values or training stability issues for deeper networks. We address these issues and propose an improved model architecture. We evaluate our model empirically in various settings from learning basic arithmetic operations to more complex functions. Our experiments indicate that our model solves stability issues and outperforms the original NALU model in means of arithmetic precision and convergence.
Background: Current imaging methods such as Magnetic Resonance Imaging (MRI), Confocal microscopy, Electron Microscopy (EM) or Selective Plane Illumination Microscopy (SPIM) yield three-dimensional (3D) data sets in need of appropriate computational methods for their analysis. The reconstruction, segmentation and registration are best approached from the 3D representation of the data set. Results: Here we present a platform-independent framework based on Java and Java 3D for accelerated rendering of biological images. Our framework is seamlessly integrated into ImageJ, a free image processing package with a vast collection of community-developed biological image analysis tools. Our framework enriches the ImageJ software libraries with methods that greatly reduce the complexity of developing image analysis tools in an interactive 3D visualization environment. In particular, we provide high-level access to volume rendering, volume editing, surface extraction, and image annotation. The ability to rely on a library that removes the low-level details enables concentrating software development efforts on the algorithm implementation parts. Conclusions: Our framework enables biomedical image software development to be built with 3D visualization capabilities with very little effort. We offer the source code and convenient binary packages along with extensive documentation at http://3dviewer.neurofly.de.
The field of small satellite formations and constellations attracted growing attention, based on recent advances in small satellite engineering. The utilization of distributed space systems allows the realization of innovative applications and will enable improved temporal and spatial resolution in observation scenarios. On the other side, this new paradigm imposes a variety of research challenges. In this monograph new networking concepts for space missions are presented, using networks of ground stations. The developed approaches combine ground station resources in a coordinated way to achieve more robust and efficient communication links. Within this thesis, the following topics were elaborated to improve the performance in distributed space missions: Appropriate scheduling of contact windows in a distributed ground system is a necessary process to avoid low utilization of ground stations. The theoretical basis for the novel concept of redundant scheduling was elaborated in detail. Additionally to the presented algorithm was a scheduling system implemented, its performance was tested extensively with real world scheduling problems. In the scope of data management, a system was developed which autonomously synchronizes data frames in ground station networks and uses this information to detect and correct transmission errors. The system was validated with hardware in the loop experiments, demonstrating the benefits of the developed approach.
The thesis looks at the question asking for the computability of the dot-depth of star-free regular languages. Here one has to determine for a given star-free regular language the minimal number of alternations between concatenation on one hand, and intersection, union, complement on the other hand. This question was first raised in 1971 (Brzozowski/Cohen) and besides the extended star-heights problem usually refered to as one of the most difficult open questions on regular languages. The dot-depth problem can be captured formally by hierarchies of classes of star-free regular languages B(0), B(1/2), B(1), B(3/2),... and L(0), L(1/2), L(1), L(3/2),.... which are defined via alternating the closure under concatenation and Boolean operations, beginning with single alphabet letters. Now the question of dot-depth is the question whether these hierarchy classes have decidable membership problems. The thesis makes progress on this question using the so-called forbidden pattern approach: Classes of regular languages are characterized in terms of patterns in finite automata (subgraphs in the transition graph) that are not allowed. Such a characterization immediately implies the decidability of the respective class, since the absence of a certain pattern in a given automaton can be effectively verified. Before this work, the decidability of B(0), B(1/2), B(1) and L(0), L(1/2), L(1), L(3/2) were known. Here a detailed study of these classes with help of forbidden patterns is given which leads to new insights into their inner structure. Furthermore, the decidability of B(3/2) is proven. Based on these results a theory of pattern iteration is developed which leads to the introduction of two new hierarchies of star-free regular languages. These hierarchies are decidable on one hand, on the other hand they are in close connection to the classes B(n) and L(n). It remains an open question here whether they may in fact coincide. Some evidence is given in favour of this conjecture which opens a new way to attack the dot-depth problem. Moreover, it is shown that the class L(5/2) is decidable in the restricted case of a two-letter alphabet.
