TY - GEN A1 - Weber, Karsten T1 - Künstliche Intelligenz als vielschichtige Herausforderung Y1 - 2025 ER - TY - RPRT A1 - Bhardwaj, Prakhar A1 - Hauer, Franziska A1 - Mücke, Vanessa A1 - Dotter, Caroline A1 - Haug, Sonja A1 - Weikl, Simone A1 - Weber, Karsten T1 - Recording and understanding multi- and intermodal mobility - a review N2 - The transition to sustainable mobility requires detailed knowledge of actual mobility behavior. Recording mobility behavior through conventional, aggregated, or retrospective survey methods (e.g., counting stations, travel diaries) may yield incomplete insights. In addition, these methods are often expensive, prone to error, or not scalable. The aim of this literature review is to provide a systematic overview of the current state of the literature on recording and analysis of individual mobility data and, based on this, to identify possible research gaps and potential. The exploratory literature review includes social science surveys on multi- and intermodal mobility behavior, as well as an overview of possible ways to analyze travel data. The focus of the latter is on identifying relevant data sources and methodological approaches for segmentation, identification of transport mode, and trip purpose determination. Results from recent studies show that sensor-based data collection alone is not sufficient to fully and accurately map mobility behavior. One possible solution is a “best-of-both-worlds” approach that combines passive, automatic mobility data collection via smartphone sensors with targeted user corrections and additions (“human-in-the-loop”) within an app. The literature review shows that this approach yields the best data quality. Combining mobility surveys with passive collection of mobility data provides a comprehensive, accurate, and complete overview of mobility behavior. KW - multimodal mobility KW - mobility Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-87427 CY - Regensburg ER - TY - GEN A1 - Hauer, Franziska A1 - Haug, Sonja A1 - Scharf, Anna A1 - Schultz, Maximilian A1 - Weber, Karsten T1 - Quantitative Befragung zu Künstlicher Intelligenz für Nichtregierungsorganisationen -Fragebogen und Codebuch N2 - Künstliche Intelligenz (KI) stellt ein wichtiges Thema der Digitalisierung dar und hat das Potenzial viele Arbeitsfelder grundlegend zu verändern. Mit dem Einsatz von KI wird die Hoffnung verbunden durch Automatisierung Zeit und Geld zu sparen. Dies könnte besonders interessant für Organisationen sein, die wenig Ressourcen zur Verfügung haben, da sie nicht gewinnorientiert oder ehrenamtlich arbeiten. Dazu zählen Nichtregierungsorganisationen (NRO), die einen wichtigen Bestandteil der Zivilgesellschaft darstellen. Aufgrund der komplexer werdenden Arbeit von NROs scheint KI Möglichkeiten zur Bewältigung aktueller und zukünftiger Herausforderungen zu bieten. Über den Einsatz von KI in NROs ist derzeit jedoch wenig bekannt. Es wurde eine quantitative Querschnittstudie deutscher Nichtregierungsorganisationen (NRO) zum Thema KI durchgeführt. Der Web-Survey baut auf den Erkenntnissen des Scoping Reviews und der qualitativen Interviews auf und vertieft den Einblick in die Bereiche aktuelle Nutzung von KI, Wissenstand in den Organisationen sowie vorhandene und noch benötigte Ressourcen. Dafür wurden 343 NROs verschiedener Größen und Handlungsfelder befragt und die Ergebnisse quantitativ analysiert. KW - Künstliche Intelligenz KW - NGO Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-87066 N1 - Hauer, Franziska; Haug, Sonja; Scharf, Anna; Schultz, Maximilian; Weber, Karsten (2024): Künstliche Intelligenz für Nichtregierungsorganisationen - Quantitative Erforschung der Umsetzung in deutschen NROs. 3. Arbeitspapier. OTH Regensburg. Regensburg. DOI: 10.13140/RG.2.2.11094.66883 CY - Regensburg ER - TY - RPRT A1 - Bhardwaj, Prakhar A1 - Hauer, Franziska A1 - Mücke, Vanessa A1 - Dotter, Caroline A1 - Haug, Sonja A1 - Weikl, Simone A1 - Weber, Karsten T1 - Multi- und intermodale Mobilität erfassen und verstehen - eine Literaturübersicht N2 - Die Transformation zu nachhaltiger Mobilität erfordert eine detaillierte Erfassung des tatsächlichen Mobilitätsverhaltens, was durch herkömmliche, aggregierte oder retrospektive Erhebungsmethoden (z.B. Zählstellen, Wegetagebücher) nicht vollständig gelingt. Zudem sind diese Methoden oft teuer, fehleranfällig oder nicht nachhaltig skalierbar. Ziel dieser Literaturübersicht ist es, einen systematischen Überblick über den aktuellen Forschungsstand zur Erhebung und Analyse individueller Mobilitätsdaten zu geben und darauf aufbauend mögliche Forschungslücken sowie Potentiale zu identifizieren. Die explorative Literaturrecherche umfasst sozialwissenschaftliche Erhebungen zu multi- und intermodalem Mobilitätsverhalten, sowie einen Überblick über Analysemöglichkeiten von Wegedaten. Der Fokus liegt auf der Identifikation relevanter Datenquellen und methodischer Ansätze zur Segmentierung, Verkehrsmittelerkennung und Wegezweckbestimmung. Ergebnisse aus aktuellen Studien verdeutlichen, dass ausschließlich sensorgestützte Datenerhebungen bisher nicht ausreichen, um Mobilitätsverhalten vollumfänglich und präzise abzubilden. Eine mögliche Lösung ist ein "Best-of-both-worlds-Ansatz, der die passive, automatische Mobilitätsdatenerfassung via Smartphone-Sensorik mit gezielten Nutzerkorrekturen und Ergänzungen ("Human-in-the-Loop") innerhalb einer App kombiniert. Die Literaturauswertung zeigt, dass auf diesem Weg die beste Datenqualität gewonnen werden kann. Durch die Kombination von Mobilitätsbefragung mit passiver Erhebung von Mobilitätsdaten kann ein umfassender, möglichst vollständiger und korrekter Überblick über das Mobilitätsverhalten gewonnen werden. KW - KI KW - Mobilität Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-87415 CY - Regensburg ER - TY - CHAP A1 - Franz, Maja A1 - Winker, Tobias A1 - Groppe, Sven A1 - Mauerer, Wolfgang T1 - Hype or Heuristic? Quantum Reinforcement Learning for Join Order Optimisation T2 - 2024 IEEE International Conference on Quantum Computing and Engineering (QCE), 15-20 September 2024, Montreal, QC, Canada N2 - Identifying optimal join orders (JOs) stands out as a key challenge in database research and engineering. Owing to the large search space, established classical methods rely on approximations and heuristics. Recent efforts have successfully explored reinforcement learning (RL) for JO. Likewise, quantum versions of RL have received considerable scientific attention. Yet, it is an open question if they can achieve sustainable, overall practical advantages with improved quantum processors. In this paper, we present a novel approach that uses quantum reinforcement learning (QRL) for JO based on a hybrid variational quantum ansatz. It is able to handle general bushy join trees instead of resorting to simpler left-deep variants as compared to approaches based on quantum(-inspired) optimisation, yet requires multiple orders of magnitudes fewer qubits, which is a scarce resource even for post-NISQ systems. Despite moderate circuit depth, the ansatz exceeds current NISQ capabilities, which requires an evaluation by numerical simulations. While QRL may not significantly outperform classical approaches in solving the JO problem with respect to result quality (albeit we see parity), we find a drastic reduction in required trainable parameters. This benefits practically relevant aspects ranging from shorter training times compared to classical RL, less involved classical optimisation passes, or better use of available training data, and fits data-stream and low-latency processing scenarios. Our comprehensive evaluation and careful discussion delivers a balanced perspective on possible practical quantum advantage, provides insights for future systemic approaches, and allows for quantitatively assessing trade-offs of quantum approaches for one of the most crucial problems of database management systems. Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-76877 N1 - Corresponding author der OTH Regensburg: Maja Franz SP - 409 EP - 420 PB - IEEE ER - TY - JOUR A1 - Krüger, Tom A1 - Mauerer, Wolfgang T1 - Out of the Loop: Structural Approximation of Optimisation Landscapes and non-Iterative Quantum Optimisation JF - Quantum N2 - The Quantum Approximate Optimisation Algorithm (QAOA) is a widely studied quantum-classical iterative heuristic for combinatorial optimisation. While QAOA targets problems in complexity class NP, the classical optimisation procedure required in every iteration is itself known to be NP-hard. Still, advantage over classical approaches is suspected for certain scenarios, but nature and origin of its computational power are not yet satisfactorily understood. By introducing means of efficiently and accurately approximating the QAOA optimisation landscape from solution space structures, we derive a new algorithmic variant of unit-depth QAOA for two-level Hamiltonians (including all problems in NP): Instead of performing an iterative quantum-classical computation for each input instance, our non-iterative method is based on a quantum circuit that is instance-independent, but problem-specific. It matches or outperforms unit-depth QAOA for key combinatorial problems, despite reduced computational effort. Our approach is based on proving a long-standing conjecture regarding instance-independent structures in QAOA. By ensuring generality, we link existing empirical observations on QAOA parameter clustering to established approaches in theoretical computer science, and provide a sound foundation for understanding the link between structural properties of solution spaces and quantum optimisation. Y1 - 2025 U6 - https://doi.org/10.22331/q-2025-11-06-1903 N1 - Die Preprint-Version ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/start/2/rows/10/sortfield/score/sortorder/desc/searchtype/simple/query/Out+of+the+Loop/docId/7938 VL - 9 PB - Verein zur Förderung des Open Access Publizierens in den Quantenwissenschaften ER - TY - CHAP A1 - Schönberger, Manuel A1 - Trummer, Immanuel A1 - Mauerer, Wolfgang T1 - Large-Scale Multiple Query Optimisation with Incremental Quantum(-Inspired) Annealing T2 - Proceedings of the ACM on Management of Data N2 - Multiple-query optimization (MQO) seeks to reduce redundant work across query batches. While MQO offers opportunities for dramatic performance improvements, the problem is NP-hard, limiting the sizes of problems that can be solved on generic hardware. We propose to leverage specialized hardware solvers for optimization, such as Fujitsu's Digital Annealer (DA), to scale up MQO to problem sizes formerly out of reach. We present a novel incremental processing approach that combines classical computation with DA acceleration. By efficiently partitioning MQO problems into sets of partial problems, and by applying a dynamic search steering strategy that reapplies initially discarded information to incrementally process individual problems, our method overcomes capacity limitations, and scales to extremely large MQO instances (up to νm1000 queries). A thorough and comprehensive empirical evaluation finds our method substantially outperforms existing approaches. Our generalisable framework lays the ground for other database use-cases on quantum-inspired hardware, and bridges towards future quantum accelerators. Y1 - 2025 U6 - https://doi.org/10.1145/3749171 VL - 3 IS - 4 PB - ACM ER - TY - INPR A1 - Köster, Dominik A1 - Porkert, Florian A1 - Volbert, Klaus T1 - Multi-Day Scheduling for Electric Vehicle Routing: A Novel Model and Comparison Of Metaheuristics N2 - The increasing use of electric vehicles (EVs) requires efficient route planning solutions that take into account the limited range of EVs and the associated charging times, as well as the different types of charging stations. In this work, we model and solve an electric vehicle routing problem (EVRP) designed for a cross-platform navigation system for individual transport. The aim is to provide users with an efficient route for their daily appointments and to reduce possible inconveniences caused by charging their EV. Based on these assumptions, we propose a multi-day model in the form of a mixed integer programming (MIP) problem that takes into account the vehicle's battery capacity and the time windows of user's appointments. The model is solved using various established metaheuristics, including tabu search (TS), adaptive large neighborhood search (ALNS), and ant colony optimization (ACO). Furthermore, the performance of the individual approaches is analyzed using generated ensembles to estimate their behavior in reality and is compared with the exact results of the Google OR-Tools solver. 6 pages, 5 figures KW - Charging stations KW - Electric vehicles KW - Metaheuristics KW - Smart cities KW - Vehicle routing Y1 - 2025 U6 - https://doi.org/10.48550/arXiv.2512.21346 ER - TY - INPR A1 - Krüger, Tom A1 - Mauerer, Wolfgang T1 - Out of the Loop: Structural Approximation of Optimisation Landscapes and non-Iterative Quantum Optimisation N2 - The Quantum Approximate Optimisation Algorithm (qaoa) is a widely studied quantum-classical iterative heuristic for combinatorial optimisation. While qaoa targets problems in complexity class NP, the classical optimisation procedure required in every iteration is itself known to be NP-hard. Still, advantage over classical approaches is suspected for certain scenarios, but nature and origin of its computational power are not yet satisfactorily understood. By introducing means of efficiently and accurately approximating the qaoa optimisation landscape from solution space structures, we derive a new algorithmic variant: Instead of performing an iterative quantum-classical computation for each input instance, our non-iterative method is based on a quantum circuit that is instance-independent, but problem-specific. It matches or outperforms unit-depth qaoa for key combinatorial problems, despite reduced computational effort. Our approach is based on proving a long-standing conjecture regarding instance-independent structures in qaoa. By ensuring generality, we link existing empirical observations on qaoa parameter clustering to established approaches in theoretical computer science, and provide a sound foundation for understanding the link between structural properties of solution spaces and quantum optimisation. Y1 - 2024 N1 - er Aufsatz wurde peer-reviewed veröffentlicht und ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/start/3/rows/10/sortfield/score/sortorder/desc/searchtype/simple/query/out+of+the+loop/docId/8843 ER - TY - CHAP A1 - Mauerer, Wolfgang ED - Ali, Shaukat ED - Barzen, Johanna ED - Delgado, Andrea ED - Müller, Hausi A. ED - Murillo, Juan Manuel T1 - Quantum Software = Quantum minus Software? T2 - Dagstuhl Reports : Quantum Software Engineering : Report from Dagsthul Seminar 24512 N2 - The Dagstuhl Seminar 24512 on "Quantum Software Engineering" was held from December 15 to 20, 2024. It brought together 26 participants from industry and academia from 13 different countries, including senior and junior researchers as well as practitioners in the field of Quantum Software Engineering. The aim of the seminar was to advance software engineering methods and tools for the engineering of hybrid quantum systems by promoting personal interaction and open discussion among researchers who are already working in this emerging area of knowledge. The first day of the seminar was devoted to the topic "When software engineering meets quantum mechanics", while the second day focused on "Quantum software engineering and its challenges." During both days, 16 invited presentations were given. The rest of the seminar was organized into three working groups to address the topics "Quantum Software Design, Modelling and Architecturing", "Adaptive Hybrid Quantum Systems", and "Quantum Software Quality Assurance". The seminar was a very fruitful experience for all participants both in terms of scientific outcomes and in terms of the personal relationships that were generated to jointly address future experiences. Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:0030-drops-230469 VL - 14 IS - 12 SP - 73 EP - 74 PB - Dagstuhl Publishing ER - TY - INPR A1 - Cenko, Egidia A1 - Weimann, Thure Georg A1 - Raptis, Georgios T1 - Navigating the DiGA Jungle: A Taxonomy and Archetypal Framework of the German Digital Therapeutics Landscape N2 - Digital therapeutics (DTx) are patient-facing apps designed to support individuals in their daily lives. Therefore, they have thepotential to revolutionize healthcare by empowering and engaging patients to become active players in their own care. Despitethe increasing adoption of DTx in national healthcare systems, research on their design remains limited. The present studyintroduces “DiGATax”, a taxonomy designed to categorize and analyze DTx, including perspectives on content, interventiondelivery logic and technology, as well as the patient’s interface, consolidating and expanding upon prior taxonomic work.Based on n = 44 applications retrieved from the German DiGA directory that demonstrated positive health outcomes, thetaxonomy is supported by empirical evidence. Additionally, the study contributes by presenting an archetype frameworkof DTx derived from a taxonomy-based cluster analysis. Further analyses offer insights into specific combinations of DTxcharacteristics across archetypes, the user interface as a key factor in their acceptance, and potential links between DTxdesign and health-related and user engagement outcomes. By offering new insights into DTx design, this study contributestowards more organized research and reporting, ultimately paving the way for the development of effective solutions. It alsomarks a further step towards Meta-DTx, which aim to align patient care for multimorbid patients under one umbrella. KW - Archetypes KW - Digital Behavior Change Interventions KW - Digital Therapeutics KW - mHealth KW - Taxonomy Y1 - 2025 U6 - https://doi.org/10.64898/2025.12.30.25343225 ER - TY - CHAP A1 - Stadler, Sebastian A1 - Seitz, Marie Sophie T1 - When algorithms design: a comparative study pits an AI-generated application against a manual application in a competition of user experience and user-friendliness T2 - Proceedings of 2025 11th international HCI and UX conference in Indonesia, 5 + 6. December 2025, Bali (CHIuXiD) N2 - The rapid growth of artificial intelligence has enabled automated website creation, including design, content, and SEO. This study compares an AI-generated application with a manually designed one in terms of usability and user experience. Forty-two participants evaluated efficiency, effectiveness, satisfaction, emotional responses, and behavioural patterns. The manual application outperformed the AI-generated one in most datasets, particularly in clarity, supportiveness, and comprehensibility, while speed, learnability, activation, creativity, and security showed no significant differences. Users reported more positive emotions and confident interactions with the manual application, whereas the AI application elicited uncertainty and frustration. These behavioural differences were also reflected in processing times and task completion rates, suggesting that AI-generated solutions currently struggle to meet established UX standards. Nevertheless, the AI builder demonstrated potential for rapid prototyping and low-effort website creation, offering advantages for users with limited technical skills or resources. The findings highlight both the opportunities and limitations of current AI-based web design tools and underline the importance of human-centered design expertise in achieving high-quality user experiences. Despite existing shortcomings, AI-supported tools may become increasingly viable as technology advances. KW - user experience KW - usability KW - artificial intelligence KW - user experience questionnaire Y1 - 2025 U6 - https://doi.org/10.1109/CHIuXiD68326.2025.11323819 SP - 55 EP - 60 PB - IEEE ER - TY - CHAP A1 - Dünnweber, Jan A1 - Lorenz, Marcel A1 - Bräu, Fabian A1 - Saito, Hiroshi A1 - Melzer, Matthias T1 - Driving out the Bears from the Suburbs by Means of Location Theory and Intelligent Agents T2 - Proceedings of the ACM 12th International Conference on Animal-Computer Interaction, ACI 2025: The International Conference on Animal-Computer Interaction ; Indiana USA ; 01-04.12.2025 N2 - Due to the recent surge in bear attacks, affected municipalities have taken various measures to drive out the intruders. Political considerations include, in many places, the expansion of shooting permits. More and more regional governmental authorities invest in technologies for efficient bear hunting, such as IP cameras and computer vision. From an ecological perspective, such procedures, which are all solely aimed at decreasing the bear population, are questionable, since ecosystems can benefit from a peaceful coexistence of bears with humans: Bears remove animal carcasses and thus return vital nutrients to the soil and prevent the spread of diseases. Moreover, the bears keep the deer population in check, which is advantageous for the forest vegetation. Conflicts start, once the bears intrude populated areas, which are often rural villages and suburbs with a low population density. Naturally, bears do not see humans as prey and respect their size. When the animal attacks, this happens rather for territorial or protective reasons, e. g., after an unexpected encounter with humans unsettled it. In this work, we analyze means to eradicate the danger: We dynamically apply Weiszfeld’s algorithm for finding the shifting geometric median between time-windowed black bear sightings and use the results for implementing an early warning system. In addition, we weigh the spotted locations and predict the expected migration behavior between bear habitats using a simulation with intelligent agents. As a case study, we apply our software to data from Romania and Fukushima and identify the positions where bear repelling facilities should be placed most effectively. Y1 - 2025 SN - 9798400721236 U6 - https://doi.org/10.1145/3768539.3768546 PB - ACM CY - New York, NY, USA ER - TY - CHAP A1 - Gutbrod, Max A1 - Rauber, David A1 - Weber Nunes, Danilo A1 - Palm, Christoph T1 - OpenMIBOOD: Open Medical Imaging Benchmarks for Out-Of-Distribution Detection T2 - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 10.-17. June 2025, Nashville N2 - The growing reliance on Artificial Intelligence (AI) in critical domains such as healthcare demands robust mechanisms to ensure the trustworthiness of these systems, especially when faced with unexpected or anomalous inputs. This paper introduces the Open Medical Imaging Benchmarks for Out-Of-Distribution Detection (OpenMIBOOD), a comprehensive framework for evaluating out-of-distribution (OOD) detection methods specifically in medical imaging contexts. OpenMIBOOD includes three benchmarks from diverse medical domains, encompassing 14 datasets divided into covariate-shifted in-distribution, nearOOD, and far-OOD categories. We evaluate 24 post-hoc methods across these benchmarks, providing a standardized reference to advance the development and fair comparison of OODdetection methods. Results reveal that findings from broad-scale OOD benchmarks in natural image domains do not translate to medical applications, underscoring the critical need for such benchmarks in the medical field. By mitigating the risk of exposing AI models to inputs outside their training distribution, OpenMIBOOD aims to support the advancement of reliable and trustworthy AI systems in healthcare. The repository is available at https://github.com/remic-othr/OpenMIBOOD. KW - Benchmark testing KW - Reliability KW - Trustworthiness KW - out-of-distribution Y1 - 2025 UR - https://openaccess.thecvf.com/content/CVPR2025/html/Gutbrod_OpenMIBOOD_Open_Medical_Imaging_Benchmarks_for_Out-Of-Distribution_Detection_CVPR_2025_paper.html SN - 979-8-3315-4364-8 U6 - https://doi.org/10.1109/CVPR52734.2025.02410 N1 - Die Preprint-Version ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/8059 SP - 25874 EP - 25886 PB - IEEE ER - TY - JOUR A1 - Heinz, Florian A1 - Schildgen, Johannes T1 - Experience Report: Hey LLM, Generate SQL! JF - Datenbank-Spektrum N2 - Large Language Models have advanced to a resourceful tool with many applications. One particularly interesting use case is the LLM-aided generation of ad-hoc database queries and the possibility of subsequent processing of the results in a way suiting the users intents. In this article, practical ways and experiences are described on how to effectively use LLMs to map a natural-language user query to an SQL query conforming to a specific database schema and post-processing the results of this query in order to, for example, create an appealing visualization. Best results are achieved under favorable circumstances, as, for example, a clean and meaningful named database schema. Y1 - 2025 U6 - https://doi.org/10.1007/s13222-025-00512-w N1 - Corresponding author der OTH Regensburg: Florian Heinz VL - 25 IS - 2 SP - 95 EP - 101 PB - Springer CY - Wiesbaden ER - TY - CHAP A1 - Preiß, Niklas A1 - Westner, Markus T1 - From Agents to Copilots: a Systematic Review of Digital Assistant Technology Adoption in Proprietary Productivity Software T2 - Annals of Computer Science and Information Systems N2 - This study provides a systematic review of how the impact and adaptation of digital assistant technologies (DATs) are defined, operationalized, and studied, synthesizing key domains where DATs generate or are expected to generate value. Based on an analysis of 61 articles published since 2013, it identifies five main areas of impact: productivity and efficiency, business development, resource optimization, quality enhancement, and the promotion of learning and creativity. The review highlights DAT adoption across various disciplines and industries, while revealing limited longitudinal research on benefits and adaptation. Key gaps remain in understanding strategic use and sustained impact. Future research should explore longitudinal comparisons of recently introduced generative AI-driven DATs and their organizational implications. This review contributes to information systems research by structuring current knowledge on DAT adoption and outcomes, and by proposing a research agenda to support deeper exploration of their value and long-term integration. KW - Digital Assistant Technology KW - Copilot KW - Generative AI KW - Technology Adoption KW - Productivity Y1 - 2025 U6 - https://doi.org/10.15439/2025F3271 SN - 2300-5963 VL - 43 SP - 565 EP - 576 PB - Polish Information Processing Society ER - TY - CHAP A1 - Köhler, Jessica A1 - Harl, Maximilian Victor A1 - Westner, Markus A1 - Strahringer, Susanne T1 - Can AI be a Scholar? A Systematic Review on the Role of Generative AI in Systematic Literature Reviews T2 - 2025 27th International Conference on Business Informatics (CBI), 09-12. September 2025, Lisbon, Portugal N2 - Systematic literature reviews (SLRs) are foundational for research but resource-intensive to conduct. With the rise of large language models (LLMs) such as ChatGPT, generative AI (GenAI) tools are being increasingly explored for their potential to support and transform the SLR process. This study presents a systematic review of peerreviewed articles that examine how LLM-based GenAI tools are used in different SLR phases. Following the PRISMA 2020 guidelines, we screened 1,846 publications published since January 2021 until April 2025 and selected 54 for in-depth analysis. Each study was coded by review phase, prompting approach, automation level, validation type and challenges. Our findings show that GenAI is most often used to support in the screening, search, and writing phases, typically through Basic Prompting and under human oversight. While many studies report efficiency gains, concerns remain regarding validity, transparency, and methodological rigor. Moreover, GenAI is frequently applied to isolated tasks but is rarely embedded in a structured, methodologically guided review processhighlighting the need for clearer phase-specific guidance and standards. We offer a structured, phase-specific synthesis that highlights both the promise and the current limitations of GenAI in literature reviews and thereby offer practical recommendations for the responsible use of GenAI in literature reviews. KW - Generative AI KW - Large Language Models KW - ChatGPT KW - Systematic Literature Review KW - Evidence Synthesis KW - AI-assisted Literature Review Y1 - 2025 U6 - https://doi.org/10.1109/CBI68102.2025.00012 PB - IEEE ER - TY - CHAP A1 - Westner, Markus ED - Hofbauer, Günter T1 - Generative AI in Business BT - Opportunities, Challenges, and Strategic Implications Authors Markus Westner T2 - Artificial Intelligence in Business and Engineering N2 - Generative Artificial Intelligence (GenAI) is transforming industries at an unprecedented rate, offering novel opportunities for productivity and innovation. This article explores the adoption of GenAI, highlighting its accelerated uptake compared to previous technologies. Key topics include productivity gains through GenAI tools (e.g., ChatGPT, Klarna’s AI assistant), challenges such as data quality and organizational readiness, and the implications for business strategy. Practical recommendations for managing generative AI adoption and maximizing its impact on both individual and organizational levels are also provided. The analysis underscores the necessity of aligning AI capabilities with customer needs and creating data-driven, adaptable business models. KW - Generative AI KW - Productivity KW - Technology Adoption Y1 - 2025 SN - 978-3-17-046742-2 SP - 38 EP - 50 PB - Kohlhammer CY - Stuttgart ER - TY - INPR A1 - Rückert, Tobias A1 - Rauber, David A1 - Maerkl, Raphaela A1 - Klausmann, Leonard A1 - Yildiran, Suemeyye R. A1 - Gutbrod, Max A1 - Nunes, Danilo Weber A1 - Moreno, Alvaro Fernandez A1 - Luengo, Imanol A1 - Stoyanov, Danail A1 - Toussaint, Nicolas A1 - Cho, Enki A1 - Kim, Hyeon Bae A1 - Choo, Oh Sung A1 - Kim, Ka Young A1 - Kim, Seong Tae A1 - Arantes, Gonçalo A1 - Song, Kehan A1 - Zhu, Jianjun A1 - Xiong, Junchen A1 - Lin, Tingyi A1 - Kikuchi, Shunsuke A1 - Matsuzaki, Hiroki A1 - Kouno, Atsushi A1 - Manesco, João Renato Ribeiro A1 - Papa, João Paulo A1 - Choi, Tae-Min A1 - Jeong, Tae Kyeong A1 - Park, Juyoun A1 - Alabi, Oluwatosin A1 - Wei, Meng A1 - Vercauteren, Tom A1 - Wu, Runzhi A1 - Xu, Mengya A1 - an Wang, A1 - Bai, Long A1 - Ren, Hongliang A1 - Yamlahi, Amine A1 - Hennighausen, Jakob A1 - Maier-Hein, Lena A1 - Kondo, Satoshi A1 - Kasai, Satoshi A1 - Hirasawa, Kousuke A1 - Yang, Shu A1 - Wang, Yihui A1 - Chen, Hao A1 - Rodríguez, Santiago A1 - Aparicio, Nicolás A1 - Manrique, Leonardo A1 - Lyons, Juan Camilo A1 - Hosie, Olivia A1 - Ayobi, Nicolás A1 - Arbeláez, Pablo A1 - Li, Yiping A1 - Khalil, Yasmina Al A1 - Nasirihaghighi, Sahar A1 - Speidel, Stefanie A1 - Rückert, Daniel A1 - Feussner, Hubertus A1 - Wilhelm, Dirk A1 - Palm, Christoph T1 - Comparative validation of surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation in endoscopy: Results of the PhaKIR 2024 challenge N2 - Reliable recognition and localization of surgical instruments in endoscopic video recordings are foundational for a wide range of applications in computer- and robot-assisted minimally invasive surgery (RAMIS), including surgical training, skill assessment, and autonomous assistance. However, robust performance under real-world conditions remains a significant challenge. Incorporating surgical context - such as the current procedural phase - has emerged as a promising strategy to improve robustness and interpretability. To address these challenges, we organized the Surgical Procedure Phase, Keypoint, and Instrument Recognition (PhaKIR) sub-challenge as part of the Endoscopic Vision (EndoVis) challenge at MICCAI 2024. We introduced a novel, multi-center dataset comprising thirteen full-length laparoscopic cholecystectomy videos collected from three distinct medical institutions, with unified annotations for three interrelated tasks: surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation. Unlike existing datasets, ours enables joint investigation of instrument localization and procedural context within the same data while supporting the integration of temporal information across entire procedures. We report results and findings in accordance with the BIAS guidelines for biomedical image analysis challenges. The PhaKIR sub-challenge advances the field by providing a unique benchmark for developing temporally aware, context-driven methods in RAMIS and offers a high-quality resource to support future research in surgical scene understanding. Y1 - 2025 N1 - Der Aufsatz wurde peer-reviewed veröffentlicht und ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/start/0/rows/10/sortfield/score/sortorder/desc/searchtype/simple/query/10.1016%2Fj.media.2026.103945/docId/8846 ER - TY - BOOK A1 - Westner, Markus A1 - Strasser, Artur T1 - Objectives and Key Results verstehen und anwenden N2 - Dieses essential gibt eine strukturierte und kompakte Einführung in die Objectives and Key Results-Managementmethode. Es erläutert Herkunft, Definition und Einsatzzweck von Objectives and Key Results (OKR) und beschreibt deren zentrale Bestandteile. Leser:innen erfahren, wie gute Objectives und Key Results formuliert werden und welche Strategien sich zur erfolgreichen Implementierung eignen. Zudem werden praxisrelevante Erfolgs- und Misserfolgsfaktoren analysiert, die den Einsatz von OKR maßgeblich beeinflussen. Ein fundierter Leitfaden für alle, die OKR als Steuerungsinstrument in Organisationen verstehen und wirksam einsetzen möchten. KW - Objectives and Key Results (OKR) KW - Zielmanagement KW - Performance Management KW - Strategieumsetzung KW - Kennzahlen / KPI Y1 - 2025 SN - 978-3-658-50381-9 U6 - https://doi.org/10.1007/978-3-658-50382-6 PB - Springer Gabler CY - Wiesbaden ER - TY - CHAP A1 - Sachenbacher, Martin A1 - Leucker, Martin T1 - Towards Monitoring and Diagnosis of Quantum Digital Twins T2 - Lecture Notes in Informatics (LNI) N2 - As access to larger-scale quantum computers is still limited, it is useful to simulate quantum algorithms on specialized programmable logic hardware such as FPGAs to exploit analogies between quantum and digital circuits, and allow for analysis and efficient debugging of quantum programs. Such simulations can be viewed as digital representations of actual physical systems, commonly referred to as Digital Twins (DTs). In previous work, we applied formal methods from runtime verification to synthesize monitors capable of real-time analysis of conventional DTs. Specifically, these monitors supervise alignment with their physical counterparts and support model-based diagnosis to identify potential root causes of observed deviations. In this short paper, we present ongoing work aimed at adapting and extending our approach to hardware-based simulation and analysis of quantum software on FPGAs, with the objective of improving both accuracy and resilience. Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-87546 VL - P-366 SP - 1767 EP - 1772 PB - Gesellschaft für Informatik e.V. CY - Bonn ER - TY - JOUR A1 - Höß, Nicole A1 - Mauerer, Wofgang T1 - Generative KI: Vom technologischen Paradigmenwechsel zur Vision einer neuen Ära im Straßenverkehr JF - Straßenverkehrstechnik N2 - Generative Künstliche Intelligenz (KI) geht bereits mit enormen Effizienzsteigerungen für Domänen einher, die von Kreativität profitieren. Die Fähigkeit, beliebig viele realitätsnahe Daten zu synthetisieren, bietet jedoch auch neue Potenziale für Anwendungen, die bisher aufgrund fehlender Datenmengen noch nicht den Sprung in die Praxis überwunden haben. Im Straßenverkehr betrifft dies z. B. das autonome Fahren in herausfordernden Situationen. Um neue Anwendungspotenziale zu erschließen, ist es allerdings erforderlich, die grundlegende Funktionsweise der Technologie gepaart mit ihren Chancen und Herausforderungen zu verstehen. Dieser Fachbeitrag dient deshalb als erste Einführung in die generative KI, indem er einen Überblick über zentrale Modelle, ihre grundlegenden Ideen und Funktionsweisen gibt. Darauf basierend werden typische Problemarten aufgezeigt, die mit generativer KI neu betrachtet und besser gelöst werden können – allgemein und spezifisch für den Straßenverkehr. Unter gleichzeitiger Betrachtung derzeitiger Limitierungen dient der Beitrag als Entscheidungshilfe für die Selektion passender generativer oder klassischer KI-Verfahren. N2 - Generative artificial intelligence (AI) is associated with enormous increases in efficiency for domains that benefit from creativity. However, the ability to synthesize any amount of realistic data also offers new potential for applications that have not yet made the leap into practice due to a lack of data. In road traffic, for example, this applies to autonomous driving in challenging situations. However, in order to unlock new application potential, it is necessary to understand the basic functioning of the technology coupled with its opportunities and challenges. This article therefore serves as a gentle introduction to generative AI by providing an overview of central models, their fundamental ideas and function. Based on this, we present typical problem types that can be revisited and profit from generative AI – in general and specifically for road traffic. While at the same time considering current limitations, the article serves as a decision support for the selection of suitable generative or classic AI methods. T2 - Generative AI: From a technological paradigm shift to a vision of a new era in road transport Y1 - 2025 U6 - https://doi.org/10.53184/SVT4-2025-2 VL - 6 IS - 4 SP - 247 EP - 252 PB - Kirschbaum Verlag GmbH ER - TY - JOUR A1 - Schildgen, Johannes T1 - Individuelle Data-Warehousing-Projekte anstelle einer Abschlussklausur JF - Datenbank-Spektrum N2 - Dieser Artikel berichtet über die Erfahrungen aus zwei Semestern, in denen Studierende der Data-Warehousing-Vorlesung individuelle Einzelprojekte bearbeitet haben. Das Projekt besteht aus fünf Challenges: die Vorstellung der Projektidee, ein Peer-Review anderer Ideen, Datenmodellierung, ETL und analytische Queries, sowie die Erstellung eines Power BI-Dashboards. KW - Data Analytics KW - Data Engineering KW - Data Mining KW - Experiential Education KW - Methodology of Data Collection and Processing KW - Project Management Y1 - 2025 U6 - https://doi.org/10.1007/s13222-025-00502-y SN - 1618-2162 PB - Springer ER - TY - JOUR A1 - Schmiedt, Anja A1 - Weiß, Christian T1 - The pair correlation function of multi-dimensional low-discrepancy sequences with small stochastic error terms JF - Journal of Number Theory N2 - In any dimension d ≥ 2, there is no known example of a low-discrepancy sequence which possesses Poisssonian pair correlations. This is in some sense rather surprising, because low-discrepancy sequences always have β-Poissonian pair correlations for all 0 < β < 1 d and are therefore arbitrarily close to having Poissonian pair correlations (which corresponds to the case β = 1 d ). In this paper, we further elaborate on the closeness of the two notions. We show that d-dimensional Kronecker sequences for badly approximable vectors � α with an arbitrary small uniformly distributed stochastic error term generically have β = 1 d -Poissonian pair correlations. Y1 - 2024 U6 - https://doi.org/10.1016/j.jnt.2023.12.011 SN - 0022-314X VL - 259 SP - 422 EP - 437 PB - Elsevier ER - TY - GEN A1 - Rückert, Tobias A1 - Rückert, Daniel A1 - Palm, Christoph T1 - Corrigendum to “Methods and datasets for segmentation of minimally invasive surgical instruments in endoscopic images and videos: A review of the state of the art” [Comput. Biol. Med. 169 (2024) 107929] T2 - Computers in Biology and Medicine N2 - The authors regret that the SAR-RARP50 dataset is missing from the description of publicly available datasets presented in Chapter 4. Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-70337 N1 - Aufsatz unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/6983 PB - Elsevier ER - TY - JOUR A1 - Hammer, Simone A1 - Nunes, Danilo Weber A1 - Hammer, Michael A1 - Zeman, Florian A1 - Akers, Michael A1 - Götz, Andrea A1 - Balla, Annika A1 - Doppler, Michael Christian A1 - Fellner, Claudia A1 - Da Platz Batista Silva, Natascha A1 - Thurn, Sylvia A1 - Verloh, Niklas A1 - Stroszczynski, Christian A1 - Wohlgemuth, Walter Alexander A1 - Palm, Christoph A1 - Uller, Wibke T1 - Deep learning-based differentiation of peripheral high-flow and low-flow vascular malformations in T2-weighted short tau inversion recovery MRI JF - Clinical hemorheology and microcirculation N2 - BACKGROUND Differentiation of high-flow from low-flow vascular malformations (VMs) is crucial for therapeutic management of this orphan disease. OBJECTIVE A convolutional neural network (CNN) was evaluated for differentiation of peripheral vascular malformations (VMs) on T2-weighted short tau inversion recovery (STIR) MRI. METHODS 527 MRIs (386 low-flow and 141 high-flow VMs) were randomly divided into training, validation and test set for this single-center study. 1) Results of the CNN's diagnostic performance were compared with that of two expert and four junior radiologists. 2) The influence of CNN's prediction on the radiologists' performance and diagnostic certainty was evaluated. 3) Junior radiologists' performance after self-training was compared with that of the CNN. RESULTS Compared with the expert radiologists the CNN achieved similar accuracy (92% vs. 97%, p = 0.11), sensitivity (80% vs. 93%, p = 0.16) and specificity (97% vs. 100%, p = 0.50). In comparison to the junior radiologists, the CNN had a higher specificity and accuracy (97% vs. 80%, p <  0.001; 92% vs. 77%, p <  0.001). CNN assistance had no significant influence on their diagnostic performance and certainty. After self-training, the junior radiologists' specificity and accuracy improved and were comparable to that of the CNN. CONCLUSIONS Diagnostic performance of the CNN for differentiating high-flow from low-flow VM was comparable to that of expert radiologists. CNN did not significantly improve the simulated daily practice of junior radiologists, self-training was more effective. KW - magnetic resonance imaging KW - deep learning KW - Vascular malformation Y1 - 2024 U6 - https://doi.org/10.3233/CH-232071 SP - 1 EP - 15 PB - IOP Press ET - Pre-press ER - TY - CHAP A1 - Schönhärl, Stefan A1 - Fuxen, Philipp A1 - Graf, Julian A1 - Schmidt, Jonas A1 - Hackenberg, Rudolf A1 - Mottok, Jürgen T1 - An Automotive Penetration Testing Framework for IT-Security Education T2 - Cloud Computing 2022: The Thirteenth International Conference on Cloud Computing, GRIDs, and Virtualization, Special Track FAST-CSP, Barcelona, Spain, 24.-28.04.2022 N2 - Automotive Original Equipment Manufacturer (OEM) and suppliers started shifting their focus towards the security of their connected electronic programmable products recently since cars used to be mainly mechanical products. However, this has changed due to the rising digitalization of vehicles. Security and functional safety have grown together and need to be addressed as a single issue, referred to as automotive security, in the following article. One way to accomplish security is automotive security education. The scientific contribution of this paper is to establish an Automotive Penetration Testing Education Platform (APTEP). It consists of three layers representing different attack points of a vehicle. The layers are the outer, inner, and core layers. Each of those contains multiple interfaces, such as Wireless Local Area Network (WLAN) or electric vehicle charging interfaces in the outer layer, message bus systems in the inner layer, and debug or diagnostic interfaces in the core layer. One implementation of APTEP is in a hardware case and as a virtual platform, referred to as the Automotive Network Security Case (ANSKo). The hardware case contains emulated control units and different communication protocols. The virtual platform uses Docker containers to provide a similar experience over the internet. Both offer two kinds of challenges. The first introduces users to a specific interface, while the second combines multiple interfaces, to a complex and realistic challenge. This concept is based on modern didactic theory, such as constructivism and problem-based learning. Computer Science students from the Ostbayerische Technische Hochschule (OTH)Regensburg experienced the challenges as part of a special topic course and provided positive feedback. KW - IT-Security KW - Education KW - Automotive KW - Penetration testing KW - Education framework Y1 - 2022 UR - https://www.thinkmind.org/index.php?view=article&articleid=cloud_computing_2022_1_10_28001 SN - 978-1-61208-948-5 SP - 1 EP - 6 PB - IARIA CY - [Wilmington, DE, USA] ER - TY - GEN A1 - Meinikheim, Michael A1 - Mendel, Robert A1 - Probst, Andreas A1 - Scheppach, Markus W. A1 - Nagl, Sandra A1 - Schnoy, Elisabeth A1 - Römmele, Christoph A1 - Prinz, Friederike A1 - Schlottmann, Jakob A1 - Messmann, Helmut A1 - Palm, Christoph A1 - Ebigbo, Alanna T1 - Einfluss von Künstlicher Intelligenz auf die Performance von niedergelassenen Gastroenterolog:innen bei der Beurteilung von Barrett-Ösophagus T2 - Zeitschrift für Gastroenterologie N2 - Einleitung  Die Differenzierung zwischen nicht dysplastischem Barrett-Ösophagus (NDBE) und mit Barrett-Ösophagus assoziierten Neoplasien (BERN) während der endoskopischen Inspektion erfordert viel Expertise. Die frühe Diagnosestellung ist wichtig für die weitere Prognose des Barrett-Karzinoms. In Deutschland werden Patient:innen mit einem Barrett-Ösophagus (BE) in der Regel im niedergelassenen Sektor überwacht. Ziele  Ziel ist es, den Einfluss von einem auf Künstlicher Intelligenz (KI) basierenden klinischen Entscheidungsunterstützungssystems (CDSS) auf die Performance von niedergelassenen Gastroenterolog:innen (NG) bei der Evaluation von Barrett-Ösophagus (BE) zu untersuchen. Methodik  Es erfolgte die prospektive Sammlung von 96 unveränderten hochauflösenden Videos mit Fällen von Patient:innen mit histologisch bestätigtem NDBE und BERN. Alle eingeschlossenen Fälle enthielten mindestens zwei der folgenden Darstellungsmethoden: HD-Weißlichtendoskopie, Narrow Band Imaging oder Texture and Color Enhancement Imaging. Sechs NG von sechs unterschiedlichen Praxen wurden als Proband:innen eingeschlossen. Es erfolgte eine permutierte Block-Randomisierung der Videofälle in entweder Gruppe A oder Gruppe B. Gruppe A implizierte eine Evaluation des Falls durch Proband:innen zunächst ohne KI und anschließend mit KI als CDSS. In Gruppe B erfolgte die Evaluation in umgekehrter Reihenfolge. Anschließend erfolgte eine zufällige Wiedergabe der so entstandenen Subgruppen im Rahmen des Tests. Ergebnis  In diesem Test konnte ein von uns entwickeltes KI-System (Barrett-Ampel) eine Sensitivität von 92,2%, eine Spezifität von 68,9% und eine Accuracy von 81,3% erreichen. Mit der Hilfe von KI verbesserte sich die Sensitivität der NG von 64,1% auf 71,2% (p<0,001) und die Accuracy von 66,3% auf 70,8% (p=0,006) signifikant. Eine signifikante Verbesserung dieser Parameter zeigte sich ebenfalls, wenn die Proband:innen die Fälle zunächst ohne KI evaluierten (Gruppe A). Wurde der Fall jedoch als Erstes mit der Hilfe von KI evaluiert (Gruppe B), blieb die Performance nahezu konstant. Schlussfolgerung  Es konnte ein performantes KI-System zur Evaluation von BE entwickelt werden. NG verbessern sich bei der Evaluation von BE durch den Einsatz von KI. KW - Barrett-Ösophagus KW - Künstliche Intelligenz Y1 - 2023 UR - https://www.thieme-connect.de/products/ejournals/abstract/10.1055/s-0043-1771711 U6 - https://doi.org/10.1055/s-0043-1771711 VL - 61 IS - 8 PB - Thieme CY - Stuttgart ER - TY - GEN A1 - Scheppach, Markus W. A1 - Mendel, Robert A1 - Probst, Andreas A1 - Nagl, Sandra A1 - Meinikheim, Michael A1 - Yip, Hon Chi A1 - Lau, Louis Ho Shing A1 - Chiu, Philip Wai Yan A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - Effekt eines Künstliche Intelligenz (KI) – Algorithmus auf die Gefäßdetektion bei third space Endoskopien T2 - Zeitschrift für Gastroenterologie N2 - Einleitung  Third space Endoskopieprozeduren wie die endoskopische Submukosadissektion (ESD) und die perorale endoskopische Myotomie (POEM) sind technisch anspruchsvoll und gehen mit untersucherabhängigen Komplikationen wie Blutungen und Perforationen einher. Grund hierfür ist die unabsichtliche Durchschneidung von submukosalen Blutgefäßen ohne präemptive Koagulation. Ziele Die Forschungsfrage, ob ein KI-Algorithmus die intraprozedurale Gefäßerkennung bei ESD und POEM unterstützen und damit Komplikationen wie Blutungen verhindern könnte, erscheint in Anbetracht des erfolgreichen Einsatzes von KI bei der Erkennung von Kolonpolypen interessant. Methoden  Auf 5470 Einzelbildern von 59 third space Endoscopievideos wurden submukosale Blutgefäße annotiert. Zusammen mit weiteren 179.681 nicht-annotierten Bildern wurde ein DeepLabv3+neuronales Netzwerk mit dem ECMT-Verfahren für semi-supervised learning trainiert, um Blutgefäße in Echtzeit erkennen zu können. Für die Evaluation wurde ein Videotest mit 101 Videoclips aus 15 vom Trainingsdatensatz separaten Prozeduren mit 200 vordefinierten Gefäßen erstellt. Die Gefäßdetektionsrate, -zeit und -dauer, definiert als der Prozentsatz an Einzelbildern eines Videos bezogen auf den Goldstandard, auf denen ein definiertes Gefäß erkannt wurde, wurden erhoben. Acht erfahrene Endoskopiker wurden mithilfe dieses Videotests im Hinblick auf Gefäßdetektion getestet, wobei eine Hälfte der Videos nativ, die andere Hälfte nach Markierung durch den KI-Algorithmus angesehen wurde. Ergebnisse  Der mittlere Dice Score des Algorithmus für Blutgefäße war 68%. Die mittlere Gefäßdetektionsrate im Videotest lag bei 94% (96% für ESD; 74% für POEM). Die mediane Gefäßdetektionszeit des Algorithmus lag bei 0,32 Sekunden (0,3 Sekunden für ESD; 0,62 Sekunden für POEM). Die mittlere Gefäßdetektionsdauer lag bei 59,1% (60,6% für ESD; 44,8% für POEM) des Goldstandards. Alle Endoskopiker hatten mit KI-Unterstützung eine höhere Gefäßdetektionsrate als ohne KI. Die mittlere Gefäßdetektionsrate ohne KI lag bei 56,4%, mit KI bei 71,2% (p<0.001). Schlussfolgerung  KI-Unterstützung war mit einer statistisch signifikant höheren Gefäßdetektionsrate vergesellschaftet. Die mediane Gefäßdetektionszeit von deutlich unter einer Sekunde sowie eine Gefäßdetektionsdauer von größer 50% des Goldstandards wurden für den klinischen Einsatz als ausreichend erachtet. In prospektiven Anwendungsstudien sollte der KI-Algorithmus auf klinische Relevanz getestet werden. KW - Künstliche Intelligenz Y1 - 2023 U6 - https://doi.org/10.1055/s-0043-1771980 VL - 61 IS - 08 PB - Thieme CY - Stuttgart ER - TY - GEN A1 - Scheppach, Markus W. A1 - Mendel, Robert A1 - Muzalyova, Anna A1 - Rauber, David A1 - Probst, Andreas A1 - Nagl, Sandra A1 - Römmele, Christoph A1 - Yip, Hon Chi A1 - Lau, Louis Ho Shing A1 - Gölder, Stefan Karl A1 - Schmidt, Arthur A1 - Kouladouros, Konstantinos A1 - Abdelhafez, Mohamed A1 - Walter, B. A1 - Meinikheim, Michael A1 - Chiu, Philip Wai Yan A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - Künstliche Intelligenz erhöht die Gefäßerkennung von Endoskopikern bei third space Endoskopie T2 - Zeitschrift für Gastroenterologie N2 - Einleitung: Künstliche Intelligenz (KI)-Algorithmen unterstützen Endoskopiker bei der Erkennung und Charakterisierung von Kolonpolypen in der klinischen Praxis und führen zu einer Erhöhung der Adenomdetektionsrate. Auch bei therapeutischen Maßnahmen wie der endoskopischen Submukosadissektion (ESD) könne relevante anatomische Strukturen durch KI mit hoher Genauigkeit erkannt und im endoskopischen Bild in Echtzeit markiert werden. Der Effekt einer solchen Applikation auf die Gefäßdetektion von Endoskopikern ist bislang nicht erforscht. Ziele:  In dieser Studie wurde der Effekt eines KI-Algorithmus zur Echtzeit-Gefäßmarkierung bei ESD auf die Gefäßdetektionsrate von Endoskopikern untersucht. Methodik:  59 third space Endoskopievideos wurde aus der Datenbank des Universitätsklinikums Augsburg extrahiert. Auf 5470 Einzelbildern dieser Untersuchungen wurde submukosale Blutgefäße annotiert. Zusammen mit weiteren 179681 unmarkierten Bildern wurde ein DeepLabV3+ neuronales Netzwerk mit einer semi-supervised learning Methode darin trainiert, submukosale Blutgefäße auf dem endoskopischen Bild zu erkennen und in Echtzeit einzuzeichnen. Anhand eines Videotests mit 101 Videoclips und 200 vordefinierten Blutgefäßen wurden 19 Endoskopiker mit und ohne KI Unterstützung getestet. Ergebnis:  Der Algorithmus erkannte in dem Videotest 93.5% der Gefäße in einer Detektionszeit von im Median 0,3 Sekunden. Die Gefäßdetektionsrate von Endoskopikern erhöhte sich durch KI Unterstützung von 56,4% auf 72,4% (p<0.001). Die Gefäßdetektionszeit reduzierte sich durch KI-Unterstützung von 6,7 auf 5.2 Sekunden (p<0.001). Der Algorithmus zeigte eine Rate an falsch positiven Detektionen in 4.5% der Einzelbilder. Falsch positiv erkannte Strukturen wurde kürzer detektiert, als richtig positive (0.7 und 6.0 Sekunden, p<0.001). Schlussfolgerung:  KI Unterstützung führte zu einer erhöhten Gefäßdetektionsrate und schnelleren Gefäßdetektionszeit von Endoskopikern. Ein möglicher klinischer Effekt auf die intraprozedurale Komplikationsrate oder Operationszeit könnte in prospektiven Studien ermittelt werden. Y1 - 2024 U6 - https://doi.org/10.1055/s-0044-1790087 VL - 62 IS - 09 SP - e830 PB - Georg Thieme Verlag KG ER - TY - GEN A1 - Roser, David A1 - Meinikheim, Michael A1 - Mendel, Robert A1 - Palm, Christoph A1 - Muzalyova, Anna A1 - Rauber, David A1 - Rückert, Tobias A1 - Parsa, Nasim A1 - Byrne, Michael F. A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - Mensch-Maschine-Interaktion: Einfluss künstlicher Intelligenz auf das diagnostische Vertrauen von Endoskopikern bei der Beurteilung des Barrett-Ösophagus