Tardigrades have fascinated researchers for more than 300 years because of their extraordinary capability to undergo cryptobiosis and survive extreme environmental conditions. However, the survival mechanisms of tardigrades are still poorly understood mainly due to the absence of detailed knowledge about the proteome and genome of these organisms. Our study was intended to provide a basis for the functional characterization of expressed proteins in different states of tardigrades. High-throughput, high-accuracy proteomics in combination with a newly developed tardigrade specific protein database resulted in the identification of more than 3000 proteins in three different states: early embryonic state and adult animals in active and anhydrobiotic state. This comprehensive proteome resource includes protein families such as chaperones, antioxidants, ribosomal proteins, cytoskeletal proteins, transporters, protein channels, nutrient reservoirs, and developmental proteins. A comparative analysis of protein families in the different states was performed by calculating the exponentially modified protein abundance index which classifies proteins in major and minor components. This is the first step to analyzing the proteins involved in early embryonic development, and furthermore proteins which might play an important role in the transition into the anhydrobiotic state.
Consider the situation where two or more images are taken from the same object. After taking the first image, the object is moved or rotated so that the second recording depicts it in a different manner. Additionally, take heed of the possibility that the imaging techniques may have also been changed. One of the main problems in image processing is to determine the spatial relation between such images. The corresponding process of finding the spatial alignment is called “registration”. In this work, we study the optimization problem which corresponds to the registration task. Especially, we exploit the Lie group structure of the set of transformations to construct efficient, intrinsic algorithms. We also apply the algorithms to medical registration tasks. However, the methods developed are not restricted to the field of medical image processing. We also have a closer look at more general forms of optimization problems and show connections to related tasks.
Globale Selbstlokalisation autonomer mobiler Roboter - Ein Schlüsselproblem der Service-Robotik
(2003)
Die Dissertation behandelt die Problemstellung der globalen Selbstlokalisation autonomer mobiler Roboter, welche folgendermaßen beschrieben werden kann: Ein mobiler Roboter, eingesetzt in einem Gebäude, kann unter Umständen das Wissen über seinen Standort verlieren. Man geht nun davon aus, dass dem Roboter eine Gebäudekarte als Modell zur Verfügung steht. Mit Hilfe eines Laser-Entfernungsmessers kann das mobile Gerät neue Informationen aufnehmen und damit bei korrekter Zuordnung zur Modellkarte geeignete hypothetische Standorte ermitteln. In der Regel werden diese Positionen aber mehrdeutig sein. Indem sich der Roboter intelligent in seiner Einsatzumgebung bewegt, kann er die ursprünglichen Sensordaten verifizieren und ermittelt im besten Fall seine tatsächliche Position.Für diese Problemstellung wird ein neuer Lösungsansatz in Theorie und Praxis präsentiert, welcher die jeweils aktuelle lokale Karte und damit alle Sensordaten mittels feature-basierter Matchingverfahren auf das Modell der Umgebung abbildet. Ein Explorationsalgorithmus bewegt den Roboter während der Bewegungsphase autonom zu Sensorpunkten, welche neue Informationen bereitstellen. Während der Bewegungsphase werden dabei die bisherigen hypothetischen Positionen bestärkt oder geschwächt, sodaß nach kurzer Zeit eine dominante Position, die tatsächliche Roboterposition,übrigbleibt.
Der große Vorteil eines q-Gramm Indexes liegt darin, dass es möglich ist beliebige Zeichenketten in einer Dokumentensammlung zu suchen. Ein Nachteil jedoch liegt darin, dass bei größer werdenden Datenmengen dieser Index dazu neigt, sehr groß zu werden, was mit einem deutlichem Leistungsabfall verbunden ist. In dieser Arbeit wird eine neuartige Technik vorgestellt, die die Leistung eines q-Gramm Indexes mithilfe zusätzlicher M-Matrizen für jedes q-Gramm und durch die Kombination mit einem invertierten Index erhöht. Eine M-Matrix ist eine Bit-Matrix, die Informationen über die Positionen eines q-Gramms enthält. Auch bei der Kombination von zwei oder mehreren Q-Grammen bieten diese M-Matrizen Informationen über die Positionen der Kombination. Dies kann verwendet werden, um die Komplexität der Zusammenführung der q-Gramm Trefferlisten für eine gegebene Suchanfrage zu reduzieren und verbessert die Leistung des n-Gramm-invertierten Index. Die Kombination mit einem termbasierten invertierten Index beschleunigt die durchschnittliche Suchzeit zusätzlich und vereint die Vorteile beider Index-Formate. Redundante Informationen werden in dem q-Gramm Index reduziert und weitere Funktionalität hinzugefügt, wie z.B. die Bewertung von Treffern nach Relevanz, die Möglichkeit, nach Konzepten zu suchen oder Indexpartitionierungen nach Wichtigkeit der enthaltenen Terme zu erstellen.
In time-sensitive networks (TSN) based on 802.1Qbv, i.e., the time-aware Shaper (TAS) protocol, precise transmission schedules and, paths are used to ensure end-to-end deterministic communication. Such resource reservations for data flows are usually established at the startup time of an application and remain untouched until the flow ends. There is no way to migrate existing flows easily to alternative paths without inducing additional delay or wasting resources. Therefore, some of the new flows cannot be embedded due to capacity limitations on certain links which leads to sub-optimal flow assignment. As future networks will need to support a large number of lowlatency flows, accommodating new flows at runtime and adapting existing flows accordingly becomes a challenging problem. In this extended abstract we summarize a previously published paper of us [1]. We combine software-defined networking (SDN), which provides better control of network flows, with TSN to be able to seamlessly migrate time-sensitive flows. For that, we formulate an optimization problem and propose different dynamic path configuration strategies under deterministic communication requirements. Our simulation results indicate that regularly reconfiguring the flow assignments can improve the latency of time-sensitive flows and can increase the number of flows embedded in the network around 4% in worst-case scenarios while still satisfying individual flow deadlines.
The strict restrictions introduced by the COVID-19 lockdowns, which started from March 2020, changed people’s daily lives and habits on many different levels. In this work, we investigate the impact of the lockdown on the communication behavior in the mobile instant messaging application WhatsApp. Our evaluations are based on a large dataset of 2577 private chat histories with 25,378,093 messages from 51,973 users. The analysis of the one-to-one and group conversations confirms that the lockdown severely altered the communication in WhatsApp chats compared to pre-pandemic time ranges. In particular, we observe short-term effects, which caused an increased message frequency in the first lockdown months and a shifted communication activity during the day in March and April 2020. Moreover, we also see long-term effects of the ongoing pandemic situation until February 2021, which indicate a change of communication behavior towards more regular messaging, as well as a persisting change in activity during the day. The results of our work show that even anonymized chat histories can tell us a lot about people’s behavior and especially behavioral changes during the COVID-19 pandemic and thus are of great relevance for behavioral researchers. Furthermore, looking at the pandemic from an Internet provider perspective, these insights can be used during the next pandemic, or if the current COVID-19 situation worsens, to adapt communication networks to the changed usage behavior early on and thus avoid network congestion.
Group-based communication is a highly popular communication paradigm, which is especially prominent in mobile instant messaging (MIM) applications, such as WhatsApp. Chat groups in MIM applications facilitate the sharing of various types of messages (e.g., text, voice, image, video) among a large number of participants. As each message has to be transmitted to every other member of the group, which multiplies the traffic, this has a massive impact on the underlying communication networks. However, most chat groups are private and network operators cannot obtain deep insights into MIM communication via network measurements due to end-to-end encryption. Thus, the generation of traffic is not well understood, given that it depends on sizes of communication groups, speed of communication, and exchanged message types. In this work, we provide a huge data set of 5,956 private WhatsApp chat histories, which contains over 76 million messages from more than 117,000 users. We describe and model the properties of chat groups and users, and the communication within these chat groups, which gives unprecedented insights into private MIM communication. In addition, we conduct exemplary measurements for the most popular message types, which empower the provided models to estimate the traffic over time in a chat group.