T2 - Zeitschrift für Gastroenterologie N2 - Ziele:  Das Ziel der Studie war es, den Einfluss von KI auf die diagnostische Sicherheit (Konfidenzniveau) von Endoskopikern anhand von BÖ-Videos zu untersuchen und mögliche Korrelationen mit der Untersuchungsqualität zu erforschen. Methodik:  22 Endoskopiker aus zwölf Zentren mit unterschiedlicher Barrett-Erfahrung untersuchten 96 standardisierte Endoskopievideos. Die Untersucher wurden in Experten und Nicht-Experten eingeteilt und nach dem Zufallsprinzip für die Bewertung der Videos mit oder ohne KI eingeteilt. Die Teilnehmer wurden in zwei Gruppen aufgeteilt: Arm A bewertete zunächst Videos ohne KI und dann mit KI, während Arm B die umgekehrte Reihenfolge einhielt. Die Untersucher hatten die Aufgabe, BÖ-assoziierte Neoplasien zu erkennen und ihr Konfidenzniveau sowohl mit als auch ohne KI auf einer Skala von 0 bis 9 anzugeben. Ergebnis:  In Arm A erhöhte der Einsatz von KI das Konfidenzniveau bei beiden signifikant (p<0.001). Bemerkenswert ist, dass jedoch nur Nicht-Experten durch die KI eine signifikante Verbesserung der Sensitivität und Spezifität (p<0.001 bzw. p<0.05) erfuhren. Während Experten ohne KI im Vergleich zu Nicht-Experten mit KI ein höheres Konfidenzniveau aufwiesen, gab es keinen signifikanten Unterschied in der Genauigkeit. In Arm B zeigten beide Gruppen eine signifikante Abnahme des Konfidenzniveaus (p<0.001) bei gleichbleibender Genauigkeit. Darüber hinaus wurden in 9% der Entscheidungen trotz korrekter KI eine falsche Wahl getroffen. Schlussfolgerung:  Der Einsatz künstlicher Intelligenz steigerte das Konfidenzniveau sowohl bei Experten als auch bei Nicht-Experten signifikant – ein Effekt, der im Studienmodell reversibel war. Darüber hinaus wiesen Experten mit oder ohne KI durchweg höhere Konfidenzniveaus auf als Nicht-Experten mit KI, trotz vergleichbarer Ergebnisse. Zudem konnte beobachtet werden, dass die Untersucher in 9% der Fälle die KI zuungunsten des Patienten ignorierten. Y1 - 2024 U6 - https://doi.org/10.1055/s-0044-1789656 VL - 62 IS - 09 SP - e575 EP - e576 PB - Georg Thieme Verlag KG ER - TY - GEN A1 - Scheppach, Markus W. A1 - Nunes, Danilo Weber A1 - Arizi, X. A1 - Rauber, David A1 - Probst, Andreas A1 - Nagl, Sandra A1 - Römmele, Christoph A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - Intraoperative Phasenerkennung bei endoskopischer Submukosadissektion mit Hilfe von künstlicher Intelligenz T2 - Zeitschrift für Gastroenterologie N2 - Einleitung:  Künstliche Intelligenz (KI) wird in der Endoskopie des Gastrointestinaltraktes zur Erkennung und Charakterisierung von Kolonpolypen eingesetzt. Die Rolle von KI bei therapeutischen Maßnahmen wurde noch nicht eingehend untersucht. Eine intraprozedurale Phasenerkennung bei endoskopischer Submukoasdissektion (ESD) könnte die Erhebung von Qualitätsindikatoren ermöglichen. Weiterhin könnte diese Technologie zu einem tieferen Verständnis über die Eigenschaften der Prozedur führen und weiterführende Applikationen zur automatischen Dokumentation oder standardisiertem Training vorbereiten. Ziele: Ziel dieser Studie war die Entwicklung eines KI Algorithmus zur intraprozeduralen Phasenerkennung bei endoskopischer Submukosadissektion. Methodik:  2071546 Einzelbilder aus 27 ESD Videos in voller Länge wurden für die übergeordneten Klassen Diagnostik, Markierung, Nadelinjektion, Dissektion und Blutung, sowie die untergeordneten Klassen Endoskop-Manipulation, Injektion und Applikation von elektrischem Strom annotiert. Mit einem Trainingsdatensatz (898440 Einzelbilder, 17 ESDs) wurde ein Video Swin Transformer mit uniformer Stichprobenentnahme trainiert und intern validiert (769523 Einzelbilder, 6 ESDs). Neben der internen Validierung wurde der Algorithmus anhand von einem separaten Testdatensatz (403583 Einzelbilder, 4 ESDs) evaluiert. Ergebnis:  Der F1 Score des Algorithmus für alle Klassen lag in der internen Validierung bei 83%, in dem separaten Test bei 90%. Anhand des separaten Tests wurden true positive (TP)-Raten für Diagnostik, Markierung, Nadelinjektion, Dissektion und Blutung von 100%, 100%, 96%, 97% und 93% ermittelt. Für Endoskopmanipulation, Injektion und Applikation von Elektrizität lagen die TP-Raten bei 92%, 98% und 91%. Schlussfolgerung:  Der entwickelte Algorithmus klassifizierte ESD Videos in voller Länge und anhand jedes einzelnen Bildes mit hoher Genauigkeit. Zukünftige Forschungsvorhaben könnten intraoperative Qualitätsindikatioren auf Basis dieser Informationen entwickeln und eine automatisierte Dokumentation ermöglichen. Y1 - 2024 U6 - https://doi.org/10.1055/s-0044-1790084 VL - 62 IS - 09 SP - e828 PB - Georg Thieme Verlag KG ER - TY - GEN A1 - Zellmer, Stephan A1 - Rauber, David A1 - Probst, Andreas A1 - Weber, Tobias A1 - Nagl, Sandra A1 - Römmele, Christoph A1 - Schnoy, Elisabeth A1 - Palm, Christoph A1 - Messmann, Helmut A1 - Ebigbo, Alanna T1 - Verwendung künstlicher Intelligenz bei der Detektion der Papilla duodeni major T2 - Zeitschrift für Gastroenterologie N2 - Einleitung Die Endoskopische Retrograde Cholangiopankreatikographie (ERCP) ist der Goldstandard in der Diagnostik und Therapie von Erkrankungen des pankreatobiliären Trakts. Jedoch ist sie technisch sehr anspruchsvoll und weist eine vergleichsweise hohe Komplikationsrate auf. Ziele  In der vorliegenden Machbarkeitsstudie soll geprüft werden, ob mithilfe eines Deep-learning-Algorithmus die Papille und das Ostium zuverlässig detektiert werden können und somit für Endoskopiker mit geringer Erfahrung ein geeignetes Hilfsmittel, insbesondere für die Ausbildungssituation, darstellen könnten. Methodik Wir betrachteten insgesamt 606 Bilddatensätze von 65 Patienten. In diesen wurde sowohl die Papilla duodeni major als auch das Ostium segmentiert. Anschließend wurde eine neuronales Netz mittels eines Deep-learning-Algorithmus trainiert. Außerdem erfolgte eine 5-fache Kreuzvaldierung. Ergebnisse Bei einer 5-fachen Kreuzvaldierung auf den 606 gelabelten Daten konnte für die Klasse Papille eine F1-Wert von 0,7908, eine Sensitivität von 0,7943 und eine Spezifität von 0,9785 erreicht werden, für die Klasse Ostium eine F1-Wert von 0,5538, eine Sensitivität von 0,5094 und eine Spezifität von 0,9970 (vgl. [Tab. 1]). Unabhängig von der Klasse zeigte sich gemittelt (Klasse Papille und Klasse Ostium) ein F1-Wert von 0,6673, eine Sensitivität von 0,6519 und eine Spezifität von 0,9877 (vgl. [Tab. 2]). Schlussfolgerung  In vorliegende Machbarkeitsstudie konnte das neuronale Netz die Papilla duodeni major mit einer hohen Sensitivität und sehr hohen Spezifität identifizieren. Bei der Detektion des Ostiums war die Sensitivität deutlich geringer. Zukünftig soll das das neuronale Netz mit mehr Daten trainiert werden. Außerdem ist geplant, den Algorithmus auch auf Videos anzuwenden. Somit könnte langfristig ein geeignetes Hilfsmittel für die ERCP etabliert werden. KW - Künstliche Intelligenz Y1 - 2023 UR - https://www.thieme-connect.de/products/ejournals/abstract/10.1055/s-0043-1772000 U6 - https://doi.org/10.1055/s-0043-1772000 VL - 61 IS - 08 SP - e539 EP - e540 PB - Thieme CY - Stuttgart ER - TY - RPRT A1 - Schultz, Maximilian T1 - 1.Bericht für das Regensburg Center of Health Sciences and Technology (RCHST) N2 - Der vorliegende Bericht zeigt den aktuellen Stand des Projekts „Künstliche Intelligenz für Nichtregierungsorganisationen (KINiro) – Bedarf, Akzeptanz und Umsetzungsmöglichkeiten“ an der OTH Regensburg am Institut für Sozialforschung und Technikfolgenabschätzung (IST). Im Projektbericht wird das Projekt KINiro mit besonderem Blick auf die Erkenntnisse und Tätigkeiten im Gesundheitswesen aufgezeigt. Das Projekt KINiro wird mit Laufzeit Januar 2023 bis Dezember 2025 durch das Bundesministerium für Familie, Senioren, Frauen und Jugend (BMFSFJ) gefördert. Eine zusätzliche Unterstützung bietet das Regensburg Center of Health Sciences and Technology (RCHST) KW - Künstliche Intelligenz KW - Gesundheit Y1 - 2024 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-73125 VL - 2024 PB - OTH Regensburg CY - Regensburg ER - TY - JOUR A1 - Rueckert, Tobias A1 - Rauber, David A1 - Maerkl, Raphaela A1 - Klausmann, Leonard A1 - Yildiran, Suemeyye R. A1 - Gutbrod, Max A1 - Nunes, Danilo Weber A1 - Moreno, Alvaro Fernandez A1 - Luengo, Imanol A1 - Stoyanov, Danail A1 - Toussaint, Nicolas A1 - Cho, Enki A1 - Kim, Hyeon Bae A1 - Choo, Oh Sung A1 - Kim, Ka Young A1 - Kim, Seong Tae A1 - Arantes, Gonçalo A1 - Song, Kehan A1 - Zhu, Jianjun A1 - Xiong, Junchen A1 - Lin, Tingyi A1 - Kikuchi, Shunsuke A1 - Matsuzaki, Hiroki A1 - Kouno, Atsushi A1 - Manesco, João Renato Ribeiro A1 - Papa, João Paulo A1 - Choi, Tae-Min A1 - Jeong, Tae Kyeong A1 - Park, Juyoun A1 - Alabi, Oluwatosin A1 - Wei, Meng A1 - Vercauteren, Tom A1 - Wu, Runzhi A1 - Xu, Mengya A1 - Wang, An A1 - Bai, Long A1 - Ren, Hongliang A1 - Yamlahi, Amine A1 - Hennighausen, Jakob A1 - Maier-Hein, Lena A1 - Kondo, Satoshi A1 - Kasai, Satoshi A1 - Hirasawa, Kousuke A1 - Yang, Shu A1 - Wang, Yihui A1 - Chen, Hao A1 - Rodríguez, Santiago A1 - Aparicio, Nicolás A1 - Manrique, Leonardo A1 - Palm, Christoph A1 - Wilhelm, Dirk A1 - Feussner, Hubertus A1 - Rueckert, Daniel A1 - Speidel, Stefanie A1 - Nasirihaghighi, Sahar A1 - Al Khalil, Yasmina A1 - Li, Yiping A1 - Arbeláez, Pablo A1 - Ayobi, Nicolás A1 - Hosie, Olivia A1 - Lyons, Juan Camilo T1 - Comparative validation of surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation in endoscopy: Results of the PhaKIR 2024 challenge JF - Medical Image Analysis N2 - Reliable recognition and localization of surgical instruments in endoscopic video recordings are foundational for a wide range of applications in computer- and robot-assisted minimally invasive surgery (RAMIS), including surgical training, skill assessment, and autonomous assistance. However, robust performance under real-world conditions remains a significant challenge. Incorporating surgical context – such as the current procedural phase – has emerged as a promising strategy to improve robustness and interpretability. To address these challenges, we organized the Surgical Procedure Phase, Keypoint, and Instrument Recognition (PhaKIR) sub-challenge as part of the Endoscopic Vision (EndoVis) challenge at MICCAI 2024. We introduced a novel, multi-center dataset comprising thirteen full-length laparoscopic cholecystectomy videos collected from three distinct medical institutions, with unified annotations for three interrelated tasks: surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation. Unlike existing datasets, ours enables joint investigation of instrument localization and procedural context within the same data while supporting the integration of temporal information across entire procedures. We report results and findings in accordance with the BIAS guidelines for biomedical image analysis challenges. The PhaKIR sub-challenge advances the field by providing a unique benchmark for developing temporally aware, context-driven methods in RAMIS and offers a high-quality resource to support future research in surgical scene understanding. KW - Surgical phase recognition KW - Instrument keypoint estimation KW - Instrument instance segmentation KW - Robot-assisted surgery Y1 - 2026 U6 - https://doi.org/10.1016/j.media.2026.103945 SN - 1361-8415 N1 - Corresponding author der OTH Regensburg: Tobias Rueckert Die Preprint-Version ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/solrsearch/index/search/start/0/rows/10/sortfield/score/sortorder/desc/searchtype/simple/query/2507.16559 VL - 109 PB - Elsevier ER - TY - CHAP A1 - Schwägerl, Felix A1 - Uhrig, Sabrina A1 - Westfechtel, Bernhard T1 - A Graph-based Algorithm for Three-way Merging of Ordered Collections in EMF Models T2 - Science of Computer Programming N2 - In EMF models, ordered collections appear as the values of multi-valued structural features. Traditional, text-based version control systems do not sufficiently support three-way merging of ordered collections inside EMF models since they cannot guarantee a consistent result. The operation three-way merging is defined as follows: based on a common base version b, two alternative versions and were developed by copying and modifying the base version. To reconcile these changes, a merged version m is to be created as a common successor of and . In this paper, we present a graph algorithm to solve the problem of three-way merging of ordered collections in EMF models. Each version of a collection can be represented by means of a linearly ordered graph. To create the merged version, these graphs are combined to a merged collection graph using set formula. To create the merged collection, a generalized topological sort is performed on the merged collection graph. Conflicts occur in case the order of elements cannot be deduced automatically; these conflicts are resolved either interactively or by default rules. We have implemented the merge algorithm in our tool BTMerge, which performs a consistency-preserving three-way merge of versions of EMF models being instances of arbitrary Ecore models. Our implementation relies on an alternative form of representing multiple versions of a collection, namely a versioned collection graph which forms a superimposition of collection versions. The algorithm presented here is purely state-based. Matching and merging of collections are clearly separated sub-problems. Insertions and deletions performed on the elements of the collection are propagated into the merged version in a consistent way. Our algorithm makes only minimal assumptions with regard to the underlying product model and thus may be applied to ordered collections inside plain text or XML files. By taking arbitrary move operations into account, the algorithm considerably goes beyond the functionality of contemporary merge tools which cannot adequately handle move operations. KW - EMF Models KW - Model Merging KW - Graph Algorithms Y1 - 2014 U6 - https://doi.org/10.1016/j.scico.2015.02.008 VL - 113 IS - Part 1 SP - 51 EP - 81 ER - TY - CHAP A1 - Buchmann, Thomas A1 - Schwägerl, Felix T1 - Breaking the Boundaries of Meta Models and Preventing Information Loss in Model-Driven Software Product Lines T2 - Proceedings of the 11th International Conference on Evaluation of Novel Software Approaches to Software Engineering, Vol 1, 2016, Rome, Italy N2 - Model-driven software product line engineering is an integrating discipline for which tool support has become available recently. However, existing tools are still immature and have several weaknesses. Among others, limitations in variability, caused by meta model restrictions, and unintended information loss are not addressed. In this paper, we present two conceptual extensions to model-driven product line engineering based on negative variability, being alternative mappings and surrogates. Alternative mappings allow for unconstrained variability, mitigating meta model restrictions by virtually extending the underlying multi-variant domain model. Surrogates prevent unintended information loss during product derivation based on a contextsensitive product analysis, which can be controlled by a declarative OCL-based language. Both extensions have been implemented in FAMILE, a model-driven product line tool that is based on EMF, provides dedicated consistency repair mechanisms, and completely automates application engineering. The added value of alternative mappings and surrogates is demonstrated by a running example. KW - Software Product Lines KW - Model-Driven Development KW - Unconstrained Variability KW - Information Loss KW - Negative Variability KW - Application Engineering Y1 - 2016 U6 - https://doi.org/10.5220/0005789100730083 SP - 83 PB - SCITEPRESS - Science and and Technology Publications ER - TY - CHAP A1 - Schwägerl, Felix A1 - Uhrig, Sabrina A1 - Westfechtel, Bernhard T1 - Model-based tool support for consistent three-way merging of EMF models T2 - ACME '13, Proceedings of the workshop on ACadeMics Tooling with Eclipse N2 - Inadequate version control has been identified as a major obstacle to the application of model-driven software engineering. In particular, sophisticated support for merging model versions is urgently needed. We present a tool for merging of EMF models which may be applied to instances of arbitrary Ecore models. The tool advances the state of the art by guaranteeing a consistent merge result. Furthermore, it detects and resolves not only context-free, but also context-sensitive conflicts. Our merge tool is model-based; it relies on EMF for its implementation. Initial experiences gained from its application confirm that the merge tool operates both accurately and efficiently. Y1 - 2013 U6 - https://doi.org/10.1145/2491279.2491281 PB - ACM CY - New York, NY, USA ER - TY - RPRT A1 - Schwarz, Philipp A1 - Bock, Sebastian A1 - Weiß, Martin T1 - Abschlussbericht Projekt LeaP - Learning Poses : Posenerkennung mit Neuronalen Netzen BT - Programm zur Förderung der angewandten Forschung und Entwicklung an Hochschulen für angewandte Wissenschaften – Fachhochschulen (Kap. 15 49 TG 78), Förderperiode 2018 – 2021; Programmsäule Strukturimpuls Forschungseinstieg - Projekt “Posenerkennung mit Neuronalen Netzen” N2 - Die Schätzung der 6D-Pose bekannter Objekte findet Anwendungen in der Robotik, der Luft- und Raumfahrtsteuerung und automatisierten Produktionsumgebungen. Es ist nach wie vor gängige Praxis, klassische Bildverarbeitungsmethoden zu verwenden, um eine hohe Präzision zu erreichen. Diese Methoden erfordern jedoch eine manuelle Parametrisierung der verwendeten Erkennungswerkzeuge für jedes einzelne Objekt. Dies wirft die Frage auf, ob maschinelle Lerntechniken, insb. convolutional neural networks, Netzwerke, so trainiert werden können, dass sie die in der Industrie geforderte Präzision ohne individuelle Programmierung erreichen können. Darüber sollen in Produktionsumgebungen Umwelteinflüsse wie Lichtverhältnisse minimiert werden. Es werden verschiedene Netzstrukturen entwickelt und auf ihre Leistung bei diesem Szenario untersucht, mit 1 bis 6 Freiheitsgraden in der Aufgabenstellung