To deliver the best user experience (UX), the human-centered design cycle (HCDC) serves as a well-established guideline to application developers. However, it does not yet cover network-specific requirements, which become increasingly crucial, as most applications deliver experience over the Internet. The missing network-centric view is provided by Quality of Experience (QoE), which could team up with UX towards an improved overall experience. By considering QoE aspects during the development process, it can be achieved that applications become network-aware by design. In this paper, the Quality of Experience Centered Design Cycle (QoE-CDC) is proposed, which provides guidelines on how to design applications with respect to network-specific requirements and QoE. Its practical value is showcased for popular application types and validated by outlining the design of a new smartphone application. We show that combining HCDC and QoE-CDC will result in an application design, which reaches a high UX and avoids QoE degradation.
State Management at line rate is crucial for critical applications in next-generation networks. P4 is a language used in software-defined networking to program the data plane. The data plane can profit in many circumstances when it is allowed to manage its state without any detour over a controller. This work is based on a previous study by investigating the potential and performance of add-on-miss insertions of state by the data plane. The state keeping capabilities of P4 are limited regarding the amount of data and the update frequency. We follow the tentative specification of an upcoming portable-NIC-architecture and implement these changes into the software P4 target T4P4S. We show that insertions are possible with only a slight overhead compared to lookups and evaluate the influence of the rate of insertions on their latency.
Studies investigating the correlates of immune protection against Yersinia infection have established that both humoral and cell mediated immune responses are required for the comprehensive protection. In our previous study, we established that the bivalent fusion protein (rVE) comprising immunologically active regions of Y pestis LcrV (100-270 aa) and YopE (50-213 aa) proteins conferred complete passive and active protection against lethal Y enterocolitica 8081 challenge. In the present study, cohort of BALB/c mice immunized with rVE or its component proteins rV, rE were assessed for cell mediated immune responses and memory immune protection against Y enterocolitica 8081 rVE immunization resulted in extensive proliferation of both CD4 and CD8 T cell subsets; significantly high antibody titer with balanced IgG1: IgG2a/IgG2b isotypes (1:1 ratio) and up regulation of both Th1 (INF-\(\alpha\), IFN-\(\gamma\), IL 2, and IL 12) and Th2 (IL 4) cytokines. On the other hand, rV immunization resulted in Th2 biased IgG response (11:1 ratio) and proliferation of CD4+ T-cell; rE group of mice exhibited considerably lower serum antibody titer with predominant Th1 response (1:3 ratio) and CD8+ T-cell proliferation. Comprehensive protection with superior survival (100%) was observed among rVE immunized mice when compared to the significantly lower survival rates among rE (37.5%) and rV (25%) groups when IP challenged with Y enterocolitica 8081 after 120 days of immunization. Findings in this and our earlier studies define the bivalent fusion protein rVE as a potent candidate vaccine molecule with the capability to concurrently stimulate humoral and cell mediated immune responses and a proof of concept for developing efficient subunit vaccines against Gram negative facultative intracellular bacterial pathogens.
The holy grail of structural biology is to study a protein in situ, and this goal has been fast approaching since the resolution revolution and the achievement of atomic resolution. A cell's interior is not a dilute environment, and proteins have evolved to fold and function as needed in that environment; as such, an investigation of a cellular component should ideally include the full complexity of the cellular environment. Imaging whole cells in three dimensions using electron cryotomography is the best method to accomplish this goal, but it comes with a limitation on sample thickness and produces noisy data unamenable to direct analysis. This thesis establishes a novel workflow to systematically analyse whole-cell electron cryotomography data in three dimensions and to find and identify instances of protein complexes in the data to set up a determination of their structure and identity for success. Mycoplasma pneumoniae is a very small parasitic bacterium with fewer than 700 protein-coding genes, is thin enough and small enough to be imaged in large quantities by electron cryotomography, and can grow directly on the grids used for imaging, making it ideal for exploratory studies in structural proteomics. As part of the workflow, a methodology for training deep-learning-based particle-picking models is established.
As a proof of principle, a dataset of whole-cell Mycoplasma pneumoniae tomograms is used with this workflow to characterize a novel membrane-associated complex observed in the data. Ultimately, 25431 such particles are picked from 353 tomograms and refined to a density map with a resolution of 11 Å. Making good use of orthogonal datasets to filter search space and verify results, structures were predicted for candidate proteins and checked for suitable fit in the density map. In the end, with this approach, nine proteins were found to be part of the complex, which appears to be associated with chaperone activity and interact with translocon machinery.
Visual proteomics refers to the ultimate potential of in situ electron cryotomography: the comprehensive interpretation of tomograms. The workflow presented here is demonstrated to help in reaching that potential.
We consider competitive location problems where two competing providers place their facilities sequentially and users can decide between the competitors. We assume that both competitors act non-cooperatively and aim at maximizing their own benefits. We investigate the complexity and approximability of such problems on graphs, in particular on simple graph classes such as trees and paths. We also develop fast algorithms for single competitive location problems where each provider places a single facilty. Voting location, in contrast, aims at identifying locations that meet social criteria. The provider wants to satisfy the users (customers) of the facility to be opened. In general, there is no location that is favored by all users. Therefore, a satisfactory compromise has to be found. To this end, criteria arising from voting theory are considered. The solution of the location problem is understood as the winner of a virtual election among the users of the facilities, in which the potential locations play the role of the candidates and the users represent the voters. Competitive and voting location problems turn out to be closely related.
In the future Internet, the people-centric communication paradigm will be complemented by a ubiquitous communication among people and devices, or even a communication between devices. This comes along with the need for a more flexible, cheap, widely available Internet access. Two types of wireless networks are considered most appropriate for attaining those goals. While wireless sensor networks (WSNs) enhance the Internet’s reach by providing data about the properties of the environment, wireless mesh networks (WMNs) extend the Internet access possibilities beyond the wired backbone. This monograph contains four chapters which present modeling and optimization methods for WSNs and WMNs. Minimizing energy consumptions is the most important goal of WSN optimization and the literature consequently provides countless energy consumption models. The first part of the monograph studies to what extent the used energy consumption model influences the outcome of analytical WSN optimizations. These considerations enable the second contribution, namely overcoming the problems on the way to a standardized energy-efficient WSN communication stack based on IEEE 802.15.4 and ZigBee. For WMNs both problems are of minor interest whereas the network performance has a higher weight. The third part of the work, therefore, presents algorithms for calculating the max-min fair network throughput in WMNs with multiple link rates and Internet gateway. The last contribution of the monograph investigates the impact of the LRA concept which proposes to systematically assign more robust link rates than actually necessary, thereby allowing to exploit the trade-off between spatial reuse and per-link throughput. A systematical study shows that a network-wide slightly more conservative LRA than necessary increases the throughput of a WMN where max-min fairness is guaranteed. It moreover turns out that LRA is suitable for increasing the performance of a contention-based WMN and is a valuable optimization tool.