und einer oder mehreren Kameras. Zusätzlich wird ein Ansatz mit reinforcement learning entwickelt und untersucht. KW - Posenermittlung KW - convolutional neural network KW - Konvergenzverhalten Y1 - 2022 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-88514 CY - Regensburg ER - TY - GEN A1 - Currle, Edda A1 - Haug, Sonja A1 - Weber, Karsten T1 - Künstliche Intelligenz und Anamnese (KI-Anamnese) — Fragebogen der Trendstudie zur Akzeptanz von KI in der Anamnese in Arztpraxen in Deutschland mit Codeplan N2 - Die Studie stützt sich auf eine Quotenstichprobe der deutschsprachigen Bevölkerung im Alter von 18 bis 74 Jahren. Sie wurde als Längsschnittstudie (Trendstudie) mit insgesamt zwei Wellen mit dem gleichen Fragebogen durchgeführt. Welle 1 war vom 18. bis 24. November 2024 im Feld, die zweite Befragungswelle vom 13. bis 17. November 2025. Mit der Datenerhebung wurde das Umfrageinstitut NielsenIQ-GfK beauftragt. Je 1.000 Personen wurden online im Rahmen des GfK eBUS®-Deutschland befragt. Die Teilnehmenden wurden mithilfe einer Kombination aus Online- und Offline-Methoden rekrutiert. Um eine möglichst genaue Abbildung der Grundgesamtheit zu gewährleisten, wurden die Merkmale Geschlecht, Alter, Region, Haushaltsgröße, Ortsgröße sowie der Bildungsstand des Haushaltsvorstands quotiert, etwaige Abweichungen durch ein iteratives Gewichtungsverfahren ausgeglichen. Der Fragebogen besteht aus neun standardisierten Fragen; zuzüglich wurden soziodemographische Merkmale erhoben. Einzelne Fragen wurden, zum Teil übersetzt sowie abgewandelt, aus anderen Studien übernommen. KW - Anamnese KW - Bevölkerungsumfrage KW - Künstliche Intelligenz KW - Technologieakzeptanz KW - Fragebogen Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-87079 N1 - Der Fragebogen ist Anhang zu folgender Veröffentlichung: Currle, E., Haug, S. & Weber, K. (2025). Akzeptanz des Einsatzes von Künstlicher Intelligenz im Anamneseprozess: Ergebnisse einer Bevölkerungsbefragung in Deutschland [Acceptance of artificial intelligence tools for medical history-taking: Findings of a population survey in Germany]. Zeitschrift für Evidenz, Fortbildung und Qualität im Gesundheitswesen, 198-199, 9–17. https://doi.org/10.1016/j.zefq.2025.10.003 Nachweis auf dem Publikationsserver unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/8671 Zugehöriges Datenset: 10.5281/zenodo.18017447 (DOI) CY - Regensburg ER - TY - GEN A1 - Rueckert, Tobias A1 - Rauber, David A1 - Klausmann, Leonard A1 - Gutbrod, Max A1 - Rueckert, Daniel A1 - Feussner, Hubertus A1 - Wilhelm, Dirk A1 - Palm, Christoph T1 - PhaKIR Dataset - Surgical Procedure Phase, Keypoint, and Instrument Recognition [Data set] N2 - Note: A script for extracting the individual frames from the video files while preserving the challenge-compliant directory structure and frame-to-mask naming conventions is available on GitHub and can be accessed here: https://github.com/remic-othr/PhaKIR_Dataset. The dataset is described in the following publications: Rueckert, Tobias et al.: Comparative validation of surgical phase recognition, instrument keypoint estimation, and instrument instance segmentation in endoscopy: Results of the PhaKIR 2024 challenge. arXiv preprint, https://arxiv.org/abs/2507.16559. 2025. Rueckert, Tobias et al.: Video Dataset for Surgical Phase, Keypoint, and Instrument Recognition in Laparoscopic Surgery (PhaKIR). arXiv preprint, https://arxiv.org/abs/2511.06549. 2025. The proposed dataset was used as the training dataset in the PhaKIR challenge (https://phakir.re-mic.de/) as part of EndoVis-2024 at MICCAI 2024 and consists of eight real-world videos of human cholecystectomies ranging from 23 to 60 minutes in duration. The procedures were performed by experienced physicians, and the videos were recorded in three hospitals. In addition to existing datasets, our annotations provide pixel-wise instance segmentation masks of surgical instruments for a total of 19 categories, coordinates of relevant instrument keypoints (instrument tip(s), shaft-tip transition, shaft), both at an interval of one frame per second, and specifications regarding the intervention phases for a total of eight different phase categories for each individual frame in one dataset and thus comprehensively cover instrument localization and the context of the operation. Furthermore, the provision of the complete video sequences offers the opportunity to include the temporal information regarding the respective tasks and thus further optimize the resulting methods and outcomes. Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.15740620 ER - TY - GEN A1 - Haug, Sonja A1 - Currle, Edda A1 - Weber, Karsten T1 - Künstliche Intelligenz und Anamnese (KI-Anamnese) - Datensätze Welle 1 und Welle 2 [Data set] N2 - Die Studie stützt sich auf eine Quotenstichprobe der deutschsprachigen Bevölkerung im Alter von 18 bis 74 Jahren. Sie wurde als Längsschnittstudie (Trendstudie) mit insgesamt zwei Wellen mit dem gleichen Fragebogen durchgeführt. Welle 1 war vom 18. bis 24. November 2024 im Feld, die zweite Befragungswelle vom 13. bis 17. November 2025. Mit der Datenerhebung wurde das Umfrageinstitut NielsenIQ-GfK beauftragt. Je 1.000 Personen wurden online im Rahmen des GfK eBUS®-Deutschland befragt. Die Teilnehmenden wurden mithilfe einer Kombination aus Online- und Offline-Methoden rekrutiert. Um eine möglichst genaue Abbildung der Grundgesamtheit zu gewährleisten, wurden die Merkmale Geschlecht, Alter, Region, Haushaltsgröße, Ortsgröße sowie der Bildungsstand des Haushaltsvorstands quotiert, etwaige Abweichungen durch ein iteratives Gewichtungsverfahren ausgeglichen. Der Fragebogen besteht aus neun standardisierten Fragen; zuzüglich wurden soziodemographische Merkmale erhoben. Einzelne Fragen wurden, zum Teil übersetzt sowie abgewandelt, aus anderen Studien übernommen. KW - Anamnese KW - Bevölkerungsumfrage KW - Künstliche Intelligenz KW - Technologieakzeptanz Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.18017447 N1 - Related works: Is described by Other: 10.35096/othr/pub-8707 (DOI) Is metadata for Journal: 10.1016/j.zefq.2025.10.003 (DOI) ER - TY - GEN A1 - Hauer, Franziska A1 - Schultz, Maximilian A1 - Haug, Sonja A1 - Weber, Karsten T1 - Künstliche Intelligenz in Nichtregierungsorganisationen - quantitativer Datensatz [Data set] N2 - Künstliche Intelligenz (KI) stellt ein wichtiges Thema der Digitalisierung dar und hat das Potenzial viele Arbeitsfelder grundlegend zu verändern. Mit dem Einsatz von KI wird die Hoffnung verbunden durch Automatisierung Zeit und Geld zu sparen. Dies könnte besonders interessant für Organisationen sein, die wenig Ressourcen zur Verfügung haben, da sie nicht gewinnorientiert oder ehrenamtlich arbeiten. Dazu zählen Nichtregierungsorganisationen (NRO), die einen wichtigen Bestandteil der Zivilgesellschaft darstellen. Aufgrund der komplexer werdenden Arbeit von NROs scheint KI Möglichkeiten zur Bewältigung aktueller und zukünftiger Herausforderungen zu bieten. Über den Einsatz von KI in NROs ist derzeit jedoch wenig bekannt. Es wurde eine quantitative Querschnittstudie deutscher Nichtregierungsorganisationen (NRO) zum Thema KI durchgeführt. Der Web-Survey baut auf den Erkenntnissen des Scoping Reviews und der qualitativen Interviews auf und vertieft den Einblick in die Bereiche aktuelle Nutzung von KI, Wissenstand in den Organisationen sowie vorhandene und noch benötigte Ressourcen. Dafür wurden 343 NROs verschiedener Größen und Handlungsfelder befragt und die Ergebnisse quantitativ analysiert. Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.17977270 N1 - Related works: Continues Working paper: 10.13140/RG.2.2.21668.69765 (DOI) Working paper: 10.13140/RG.2.2.10671.75685/1(DOI) Is continued by Working paper: 10.13140/RG.2.2.25460.46720 (DOI) Is described by Working paper: 10.13140/RG.2.2.11094.66883 (DOI) Journal article: 10.5771/1613-0707-2025-5-35 (DOI) Is referenced by Conference paper: 10.3233/SHTI250172 (DOI) ER - TY - GEN A1 - Gutbrod, Max A1 - Rauber, David A1 - Weber Nunes, Danilo A1 - Palm, Christoph T1 - A cleaned subset of the first five CATARACTS test videos [Data set] N2 - This dataset is a subset of the original CATARACTS test dataset and is used by the OpenMIBOOD framework to evaluate a specific out-of-distribution setting. When using this dataset, it is mandatory to cite the corresponding publication (OpenMIBOOD (10.1109/CVPR52734.2025.02410)) and follow the acknowledgement and citation requirements of the original dataset (CATARACTS). The original CATARACTS dataset (associated publication,Homepage) consists of 50 videos of cataract surgeries, split into 25 train and 25 test videos. This subset contains the frames of the first 5 test videos. Further, black frames at the beginning of each video were removed. Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.14924735 N1 - Related works: Is derived from: Dataset: 10.21227/ac97-8m18 (DOI) Software: Repository URL: https://github.com/remic-othr/OpenMIBOOD ER - TY - JOUR A1 - Weber, Karsten T1 - Public images of artificial intelligence: an overview JF - media-culture-social communication N2 - Artificial intelligence (AI) is not only a ubiquitous topic in scholarly debates but in recent years has also received increased media and political attention. Discussions, initially strongly influenced by autonomous vehicles but currently dominated by expectations concerning generative AI such as ChatGPT, have contributed to public awareness. There are debate cycles, often characterized by a significant decline in public attention after a certain period of time, and the issue to a large extent or completely disappears from mass media’s agenda; also, they are often specific to certain countries. In this paper, a cursory cross-country overview concerning studies on media representations and the public perception of AI is provided. The use of AI in healthcare will be particularly highlighted, as there are currently high expectations regarding the benefits of AI, but other areas of application will also be considered. It will be tentatively concluded that in many countries, public perception of and public attitudes toward AI often is based on superficial knowledge and even prejudices. Y1 - 2025 U6 - https://doi.org/10.31648/mcsc.10343 IS - 21 SP - 9 EP - 24 PB - Wydawnictwo Uniwersytetu Warmińsko-Mazurskiego w Olsztynie CY - Olsztyn ER - TY - GEN A1 - Mauerer, Wolfgang A1 - Joblin, Mitchell A1 - Tamburri, Damian Andrew A1 - Paradis, Carlos A1 - Kazman, Rick A1 - Apel, Sven T1 - In Search of Socio-Technical Congruence: A Large-Scale Longitudinal Study [Data set] Y1 - 2021 U6 - https://doi.org/10.5281/zenodo.4766388 N1 - Archive of the accompanying website including all datasets for "In Search of Socio-Technical Congruence: A Large-Scale Longitudinal Study" (IEEE Transactions on Software Engineering): https://doi.org/10.1109/TSE.2021.3082074 ER - TY - GEN A1 - Mauerer, Wolfgang A1 - Scherzinger, Stefanie T1 - Nullius in Verba - Reproducibility Tutorial [Data set] N2 - Reproduction package (docker image) for the ICDE 2021 tutorial "Nullius in Verba". Y1 - 2021 U6 - https://doi.org/10.5281/zenodo.4730023 N1 - In Verbindung mit dem Konferenzbeitrag: Nullius in Verba: Reproducibility for Database Systems Research, Revisited - https://doi.org/10.1109/ICDE51399.2021.00270 ER - TY - GEN A1 - Schmidbauer, Lukas A1 - Lobe, Elisabeth A1 - Schäfer, Ina A1 - Mauerer, Wolfgang T1 - It's Quick to be Square: Fast Quadratisation for Quantum Toolchains [Data set] N2 - Many of the envisioned use-cases for quantum computers involve optimisation processes. While there are many algorithmic primitives to perform the required calculations, all eventually lead to quantum gates operating on quantum bits, with an order as determined by the structure of the objective function and the properties of target hardware. When the structure of the problem representation is not aligned with structure and boundary conditions of the executing hardware, various overheads to degrade the computation may arise, possibly negating any possible quantum advantage. Therefore, automatic transformations of problem representations play an important role in quantum computing when descriptions (semi-)targeted at humans must be cast into forms that can be executed on quantum computers. Mathematically equivalent formulations are known to result in substantially different non-functional properties depending on hardware, algorithm and detail properties of the problem. Given the current state of noisy-intermediate scale quantum hardware (NISQ), these effects are considerably more pronounced than in classical computing. Likewise, efficiency of the transformation itself is relevant because possible quantum advantage may easily be eradicated by the overhead of transforming between representations. In this paper we consider a specific class of higher-level representations (polynomial unconstrained binary optimisation problems), and devise novel automatic transformation mechanisms into widely used quadratic unconstrained binary optimisation problems that substantially improve efficiency and versatility over the state of the art. We also identify what influence factors of lower-level details can be abstracted away in the transformation process, and which details must be made available to higher-level abstractions. Y1 - 2024 U6 - https://doi.org/10.5281/zenodo.14245588 N1 - Paper: 10.48550/arXiv.2411.19934 Software: https://github.com/lfd/Fast_Quadratisation ER - TY - GEN A1 - Schmidbauer, Lukas A1 - Mauerer, Wolfgang T1 - SAT Strikes Back: Parameter and Path Relations in Quantum Toolchains [Data set] N2 - In the foreseeable future, toolchains for quantum computing should offer automatic means of transforming a high level problem formulation down to a hardware executable form. Thereby, it is crucial to find (multiple) transformation paths that are optimised for (hardware specific) metrics. We zoom into this pictured tree of transformations by focussing on k-SAT instances as input and their transformation to QUBO, while considering structure and characteristic metrics of input, intermediate and output representations. Our results can be used to rate valid paths of transformation in advance—also in automated (quantum) toolchains. We support the automation aspect by considering stability and therefore predictability of free parameters and transformation paths. Moreover, our findings can be used in the manifesting era of error correction (since considering structure in a high abstraction layer can benefit error correcting codes in layers below). We also show that current research is closely linked to quadratisation techniques and their mathematical foundation. Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.15464391 N1 - Repositoriy URL: https://github.com/lfd/QSW25-SAT-Strikes-Back ER - TY - GEN A1 - Moser, Elisabeth A1 - Eberhardt, Matthias A1 - Meyer, Selina A1 - Schmidhuber, Maximilian A1 - Ketterer, Daniel T1 - ARGO ship classification dataset [Data set] N2 - The ARGO ship classification dataset holds 1750 labelled images from PlanetScope-4-Band satelites. The dataset creation process and results on the dataset are published in the demo paper: {CITE} The imagery is provided as numpy binary files. All image data is licensed by Planet Labs PBC. The channel ordering is BGRN. The dataset is provided in two folders named "ship" and "non_ship". Those folders correspond to the original labels created during automated dataset creation. The files are numbered. Two additional .csv files are provided. The shipsAIS_2017_Zone17.csv file holds the AIS information on the imagery contained in the ship folder. The data was retrieved from marinecadastre.gov. During the experiments errors in the automatically created dataset emerged which are further described in the paper. The manual relabelling is supplied in the corrected_labels.csv file. Y1 - 2022 U6 - https://doi.org/10.5281/zenodo.6058710 ER - TY - CHAP A1 - Deutel, Mark A1 - Mallah, Maen A1 - Wissing, Julio A1 - Scheele, Stephan ED - Vermesan, Ovidiu ED - Pagani, Alain ED - Meloni, Paolo T1 - Recent Trends in Edge AI: Efficient Design, Training and Deployment of Machine Learning Models T2 - Charting the Intelligence Frontiers – Edge AI Systems Nexus N2 - With a rising demand for ubiquitous smart systems, processing and interpreting large quantities of data generated on the edge at a high velocity is becoming an increasingly important challenge. Machine learning (ML) models such as Deep Neural Networks (DNNs) are an essential tool of today’s artificial intelligence due to their ability to make accurate predictions given complex tasks and environments. However, Deep Learning is computationally complex and energy intensive. This seems to contradict the characteristics of many edge devices, which have only limited memory, computational resources, and energy budget available. To overcome this challenge, an efficient ML model design is crucial that incorporates available optimization techniques from hardware, software, and methodological perspective to enable energy-efficient deployment and operation on the edge. This work comprehensively summarizes recent techniques for training, optimizing, and deploying ML models targeting edge devices. We discuss different strategies for finding deployable ML models, scalable DNN architectures, neural architecture search, and multi-objective optimization approaches, to enable feasible trade-offs considering available resources and latency. Furthermore, we give insight into DNN compression methods such 182as quantization and pruning. We conclude by investigating different forms of cascaded processing, from simple multi-level approaches to highly branched compute graphs and early-exit DNNs. Y1 - 2026 SN - 9788743808862 U6 - https://doi.org/10.1201/9788743808862-9 SP - 181 EP - 220 PB - River Publishers CY - New York ER - TY - GEN A1 - Haug, Sonja A1 - Vetter, Miriam A1 - Dotter, Caroline A1 - Weber, Karsten T1 - Forschungsprojekt „EVEKT „Erhöhung der Verbraucherpartizipation an der Energiewende und datenbasierte Mehrwertdienste“ — Fragebogen zur Online-Befragung und Codeplan N2 - Um Leistungsschwankungen im Stromnetz durch fluktuierende erneuerbare Energien auszugleichen, sollen in Deutschland intelligente Messtechnologien in Privathaushalten verbaut werden. Diese gelten als zentraler Bestandteil sog. Smart-Grids (intelligente Stromnetze), die Stromnachfrage und -angebot steuern, um Stromnetzstabilität zu gewährleisten. Die Werte, die bei der Messung des Stromverbrauchs in Privathaushalten erhoben werden, können an die Verbraucher*innen mithilfe von Apps oder anderer Smart-Meter-Plattformen (datenbasierte Mehrwertdienste) rückgemeldet werden. Vor diesem Hintergrund wurde 2023 eine Online-Befragung der Wohnbevölkerung in Deutschland durchgeführt (n=2.027). Untersucht wurden Bekanntheit, Nutzungsbereitschaft und Akzeptanz intelligenter Stromzähler bzw. Smart-Meter sowie die Nutzungsintention bei Anwendungsszenarien und Datenschutzbedenken. Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-87409 N1 - Vetter, Miriam/Haug, Sonja/Dotter, Caroline/Weber, Karsten (2024). 4. Arbeitspapier. Ak-zeptanz und Nutzungsbereitschaft von Smart-Meter-Anwendungen und datenbasierten Mehrwertdiensten. Erste Auswertungen im Rahmen von EVEKT. Ostbayerische Technische Hochschule (OTH) Regensburg. Regensburg. EVEKT – Erhöhung der Verbraucherpartizipa-tion an der Energiewende durch KI-Technologien und datenbasierte Mehrwertdienste. Teil-projekt Ethische und soziale Aspekte der Erhöhung der Verbraucherpartizipation an der Energiewende durch KI-Technologien. https://doi.org/10.13140/RG.2.2.33227.60968 Zugehöriges Datenset: https://doi.org/10.5281/zenodo.17962728 CY - Regensburg ER - TY - INPR A1 - Feeß, Alexander A1 - Weiß, Martin T1 - Uniqueness Domains of Offset Wrist Robots via a 4D Workspace Representation N2 - We present a new method for studying the workspace of 6R robots using a four-dimensional representation. This is applied to numerically determine connected workspace components with a constant number of solutions of the inverse kinematics, and to describe critical values that form the boundaries these regions. To define and maximize uniqueness domains, nonsingular transition of these boundaries are investigated. This provides a method to determine workspace regions in which the robot can move avoiding configuration changes and singularities. KW - kinematic workspace analysis KW - kinematic singularities KW - cuspidal robot Y1 - 2026 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-88635 CY - Regensburg ER - TY - INPR A1 - Feeß, Alexander A1 - Weiß, Martin T1 - Analysis of a Cuspidal 6R Robot N2 - We present a theoretical and numerical analysis of the kinematics for the Transpressor, a cuspidal 6R robot. It admits up to 16 inverse kinematics solutions which are described geometrically. For special target poses, we provide the solutions analytically and present a simple numerical solver for the general case. Moreover, an analytical estimate of the Jacobian determinant on a path between two solutions proves cuspidality for a class of robots similar to the transpressor. KW - Kinematik KW - Industrierobotik KW - inverse kinematics KW - kinematic singularities Y1 - 2026 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-88626 CY - Regensburg ER - TY - GEN A1 - Haug, Sonja A1 - Vetter, Miriam A1 - Dotter, Caroline A1 - Weber, Karsten T1 - Forschungsprojekt "EVEKT "Erhöhung der Verbraucherpartizipation an der Energiewende und datenbasierte Mehrwertdienste" — Datensatz [Data set] N2 - Um Leistungsschwankungen im Stromnetz durch fluktuierende erneuerbare Energien auszugleichen, sollen in Deutschland intelligente Messtechnologien in Privathaushalten verbaut werden. Diese gelten als zentraler Bestandteil sog. Smart-Grids (intelligente Stromnetze), die Stromnachfrage und -angebot steuern, um Stromnetzstabilität zu gewährleisten. Die Werte, die bei der Messung des Stromverbrauchs in Privathaushalten erhoben werden, können an die Verbraucher*innen mithilfe von Apps oder anderer Smart-Meter-Plattformen (datenbasierte Mehrwertdienste) rückgemeldet werden. Vor diesem Hintergrund wurde 2023 eine Online-Befragung der Wohnbevölkerung in Deutschland durchgeführt (n=2.027). Untersucht wurden Bekanntheit, Nutzungsbereitschaft und Akzeptanz intelligenter Stromzähler bzw. Smart-Meter sowie die Nutzungsintention bei Anwendungsszenarien und Datenschutzbedenken. KW - Smart Meter KW - Technikfolgenabschätzung KW - Energiesparen KW - Datenschutzsorgen KW - Bevölkerungsbefragung KW - Akzeptanz Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.17962728 N1 - Related works: Is derived from: Working paper: 10.13140/RG.2.2.11998.72009 (DOI) Working paper: 10.13140/RG.2.2.11159.85925 (DOI) Is described by: Other: 10.35096/othr/pub-8740 (DOI) Is metadata for: Working paper: 10.13140/RG.2.2.33227.60968 (DOI) Project deliverable: 10.34657/20059 (DOI) Working paper: 10.13140/RG.2.2.17667.11043 (DOI) Journal: 10.1007/s11623-025-2065-8 Titel anhand dieser DOI in Citavi-Projekt übernehmen (DOI) ER - TY - JOUR A1 - Grabinger, Lisa A1 - Hauser, Florian A1 - Wolff, Christian A1 - Mottok, Jürgen T1 - On Eye Tracking in Software Engineering JF - SN Computer Science N2 - Eye tracking is becoming more and more important as a research method within the field of software engineering (SE). Existing meta-analyses focus on the design or conduct of SE eye tracking studies rather than the analysis phase. This article attempts to fill this gap; it presents a systematic literature review of eye tracking studies in the field of SE—focusing mainly on the data analysis methods used. From the IEEE Xplore and ACM digital libraries we gather 125 papers up to the first quarter of 2024. Detailed evaluation provides information on the number of papers that use specific methods of analysis (i.e., descriptive or inferential statistics, and gaze visualization) or settings (e.g., sample size, technical setup, and selected aspects of research design). With the data obtained we can infer the popularity of specific analysis methods in the field. Those results enable efficient work on data analysis tools or education of aspiring researchers and can serve as basis for standardization or guidelines within the community—providing for methods to include as well as current inconsistencies. KW - Eye tracking KW - Software engineering KW - Empirical KW - Systematic literature review Y1 - 2024 U6 - https://doi.org/10.1007/s42979-024-03045-3 SN - 2661-8907 N1 - Zugehöriges Datenset: doi:10.5281/zenodo.11279180 VL - 5 IS - 6 PB - Springer CY - Singapore ER - TY - JOUR A1 - Zellner, Gregor A1 - Paul, Dennis T1 - Ein strukturierter Vergleich ausgewählter Process Mining Tools JF - BIT Banking and Information Technology N2 - Die Notwendigkeit zur Optimierung und Automatisierung von Ablaufen in Unternehmen nimmt mit zunehmender Digitalisierung an Fahrt auf. Zur Unterstützung dieses Trends haben sich in den letzten Jahren einige Tools im Themenbereich Process Mining entwickelt und etabliert. Ziel dieser Untersuchung ist es, systematisch ausgewählte Tools anhand eines vorhandenen, aber weiterentwickelten Kriterienkatalogs zu vergleichen und damit dem Anwender eine strukturierte Einsicht in den aktuellen Markt der Process Mining Tools zu geben. Aus einer Auswahl von 44 Tools werden acht Tools genauer untersucht. Die letztendliche Auswahl hängt von vielen unternehmensindividuellen Faktoren ab, so dass kein pauschaler Vorschlag für das geeignetste Tool gegeben werden kann. KW - Process Mining KW - Vergleich KW - Tools Y1 - 2023 VL - Band 24, Oktober 2023 IS - 2 ER - TY - CHAP A1 - Bock, Sebastian A1 - Schwarz, Philipp A1 - Weiß, Martin G. ED - Yang, Xin-She T1 - U-Shape Phenomenon with Gaussian Noise and Clipped Inputs T2 - Proceedings of Eighth International Congress on Information and Communication Technology. - (ICICT 2023, London, Volume 3) N2 - Object detection and recognition is a core task in machine vision. Both the position and the rotation of the objects are of interest. Common algorithms for object recognition based on neural networks use a regression formulation and are aimed at changing environments. In industrial applications, the setting is different: One would like to determine with high accuracy, the pose of known objects. We are investigating a prototype application in this field, where we identify the rotation of an object in a plane. We use common network architectures and provide an overview of how well they are suited for this purpose. In addition, we compare our results with standard image processing algorithms, paying particular attention to the accuracy with metrics relevant in the industry. Moreover, an U-shape phenomenon with disturbed data occurred during the experiments, which is partially explained analytically. KW - Neural networks KW - Pose estimation KW - U-shape Y1 - 2023 SN - 978-981-99-3042-5 U6 - https://doi.org/10.1007/978-981-99-3043-2_45 SP - 569 EP - 579 PB - Springer CY - Singapore ER - TY - JOUR A1 - Zellner, Gregor T1 - Ein Grundriss der Wissenschaftsdisziplin Wirtschaftsinformatik JF - BIT Banking and Information Technology N2 - Dieser Beitrag gibt einen Grundriss der Wissenschaftsdisziplin Wirtschaftsinformatik (WI) und ihrer Entwicklung. Hierzu wird zunächst der Gegenstand der WI näher betrachtet, nämlich die Informationssysteme. Dann wird erläutert, welche Ziele die WI verfolgt. Da WI ein interdisziplinäres Fach ist, werden anschließend die bestimmenden Basiswissenschaften der WI im Rahmen des Grundsystems der Wissenschaften dargestellt. Dabei wird deutlich, dass der Einfluss unterschiedlicher Disziplinen den Methodenpluralismus in der WI befördert. Im Beitrag werden die zwei grundlegenden Forschungsparadigmen in Form des verhaltensorientierten und des gestaltungsorientierten Ansatzes näher betrachtet. Nach einem kurzen Einblick in die zeitliche Entwicklung der deutschsprachigen WI werden die Themen-und Tätigkeitsfelder der WI näher erörtert. Abschließend wird ein Ausblick auf die Herausforderungen in der WI gegeben. KW - Wissenschaftsdisziplin Wirtschaftsinformatik Y1 - 2024 VL - Band 25, März 2024 IS - 1 ER - TY - GEN A1 - Gutbrod, Max A1 - Rauber, David A1 - Weber Nunes, Danilo A1 - Palm, Christoph T1 - Cropped single instrument frames subset from Cholec80 [Data set] N2 - This dataset is a subset of the original Cholec80 dataset and is used by the OpenMIBOOD framework to evaluate a specific out-of-distribution setting. When using this dataset, it is mandatory to cite the corresponding publication (OpenMIBOOD) and to follow the acknowledgement and citation requirements of the original dataset (Cholec80). The original Cholec80 dataset (associated paper,Homepage) consists of 80 cholecystectomy surgery videos recorded at 25 fps, performed by 13 surgeons. It includes phase annotations (25 fps) and tool presence labels (1 fps), with phase definitions provided by a senior surgeon. A tool is considered present if at least half of its tip is visible. The dataset categorizes tools into seven types: Grasper, Bipolar, Hook, Scissors, Clipper, Irrigator, and Specimen bag. Multiple tools may be present in each frame. Additionally, 76 of the 80 videos exhibit a strong black vignette. For this dataset subset, frames were extracted based on tool presence labels, selecting only those containing Grasper, Bipolar, Hook, or Clipper while ensuring that only a single tool appears per frame. To enhance visual consistency, the black vignette was removed by extracting an inner rectangular region, where applicable. KW - Tool Presence Detection KW - Cholecystectomy KW - Laparoscopic KW - Deep Learning KW - Out-Of-Distribution Detection Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.14921670 N1 - Related works Is derived from Journal article: 10.1109/TMI.2016.2593957 Software Repository URL https://github.com/remic-othr/OpenMIBOOD ER - TY - GEN A1 - Gutbrod, Max A1 - Rauber, David A1 - Weber Nunes, Danilo A1 - Palm, Christoph T1 - OpenMIBOOD's classification models for the MIDOG, PhaKIR, and OASIS-3 benchmarks [Data set] N2 - These models are provided for evaluating post-hoc out-of-distribution methods on the three OpenMIBOOD benchmarks: MIDOG, PhaKIR, and OASIS-3. When using these models, make sure to give appropriate credit and cite the OpenMIBOOD publication. Y1 - 2025 U6 - https://doi.org/10.5281/zenodo.14982267 N1 - Software Repository URL https://github.com/remic-othr/OpenMIBOOD ER - TY - THES A1 - Stadler, Sebastian Rudolf Johann T1 - The integration of virtual reality into the design process T1 - Die Integration von virtueller Realität in den Designprozess N2 - Technological advances strive towards automation and digitalization with the aim to increase efficiency in terms of decreased costs and product development time, resulting in shorter product life cycles. As this causes industries to release products more frequently and more quickly, industrial designers face problems in developing more complex systems in a decreased amount of time while aiming to meet individualized user requirements and stakeholder demands. If prototypes are infeasible or unaffordable, a successful market implementation of products is jeopardized. As design history proves, the introduction of new technologies to the profession of design has the potential to revolutionize the way of working for industrial designers. Virtual Reality may be one of these revolutions, as Computer-Aid Design has been in the 1980s. The present dissertation investigates the integration of Virtual Reality and its impact on the design process, the respective design method, but also the designer and design as a profession. After defining industrial design and Virtual Reality, a basic framework is proposed that describes a design process consensus model, the transfer of conventional design methods into VR, as well as a preliminary set of criteria for evaluating design studies. Based on this framework, four research questions are defined that initiate the investigation of how Virtual Reality can impact the design process, design methods, the role of designers, and design as a profession. To allow this, a refined set of criteria for evaluating design studies is developed and a benchmark study is conducted that proves the suitability for using Virtual Reality in the chosen context. Subsequently, five design studies are conducted in the field of autonomous mobility for public transport in Singapore to investigate the previously defined questions. The five design methods that are combined with Virtual Reality are heuristic evaluation, surveys, Computer-Aided Design prototyping, usability testing, and product presentation. For each design study, the development of the VR apparatus as well as the data collection and test of the method is explained. Additionally, two conventional design methods are consulted for a subsequent comparative analysis of each VR methods. Beyond the definition of the refined evaluation criteria, consisting of safety, validity, time and cost, variability, interactivity, immersion, and effort, the suitability of using VR for behavioural observation in the present application domain is confirmed. Based on the present investigations, the Virtual Design Toolbox is presented. The toolbox consists of five method cards that explain the usage of Virtual Reality for industrial designers throughout the product development and function as support for successfully developing and applying design methods utilizing this technology. The case studies prove that the usage of Virtual Reality has the potential to impact the design process, as divergent thinking and convergent thinking is fostered through valid data collection, decreased time- and cost- involvement, and increased variability. The roles of the involved designers throughout all design studies shifted from being concept creators towards coordinators and facilitators. Implications are derived on different levels, starting with the technology of Virtual Reality that could be a future game changer for research and design, followed by increased motivation to participate in design studies, the need for interdisciplinary teams to establish the required expertise, and lastly integration of Virtual Reality in design education. The usage of Virtual Reality in design will not replace conventional design methods but has the potential to lead to a paradigm shift for the whole profession of design. N2 - Der technologische Fortschritt entwickelt sich in Richtung Automatisierung und Digitalisierung, was zu erhöhter Effizienz und damit zu verkürzten Produktentwicklungszeiten und verkürzten Produktlebenszyklen führt. Da dies die Industrie dazu veranlasst, Produkte häufiger und schneller zur Marktreife zu bringen, stehen Industriedesigner vor dem Problem, komplexere Systeme in kürzerer Zeit entwickeln zu müssen und gleichzeitig die individuellen Anforderungen der Nutzenden und anderer Stakeholder zu erfüllen. Wenn Prototypen nicht fertigbar oder nicht bezahlbar sind, ist eine erfolgreiche Markteinführung gefährdet. Wie die Designgeschichte beweist, hat die Implementierung neuartiger Technologien in den Beruf des Designers das Potenzial, die Arbeitsweise des Industriedesigners zu revolutionieren. In der vorliegenden Dissertation wird daher die Integration und die Auswirkung von Virtual Reality auf den Designprozess, Designmethoden, aber auch den Designer und das Design als Profession untersucht. Nach der Definition von Industriedesign und Virtual Reality wird ein grundlegendes Vorgehen vorgeschlagen, das ein Konsensmodell für den Designprozess, die Übertragung konventioneller Designmethoden in die VR sowie vorläufige Evaluierungskriterien zur Bewertung von Designstudien beschreibt. Basierend auf diesem Vorgehen werden vier Forschungsfragen definiert, die die Untersuchung einleiten, wie Virtual Reality den Designprozess, Designmethoden, die Rolle von Designern und Design als Profession beeinflussen kann. Um dies zu ermöglichen, wird mit Hilfe von Experteninterviews ein ausgearbeitetes Set von Evaluierungskriterien zur Bewertung von Designstudien entwickelt und eine Benchmark-Studie durchgeführt, die die Eignung für die Verwendung von Virtual Reality in der gewählten Applikationsdomäne belegt. Anschließend werden fünf Designstudien im Bereich autonomer Mobilität für den öffentlichen Verkehr in Singapur durchgeführt, um die zuvor definierten Forschungsfragen zu untersuchen. Die fünf Designmethoden, die mit Virtual Reality kombiniert werden, sind heuristische Evaluation, Umfragen, Computer-Aided-Design Prototyping, Usability-Tests und Produktpräsentation. Für jede Designstudie wird die Entwicklung des VR Apparats sowie die Datenerhebung und der Test der Methode erläutert. Zusätzlich werden zwei konventionelle Designmethoden für eine anschließende vergleichende Analyse herangezogen. Nach der Definition der verfeinerten Bewertungskriterien, bestehend aus Sicherheit, Validität, Zeit und Kosten, Variabilität, Interaktivität, Immersion und Aufwand, wird die Eignung der Verwendung von Virtual Reality zur Datenerhebung in der vorliegenden Anwendungsdomäne bestätigt. Basierend auf den vorliegenden Untersuchungen wird die Virtual Design Toolbox vorgestellt. Die Toolbox besteht aus fünf Methodenkarten, die den Einsatz von Virtual Reality für Industriedesigner während der Produktentwicklung erklären und als Unterstützung für die erfolgreiche Entwicklung und Anwendung von Designmethoden mit dieser Technologie dienen. Die Fallstudien belegen, dass der Einsatz von Virtual Reality das Potenzial hat, den Designprozess zu beeinflussen, da divergentes und konvergentes Denken durch valide Datenerhebung, verringerten Zeit- und Kostenaufwand und erhöhte Variabilität gefördert wird. Die Rollen der beteiligten Designer verschoben sich in allen Designstudien von der Rolle des Konzepterstellers hin zum Koordinator und Moderator. Implikationen werden auf verschiedenen Ebenen abgeleitet, beginnend mit der Technologie von Virtual Reality, die ein zukünftiger Game-Changer für Forschung und Design sein könnte, gefolgt von der gesteigerten Motivation, an Designstudien teilzunehmen, der Notwendigkeit interdisziplinärer Teams, um die erforderliche Expertise zu etablieren, und schließlich der Integration von Virtual Reality in die Designausbildung. Der Einsatz von Virtual Reality im Design wird herkömmliche Designmethoden nicht ersetzen, hat jedoch das Potenzial, zu einem Paradigmenwechsel für das gesamte Berufsfeld des Designs zu führen. Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?https://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:91-diss-20211103-1612177-1-3 UR - https://www.researchgate.net/publication/356638451_The_Integration_of_Virtual_Reality_into_the_Design_Process UR - https://mediatum.ub.tum.de/?id=1612177 ER - TY - JOUR ED - Herrmann, Frank T1 - Production Scheduling and Planning in Manufacturing Systems T2 - Systems KW - planning hierarchy KW - lot-sizing KW - scheduling KW - machine learning, neuronal networks KW - (meta) heuristics Y1 - 2025 SN - 2079-8954 N1 - Die Ausgabe 'Production Scheduling and Planning in Manufacturing Systems' ist eine Spezialausgabe des Online-Journals 'Systems' mit Prof. Dr. Frank Herrmann als Gast-Editor. VL - 2025 PB - MDPI CY - Basel ER - TY - THES A1 - Kratzer, Simon T1 - The Fractional CIO as a new form of IT management for SMEs : essays on current issues and new developments in IT executive management N2 - For decades, the importance of information technology (IT) for organizations has continuously increased. To ensure effective IT management and operations, to enable the business, and to stay competitive, many large organizations appoint a Chief Information Officer (CIO) as their highest-ranked IT executive (Preston et al., 2008). While many small to medium-sized enterprises (SMEs) face similar challenges concerning IT, they often lack the resources to employ a full-time IT executive (Bhagwat & Sharma, 2007; Cragg et al., 2013). However, Kamariotou and Kitsios (2022) find that strategic planning and alignment with IT are important determinants for increasing an SME’s business value. An increasing number of SMEs started to employ CIOs part-time to avoid the risk of not staying competitive and losing business value (Moise, 2021; Pratt, 2022). Such so-called “Fractional CIOs” are often highly experienced and skilled individuals working for multiple SMEs simultaneously (Kratzer et al., 2022). By contracting a Fractional CIO, SMEs benefit in various ways. First, they get an experienced IT leader at a fraction of the cost. Second, they remain flexible concerning billable hours and the time horizon of the engagement. Third, they are receiving an outside-in perspective and might profit from the Fractional CIO’s experience gained from simultaneous engagements. Even though the Fractional CIO role benefits SMEs, adoption and awareness are still low. Further, substantial research on the novel role is missing. This dissertation aims to establish a new stream in CIO research that explores the role of the Fractional CIO. Our comprehensive literature review of the CIO research field provides the basis for this research (Kratzer et al., 2023b). As we explore this novel phenomenon, we are also interested in how to best report research findings. We, therefore, aim to conceptualize techniques to convey novel phenomena in information systems (IS) research in an illustrative way. Hence, we posed the following overarching research questions: RQ1: How can novel phenomena be communicated in an illustrative yet rigorous way in research papers? RQ2: What is the state-of-the-art in the CIO research field, and what are further avenues for research? RQ3: What concepts provide SMEs with the experienced IT executive management they need? RQ4: What makes Fractional CIO engagements successful? RQ5: How does the role of the Fractional CIO take shape in practice in the German market? These overarching research questions are answered through six consecutive publications. 1) 'Literary Sketches in Information Systems Research: Conceptualization and Guidance for Using Vignettes as a Narrative Form' This paper is conceptual and has a methodological focus on the use of vignettes. The paper contributes to the usage of vignettes as a narrative form in academic research, providing a taxonomy that structures vignettes and identifies three archetypes. By conceptualizing vignettes through a taxonomy and archetypes, we create transparency for the usage of vignettes, encourage the use of vignettes, and provide recommendations how to use them more rigorously. Long term, this might contribute to a better exchange between academia and practice through better comprehensibility of academic papers. 2) 'Four Decades of Chief Information Officer Research: A Literature Review and Research Agenda Based on Main Path Analysis' In this paper, we conducted a bibliographic literature review using main path analysis, which helped to objectively identify existing topics based on their importance. We also illustrated the knowledge flow in CIO research by identifying major and emerging research streams and analyzing their evolution over time. Additionally, we aggregated central papers in CIO research and developed a research agenda to provide guidance for future research. Overall, this paper helps to advance the understanding of the CIO research field and provides insights for researchers and practitioners. 3) 'The Fractional CIO in SMEs: Conceptualization and Research Agenda' With this paper, we were the first to conceptualize the role of the Fractional CIO and to develop a research agenda for this novel research field. The results from this paper contribute to research and practice in several ways. First, we are the first authors to conceptualize the role of the Fractional CIO, derive a definition, compare it to existing CIO role research, and derive four engagement types. Second, we develop a research agenda to guide future research in the new Fractional CIO research field. Third, we contribute to practice by promoting the role’s awareness among potential Fractional CIOs and organizations. Fourth, we show that Fractional CIO services are versatile and can benefit organizations of different sizes and maturity levels, and there are diverse possibilities for evolution. 4) 'What Makes Fractional CIO Engagements in SMEs Successful? – A Research Framework' In this paper, we developed a research framework for Fractional CIO engagement success. It makes contributions to research and practice. First, we propose a research framework for Fractional CIO engagement success and, therefore, develop a common ground for future research. Second, we provide practical advice for Fractional CIOs and SMEs regarding factors for engagement success. Third, we raise awareness about the Fractional CIO role that may benefit many SMEs around the world. 5) 'Factors for Fractional CIO Engagement Success' In this paper, we used Q-methodology to empirically evaluate the relative importance of factors influencing Fractional CIO engagement success and to identify different viewpoints on engagement success. Our paper makes several contributions to research and practice. First, our paper is the first to identify and evaluate the importance of factors for Fractional CIO engagement success. Therefore, it serves as a basis for further research in the new Fractional CIO research field. Second, we find that upper echelon theory (Hambrick & Mason, 1984) can be applied to executives independent of their contractual relationship in cases of comparable responsibilities. Third, by thoroughly applying Q-methodology, we showcase an exploratory tool for identifying opinions and preferences of participants and clustering them accordingly. This approach provides guidance for other IS scholars to apply it. Fourth, we develop preliminary archetypes of Fractional CIOs and, based on that, provide tentative advice for Fractional CIOs and organizations that hire them. These preliminary archetypes enabled us to identify further theoretical explanations of this phenomenon. In addition to transaction cost theory, agency theory, and upper echelon perspective, we found that stewardship theory and dynamic capabilities explain additional characteristics of Full-ownership CIOs and Change Agents. 6) 'Mehr als Vollzeit: Fractional CIOs in KMUs' This paper summarizes the results of our previous studies regarding the Fractional CIO role in German language and conducts semi-structured interviews with three Fractional CIOs from Germany to complement it with a perspective on the German market. Overall, we find that the Fractional CIO role in the German market is performed similarly to the international market. The interviewed German Fractional CIOs/CTOs could not clearly identify any hurdles that might explain the low adoption of the role. However, all three German Fractional CIOs agree that German SMEs would strongly benefit from Fractional CIOs. (References to be found in the full text) KW - Fractional CIO KW - CIO KW - Chief Information Officer KW - IT Management KW - KMUs KW - SMEs Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bsz:14-qucosa2-894186 ER - TY - THES A1 - Baumann, Martina T1 - Health Technology Assessment und ethische Aspekte der Prothesenversorgung N2 - Innovative Hand- und Beinprothesen haben das Potenzial, die Gesundheit und Lebensqualität von Menschen mit Gliedmaßenamputationen und Dysmelien zu verbessern. Aus einer Behindertenrechtsperspektive haben Menschen ein Recht auf eine Versorgung mit solchen Hilfsmitteln, die ihnen mehr gesellschaftliche Inklusion ermöglichen. Gleichzeitig können neue Prothesentechnologien Gesundheitssysteme aufgrund ihrer hohen Kosten vor Herausforderungen in Bezug auf die gerechte Ressourcenallokation stellen. Auch die zunehmende Verschmelzung von Prothesen mit menschlichen Körpern wirft ethische Fragen nach Autonomie und Sicherheit auf. Das Health Technology Assessment (HTA) ist ein wissenschaftlicher Ansatz zur Effizienzsteigerung und gerechten Rationierung in Gesundheitswesen. HTA soll Entscheidungen zur Erstattungsfähigkeit von Gesundheitstechnologien unterstützen, indem es deren Kosten und Nutzen systematisch untersucht – aber auch ethische, rechtliche und soziale Aspekte ihres Einsatzes beleuchtet. Diese Arbeit gibt einen Überblick über den Stand des HTA zu Prothesen auf Basis einer Recherche zu HTA-Berichten und Einschätzungen von Stakeholder*innen des deutschen Gesundheitswesens. Ethische Aspekte der Prothesentechnologie und -versorgung werden auf Basis eines Literaturreviews und einer Befragung von Prothesenträger*innen beleuchtet. Dieser Teil der Arbeit untersucht auch, inwiefern solch eine Kombination aus literaturbasiertem und empirischem Ansatz („empirische Ethik“) einen Mehrwert für Ethik im HTA bietet. In einem Fokus auf Gerechtigkeit wird untersucht, welche Hürden von Prothesennutzer*innen und Stakeholder*innen beim Zugang zu Prothesen im deutschen Gesundheitssystem wahrgenommen werden und inwiefern diese Hürden zu ungerechtem Zugang führen können. Es werden dann basierend auf den empirisch identifizierten Defiziten in der Versorgungspraxis und Norman Daniels Ansatz für Gesundheitsgerechtigkeit Empfehlungen für eine gerechtere und bessere Prothesenversorgung vorgestellt. Die Arbeit liefert somit Beiträge zur methodischen Diskussion um Ethik im Health Technology Assessment und zur bislang in Deutschland eher zögerlich geführten Diskussion, wie mit Ressourcenknappheit im Gesundheitswesen umgegangen werden sollte. N2 - Innovative hand and leg prostheses have the potential to improve the health and quality of life of people with limb amputations and dysmelia. From a disability rights perspective, people have a right to fair access to medical aids that enable them to be better included in society. At the same time, new prosthetic technologies may pose challenges for healthcare systems in terms of just resource allocation due to their high costs. The increasing fusion of prostheses with human bodies also raises ethical questions about autonomy and safety. Health Technology Assessment (HTA) is a scientific approach to increasing efficiency and fair rationing in the healthcare system. HTA supports decisions on the reimbursement of health technologies by systematically analysing their costs and benefits, as well as the ethical, legal, and social aspects of their implementation. This dissertation provides an overview of the status of HTA on prostheses based on a search of HTA reports and assessments by stakeholders in the German healthcare system. Ethical aspects of prosthesis technology and care are analysed based on a literature review and an interview study of prosthesis users. This part of the thesis also examines if such a combination of literature-based and empirical approaches (empirical ethics) offers added value for ethics in HTA. With a focus on ethics of justice, the barriers perceived by prosthesis users and stakeholders in access to prostheses in the German healthcare system, and the extent to which these barriers can lead to unfair access are analysed. Based on the empirically identified deficits in care practice and Norman Daniel's approach to just health, recommendations for fairer and better prosthesis care are presented. The work thus contributes to the methodological discussion on ethics in health technology assessment and to the rather hesitant debate in Germany on how to deal with resource scarcity in the healthcare system. Y1 - 2023 UR - https://publikationen.bibliothek.kit.edu/1000170547 U6 - https://doi.org/10.5445/IR/1000170547 ER - TY - THES A1 - Munninger, Maximilian T1 - Betrachtung der inhärenten Stochastizität durch das Scheduling in Verbindung mit der mehrstufigen Losbildung in einem hierarchischen Planungskonzept N2 - In der Dissertationsschrift „Betrachtung der inhärenten Stochastizität durch das Scheduling in Verbindung mit der mehrstufigen Losbildung in einem hierarchischen Planungskonzept“ werden mehrere Modellformulierungen für die Lösung von mehrstufigen kapazitierten Losgrößenproblemen erarbeitet. Bei der Losgrößenplanung mit Mikroperiodenmlodelen entstehen durch die üblicherweise integrierte Reihenfolgeplanung keine Verspätungen durch Wartezeiten. Jedoch sind die Rechenzeiten für optimale Lösungen in praxisrelevanten Problemgrößen nahezu unbeherrschbar. Bei einer Lösung mit Makroperiodenmodellen und einer nachfolgenden Reihenfolgeplanung mit Prioritätsregeln entstehen hingegen durch die unzureichenden Kapazitätsberechnungen immer wieder unvorhergesehene Wartezeiten. Deshalb werden in der Arbeit drei Modifikationen für das Makroperiodenmodell MLCLSP vorgeschlagen, mit denen diese Wartezeiten geschätzt werden können. Im ersten Ansatz werden Kapazitätskorrekturfaktoren eingeführt und deren Werte simulativ ermittelt, womit das Problem der reihenfolgebedingten Wartezeiten teilweise behoben werden kann. In einem zweiten Modellvorschlag werden diese Kapazitätskorrekturfaktoren mit einer Analogie aus der Warteschlangetheorie geschätzt. Ein drittes Modell wird vorgeschlagen, in dem ein Parameter eingeführt wird mit dem die Robustheit gegenüber der inhärenten Stochastizität zwischen den Losgrößen- und Scheduling-Problemen gesteuert werden kann. KW - Hierarchische Produktionsplanung KW - Mehrstufige kapazitierte Losgrößenplanung KW - Materialbedarfsplanung KW - Scheduling KW - Reihenfolgeplanung KW - Ressourceneinsatzplanung KW - Robuste Planung KW - Makroperiodenmodelle KW - Big-Bucket-Modelle KW - Sukzessivplanung KW - MLCLSP KW - Multi-level Capacitated Lotsizing Problem KW - Kapazitätskorrekturfaktoren KW - Warteschlangentheorie KW - Verspätungen KW - gemischt-ganzzahlige lineare Optimierung KW - Operations Research KW - Werkstattfertigung KW - Hierarchical production planning KW - Multi-level Capacitated Lotsizing Problem KW - Material requirements planning KW - Scheduling KW - Robust planning KW - Big-Bucket-Models KW - Consecutive planning KW - Capacity Correction Factors KW - Queueing Theory KW - Tardiness KW - mixed-integer linear programming KW - Job shop KW - stochasticity KW - Stochastizität KW - Kapazitätslotsenproblem KW - Zentrale