Recently, several classifiers that combine primary tumor data, like gene expression data, and secondary data sources, such as protein-protein interaction networks, have been proposed for predicting outcome in breast cancer. In these approaches, new composite features are typically constructed by aggregating the expression levels of several genes. The secondary data sources are employed to guide this aggregation. Although many studies claim that these approaches improve classification performance over single genes classifiers, the gain in performance is difficult to assess. This stems mainly from the fact that different breast cancer data sets and validation procedures are employed to assess the performance. Here we address these issues by employing a large cohort of six breast cancer data sets as benchmark set and by performing an unbiased evaluation of the classification accuracies of the different approaches. Contrary to previous claims, we find that composite feature classifiers do not outperform simple single genes classifiers. We investigate the effect of (1) the number of selected features; (2) the specific gene set from which features are selected; (3) the size of the training set and (4) the heterogeneity of the data set on the performance of composite feature and single genes classifiers. Strikingly, we find that randomization of secondary data sources, which destroys all biological information in these sources, does not result in a deterioration in performance of composite feature classifiers. Finally, we show that when a proper correction for gene set size is performed, the stability of single genes sets is similar to the stability of composite feature sets. Based on these results there is currently no reason to prefer prognostic classifiers based on composite features over single genes classifiers for predicting outcome in breast cancer.
Virtual environments (VEs) can evoke and support emotions, as experienced when playing emotionally arousing games. We theoretically approach the design of fear and joy evoking VEs based on a literature review of empirical studies on virtual and real environments as well as video games’ reviews and content analyses. We define the design space and identify central design elements that evoke specific positive and negative emotions. Based on that, we derive and present guidelines for emotion-inducing VE design with respect to design themes, colors and textures, and lighting configurations. To validate our guidelines in two user studies, we 1) expose participants to 360° videos of VEs designed following the individual guidelines and 2) immerse them in a neutral, positive and negative emotion-inducing VEs combining all respective guidelines in Virtual Reality. The results support our theoretically derived guidelines by revealing significant differences in terms of fear and joy induction.
Climate models are the tool of choice for scientists researching climate change. Like all models they suffer from errors, particularly systematic and location-specific representation errors. One way to reduce these errors is model output statistics (MOS) where the model output is fitted to observational data with machine learning. In this work, we assess the use of convolutional Deep Learning climate MOS approaches and present the ConvMOS architecture which is specifically designed based on the observation that there are systematic and location-specific errors in the precipitation estimates of climate models. We apply ConvMOS models to the simulated precipitation of the regional climate model REMO, showing that a combination of per-location model parameters for reducing location-specific errors and global model parameters for reducing systematic errors is indeed beneficial for MOS performance. We find that ConvMOS models can reduce errors considerably and perform significantly better than three commonly used MOS approaches and plain ResNet and U-Net models in most cases. Our results show that non-linear MOS models underestimate the number of extreme precipitation events, which we alleviate by training models specialized towards extreme precipitation events with the imbalanced regression method DenseLoss. While we consider climate MOS, we argue that aspects of ConvMOS may also be beneficial in other domains with geospatial data, such as air pollution modeling or weather forecasts.
In many real world settings, imbalanced data impedes model performance of learning algorithms, like neural networks, mostly for rare cases. This is especially problematic for tasks focusing on these rare occurrences. For example, when estimating precipitation, extreme rainfall events are scarce but important considering their potential consequences. While there are numerous well studied solutions for classification settings, most of them cannot be applied to regression easily. Of the few solutions for regression tasks, barely any have explored cost-sensitive learning which is known to have advantages compared to sampling-based methods in classification tasks. In this work, we propose a sample weighting approach for imbalanced regression datasets called DenseWeight and a cost-sensitive learning approach for neural network regression with imbalanced data called DenseLoss based on our weighting scheme. DenseWeight weights data points according to their target value rarities through kernel density estimation (KDE). DenseLoss adjusts each data point’s influence on the loss according to DenseWeight, giving rare data points more influence on model training compared to common data points. We show on multiple differently distributed datasets that DenseLoss significantly improves model performance for rare data points through its density-based weighting scheme. Additionally, we compare DenseLoss to the state-of-the-art method SMOGN, finding that our method mostly yields better performance. Our approach provides more control over model training as it enables us to actively decide on the trade-off between focusing on common or rare cases through a single hyperparameter, allowing the training of better models for rare data points.
Object six Degrees of Freedom (6DOF) pose estimation is a fundamental problem in many practical robotic applications, where the target or an obstacle with a simple or complex shape can move fast in cluttered environments. In this thesis, a 6DOF pose estimation algorithm is developed based on the fused data from a time-of-flight camera and a color camera. The algorithm is divided into two stages, an annealed particle filter based coarse pose estimation stage and a gradient decent based accurate pose optimization stage. In the first stage, each particle is evaluated with sparse representation. In this stage, the large inter-frame motion of the target can be well handled. In the second stage, the range data based conventional Iterative Closest Point is extended by incorporating the target appearance information and used for calculating the accurate pose by refining the coarse estimate from the first stage. For dealing with significant illumination variations during the tracking, spherical harmonic illumination modeling is investigated and integrated into both stages. The robustness and accuracy of the proposed algorithm are demonstrated through experiments on various objects in both indoor and outdoor environments. Moreover, real-time performance can be achieved with graphics processing unit acceleration.
The issue of sustainability is at the top of the political and societal agenda, being considered of extreme importance and urgency. Human individual action impacts the environment both locally (e.g., local air/water quality, noise disturbance) and globally (e.g., climate change, resource use). Urban environments represent a crucial example, with an increasing realization that the most effective way of producing a change is involving the citizens themselves in monitoring campaigns (a citizen science bottom-up approach). This is possible by developing novel technologies and IT infrastructures enabling large citizen participation. Here, in the wider framework of one of the first such projects, we show results from an international competition where citizens were involved in mobile air pollution monitoring using low cost sensing devices, combined with a web-based game to monitor perceived levels of pollution. Measures of shift in perceptions over the course of the campaign are provided, together with insights into participatory patterns emerging from this study. Interesting effects related to inertia and to direct involvement in measurement activities rather than indirect information exposure are also highlighted, indicating that direct involvement can enhance learning and environmental awareness. In the future, this could result in better adoption of policies towards decreasing pollution.
Parametric weighted finite automata (PWFA) are a multi-dimensional generalization of weighted finite automata. The expressiveness of PWFA contains the expressiveness of weighted finite automata as well as the expressiveness of affine iterated function system. The thesis discusses theory and applications of PWFA. The properties of PWFA definable sets are studied and it is shown that some fractal generator systems can be simulated using PWFA and that various real and complex functions can be represented by PWFA. Furthermore, the decoding of PWFA and the interpretation of PWFA definable sets is discussed.
Background
Information extraction techniques that get structured representations out of unstructured data make a large amount of clinically relevant information about patients accessible for semantic applications. These methods typically rely on standardized terminologies that guide this process. Many languages and clinical domains, however, lack appropriate resources and tools, as well as evaluations of their applications, especially if detailed conceptualizations of the domain are required. For instance, German transthoracic echocardiography reports have not been targeted sufficiently before, despite of their importance for clinical trials. This work therefore aimed at development and evaluation of an information extraction component with a fine-grained terminology that enables to recognize almost all relevant information stated in German transthoracic echocardiography reports at the University Hospital of Würzburg.
Methods
A domain expert validated and iteratively refined an automatically inferred base terminology. The terminology was used by an ontology-driven information extraction system that outputs attribute value pairs. The final component has been mapped to the central elements of a standardized terminology, and it has been evaluated according to documents with different layouts.
Results
The final system achieved state-of-the-art precision (micro average.996) and recall (micro average.961) on 100 test documents that represent more than 90 % of all reports. In particular, principal aspects as defined in a standardized external terminology were recognized with f 1=.989 (micro average) and f 1=.963 (macro average). As a result of keyword matching and restraint concept extraction, the system obtained high precision also on unstructured or exceptionally short documents, and documents with uncommon layout.
Conclusions
The developed terminology and the proposed information extraction system allow to extract fine-grained information from German semi-structured transthoracic echocardiography reports with very high precision and high recall on the majority of documents at the University Hospital of Würzburg. Extracted results populate a clinical data warehouse which supports clinical research.