Recheneinheit KW - capacity correction factors KW - Arbeitsgangfolge KW - work sequence KW - Lagerkosten KW - storage costs KW - Totalplanungsmodelle KW - total planning models KW - gleichgestellte Partialmodelle KW - equivalent partial models KW - Beschäftigungsglättung KW - employment smoothing KW - Direktbedarfskoeffizient KW - direct demand coefficient KW - Produktionslogistik KW - production logistics KW - Informationstechnik KW - information technology KW - simulation software KW - Simulationssoftware KW - delay values KW - Verzögerungswerte KW - Rüstzeit KW - set-up time KW - Mindestvorlaufzeit KW - minimum lead time Y1 - 2022 SN - 9783736975958 VL - 2022 PB - Cuvillier Verlag CY - Göttingen ET - 1. Auflage ER - TY - THES A1 - Wulf, Frederik T1 - Essays on cloud computing adoption, post-adoption, and security Y1 - 2021 ER - TY - THES A1 - Schönberger, Manuel T1 - Eine zweigleisige Strategie hin zu ausgereiftem Quanten-Datenmanagement T1 - A Twofold Strategy Towards Mature Quantum Data Management N2 - By relying on stagnant general-purpose devices limited by the end of Moore's law, conventional data management (DM) methods are challenged by ever-increasing computational loads. As a promising alternative, we may rely on quantum processing units (QPU) as special-purpose accelerators, which may overcome the limitations of conventional systems by exploiting quantum phenomena. However, given their early-stage, prototypical nature, contemporary QPUs remain limited in size and reliability, which renders them unfit for practical scenarios. In this thesis, we therefore propose a twofold strategy to aid the shift towards mature quantum data management. As a first fundamental pillar, our strategy rests on QPU-DM co-design: Inspired by the conventional hardware-software co-design paradigm, our method yields recommendations for the tailored design of QPU architecture, to craft quantum systems conforming to the specific requirements of DM tasks. Based on an empirical assessment involving contemporary quantum systems, and simulating enhanced QPU architectures, our method identifies the most effective among a variety of architectural enhancements, and allows us to derive recommendations on the design of QPUs tailored to DM tasks. However, to decide whether QPU-DM co-design efforts are truly worth pursuing, we require insights into the true potential of quantum systems for DM up-front. As a second fundamental pillar, our strategy hence assesses the algorithmic aptness of our methods on quantum-inspired special-purpose hardware (HW): While classical in nature, such systems mimic the workings of pure QPUs, and thus provide a lower bound on the capabilities of future QPUs. Thereby, our strategy yields insights into the true potential of future QPUs, which are expected to achieve still higher performance upon obtaining a sufficient level of maturity. We demonstrate the aptness of our twofold strategy on fundamental DM and query optimisation tasks by (1) deriving QPU design recommendations for the join ordering (JO) problem by applying our QPU-DM co-design method, (2) demonstrating the scalability aptness of quantum-inspired HW for extremely large JO solution spaces that prove challenging for conventional approaches, (3) deriving methods to solve the most extensive class of JO problems on quantum(-inspired) systems, and (4) conducting a still more extensive assessment of quantum-inspired HW for multiple query optimisation. Thereby, we aid the shift from currently imperfect QPUs towards systems fit for mature quantum data management. N2 - Anforderungen zur Verarbeitung stetig ansteigender Datenmengen stellen eine wachsende Herausforderung im Bereich des Datenmanagement (DM) dar, dessen etablierte Methoden auf traditionellen Allzweck-Systemen beruhen, deren Entwicklung durch das Ende des Mooreschen Gesetzes stagniert. Eine vielversprechende Alternative bieten Quantum Processing Units (QPU), welche das Potential aufweisen, die Limitationen traditioneller Systeme durch das Ausnutzen von Phänomenen der Quantenmechanik zu überwinden. Aufgrund ihres frühen Entwicklungsstadiums und ihrer prototypischen Natur sind gegenwärtige QPUs jedoch trotz dieses Potentials in ihrer Kapazität und Zuverlässigkeit begrenzt, und folglich ungeeignet für praktische Anwendungen. In dieser Arbeit schlagen wir daher eine zweigleisige Strategie für die Entwicklung hin zu ausgereiftem Quanten-Datenmanagement vor. Als erste fundamentale Säule beruht unsere Strategie auf Techniken des QPU-DM Co-Design: Inspiriert durch das konventionelle Hardware-Software Co-Design Paradigma können wir anhand unserer Methode Empfehlungen für das maßgeschneiderte Design von QPU Architekturen ermitteln, um Quanten-Systeme zu entwerfen, welche an spezifische Anforderungen fundamentaler DM Aufgaben angepasst sind. Basierend auf empirischen Analysen gegenwärtiger Quanten-Systeme und der Simulation strukturell erweiterter QPU Architekturen identifiziert unsere Methode unter Betrachtung einer Vielzahl möglicher Alternativen die effektivsten Erweiterungen für das Design von maßgeschneiderten QPUs. Zur Entscheidung über die Anwendung ressourcen-intensiver QPU-DM Co-Design Methoden benötigen wir vorab jedoch Erkenntnisse über das tatsächliche Potential von Quanten-Systemen für DM. Als zweite fundamentale Säule sieht unsere Strategie daher eine algorithmische Analyse von Quanten-Methoden mittels quanten-inspirierten Systemen vor: Derartige Systeme imitieren die Funktionsweise tatsächlicher QPUs, und geben daher trotz ihrer klassischen Natur Aufschluss über die Mindest-Leistungsfähigkeit und das Potential tatsächlicher, ausgereifter QPUs, welche zur Ausnutzung von Quantenphänomenen befähigt sind. Wir demonstrieren die Tauglichkeit unserer zweigleisigen Strategie für fundamentale DM Probleme durch (1) das Ermitteln von Design-Empfehlungen für das Join Ordering (JO) Problem durch Anwendung unserer QPU-DM Co-Design Methode, (2) das Demonstrieren der Skalierbarkeit quanten-inspirierter Methoden für extrem große JO Lösungsräume, (3) das Konzipieren von Methoden zum Lösen der umfangreichsten Klasse von JO Problemen auf quanten(-inspirierten) Systemen, und (4) das Durchführen einer noch umfangreicheren Analyse quanten-inspirierter Methoden für Multiple Query Optimisation. Dadurch unterstützen wir die Transformation von aktuellen QPU-Prototypen hin zu Quanten-Systemen, die für ausgereiftes Quanten-Datenmanagement geeignet sind. Y1 - 2025 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:898-opus4-84774 ER - TY - JOUR A1 - Hoegele, W. A1 - Löschel, Rainer A1 - Dobler, Barbara A1 - Hesser, J. A1 - Kölbl, Oliver A1 - Zygmanski, Piotr T1 - Stochastic formulation of patient positioning using linac-mounted cone beam imaging with prior knowledge JF - Medical physics N2 - PURPOSE In this work, a novel stochastic framework for patient positioning based on linac-mounted CB projections is introduced. Based on this formulation, the most probable shifts and rotations of the patient are estimated, incorporating interfractional deformations of patient anatomy and other uncertainties associated with patient setup. METHODS The target position is assumed to be defined by and is stochastically determined from positions of various features such as anatomical landmarks or markers in CB projections, i.e., radiographs acquired with a CB-CT system. The patient positioning problem of finding the target location from CB projections is posed as an inverse problem with prior knowledge and is solved using a Bayesian maximum a posteriori (MAP) approach. The prior knowledge is three-fold and includes the accuracy of an initial patient setup (such as in-room laser and skin marks), the plasticity of the body (relative shifts between target and features), and the feature detection error in CB projections (which may vary depending on specific detection algorithm and feature type). For this purpose, MAP estimators are derived and a procedure of using them in clinical practice is outlined. Furthermore, a rule of thumb is theoretically derived, relating basic parameters of the prior knowledge (initial setup accuracy, plasticity of the body, and number of features) and the parameters of CB data acquisition (number of projections and accuracy of feature detection) to the expected estimation accuracy. RESULTS MAP estimation can be applied to arbitrary features and detection algorithms. However, to experimentally demonstrate its applicability and to perform the validation of the algorithm, a water-equivalent, deformable phantom with features represented by six 1 mm chrome balls were utilized. These features were detected in the cone beam projections (XVI, Elekta Synergy) by a local threshold method for demonstration purposes only. The accuracy of estimation (strongly varying for different plasticity parameters of the body) agreed with the rule of thumb formula. Moreover, based on this rule of thumb formula, about 20 projections for 6 detectable features seem to be sufficient for a target estimation accuracy of 0.2 cm, even for relatively large feature detection errors with standard deviation of 0.5 cm and spatial displacements of the features with standard deviation of 0.5 cm. CONCLUSIONS The authors have introduced a general MAP-based patient setup algorithm accounting for different sources of uncertainties, which are utilized as the prior knowledge in a transparent way. This new framework can be further utilized for different clinical sites, as well as theoretical developments in the field of patient positioning for radiotherapy. Y1 - 2011 U6 - https://doi.org/10.1118/1.3532959 VL - 38 IS - 2 SP - 668 EP - 681 PB - American Association of Physicists in Medicine ER - TY - JOUR A1 - Heckner, Markus A1 - Schneidermeier, Tim A1 - Hastreiter, Isabella A1 - Wolff, Christian T1 - Der Newbie, der Fortgeschrittene und der Power-User. Warum es nicht reicht, “einfach mal die Nutzer zu fragen” JF - i-com N2 - Innerbetriebliche Software entwickelt sich häufig über Jahre hinweg inkrementell weiter, sammelt so Schritt für Schritt Funktionen an und verliert an Übersichtlichkeit. Soll diese Software hinsichtlich ihrer Gebrauchstauglichkeit optimiert werden, müssen aus Sicht eines externen Usability-Beraters zunächst die Domäne des Kunden verstanden und die Arbeitsprozesse analysiert werden. In der Analyse muss insbesondere der Umgang unterschiedlicher Nutzergruppen mit der Software berücksichtigt werden, um fundierte Neuvorschläge in einem Redesign erarbeiten zu können. In diesem Erfahrungsbericht beschreiben wir anhand eines konkreten Praxisbeispiels Eigenschaften von Nut-zergruppen und deren Einfluss auf die Analyse, und warum es nicht reicht, die Nutzer „einfach mal zu fragen, wie sie die Software denn finden". Y1 - 2012 U6 - https://doi.org/10.1524/icom.2012.0038 VL - 11 IS - 3 SP - 47 EP - 50 PB - DeGruyter ER - TY - JOUR A1 - Heinrich, Bernd A1 - Zellner, Gregor A1 - Leist, Susanne T1 - Service integrators in business networks - The importance of relationship values JF - Electronic Markets N2 - To realise the potentials of CRM (customer relationship management), relationship-specific processes need to be designed and implemented in companies. This is all the more important and complicated in business networks where two or more actors collaborate to serve the customers. A good collaboration within business networks is the basis for understanding the customer process and identifying customer needs. But, even in these days of customer orientation, transaction orientation is still a matter of strong interest, and the following questions remain to be answered: What is the difference between relationship-oriented processes and transaction-oriented processes, and how can relationship-oriented processes be designed for a business network? The authors give first answers to both questions by using a systematic, goal-oriented specialisation of generic actions. To give an example, one relationship-oriented process will be designed and specified for a certain customer process in the course of this paper. Y1 - 2011 U6 - https://doi.org/10.1007/s12525-011-0075-x VL - 21 IS - 4 SP - 215 EP - 235 PB - Springer ER - TY - CHAP A1 - Griesberger, Philipp A1 - Leist, Susanne A1 - Zellner, Gregor T1 - Analysis of techniques for business process improvement T2 - 19th European Conference on Information Systems, ECIS 2011, Helsinki, Finland, June 9-11, 2011 N2 - This paper is about identifying and analyzing techniques that can be used in a business process improvement (BPI) situation. To determine suitable techniques, the term BPI is defined and criteria are derived to analyze the usability of the selected techniques. Over 300 techniques from various improvement methods were reduced to those techniques which can be applied to help improve business processes and, in particular, to support the act of improvement. Identifying these techniques is helpful, because in most cases the act of improvement is a black box that lacks guidelines or structured procedures. In our research we developed an evaluation scheme to analyze the usability of BPI techniques and give hints about how to select a suitable technique for a certain improvement situation. Y1 - 2011 UR - https://www.researchgate.net/publication/221408836_Analysis_of_techniques_for_business_process_improvement ER - TY - JOUR A1 - Hoegele, W. A1 - Löschel, Rainer A1 - Merkle, N. A1 - Zygmanski, Piotr T1 - An efficient inverse radiotherapy planning method for VMAT using quadratic programming optimization JF - Medical physics N2 - PURPOSE The purpose of this study is to investigate the feasibility of an inverse planning optimization approach for the Volumetric Modulated Arc Therapy (VMAT) based on quadratic programming and the projection method. The performance of this method is evaluated against a reference commercial planning system (eclipse(TM) for rapidarc(TM)) for clinically relevant cases. METHODS The inverse problem is posed in terms of a linear combination of basis functions representing arclet dose contributions and their respective linear coefficients as degrees of freedom. MLC motion is decomposed into basic motion patterns in an intuitive manner leading to a system of equations with a relatively small number of equations and unknowns. These equations are solved using quadratic programming under certain limiting physical conditions for the solution, such as the avoidance of negative dose during optimization and Monitor Unit reduction. The modeling by the projection method assures a unique treatment plan with beneficial properties, such as the explicit relation between organ weightings and the final dose distribution. Clinical cases studied include prostate and spine treatments. The optimized plans are evaluated by comparing isodose lines, DVH profiles for target and normal organs, and Monitor Units to those obtained by the clinical treatment planning system eclipse(TM). RESULTS The resulting dose distributions for a prostate (with rectum and bladder as organs at risk), and for a spine case (with kidneys, liver, lung and heart as organs at risk) are presented. Overall, the results indicate that similar plan qualities for quadratic programming (QP) and rapidarc(TM) could be achieved at significantly more efficient computational and planning effort using QP. Additionally, results for the quasimodo phantom [Bohsung et al., "IMRT treatment planning: A comparative inter-system and inter-centre planning exercise of the estro quasimodo group," Radiother. Oncol. 76(3), 354-361 (2005)] are presented as an example for an extreme concave case. CONCLUSION Quadratic programming is an alternative approach for inverse planning which generates clinically satisfying plans in comparison to the clinical system and constitutes an efficient optimization process characterized by uniqueness and reproducibility of the solution. Y1 - 2012 U6 - https://doi.org/10.1118/1.3671922 VL - 39 IS - 1 SP - 444 EP - 454 PB - American Association of Physicists in Medicine ER - TY - JOUR A1 - Hoegele, W. A1 - Löschel, Rainer A1 - Zygmanski, Piotr T1 - An alternative VMAT with prior knowledge about the type of leaf motion utilizing projection method for concave targets JF - Medical physics N2 - The authors present an alternative approach to inverse planning optimization and apply it to volumetric modulated are therapy (VMAT) in one rotation with a prior knowledge about the type of leaf motions. The optimization is based on the projection theorem in inner product spaces. MLC motion is directly considered in the optimization, thus avoiding leaf segmentation characteristic of IMRT optimization. In this work they realize the method for concave irregular targets encompassing an organ at risk leading to a repetitive MLC motion pattern. Applying the projection theorem leads to a noniterative optimization method and reduces to solving few systems of linear equations with small numbers of dimensions. The solution of the inverse problem is unique, and false minima are naturally excluded. They divided the full rotation into about 50 short arc segments and for each segment decomposed dose into separate contributions related to stages of MLC motion. This results generally in an inverse problem with just four free parameters per arc segment. Practically three degrees of freedom will be used for the purpose of a constant angular speed of the gantry. Therefore the total number of degrees of freedom for a 3D problem is about 3 x 50 x number of collimator leaf pairs for irradiating the whole target volume in one rotation. Two 2D and one 3D concave target volumes are applied for a slice by slice optimization. A 6 MV photon beam model is used, including realistic scattering and attenuation, and a maximal leaf velocity of 3 cm/s is regarded. The resulting dose distributions cover the PTVs very well and have maxima at about 108% of dose in the PTVs. The OAR is spared very strong in all cases. As a result of optimization, the MLC apertures are repetitively opening and closing and can be interpreted in an intuitive way. Applying the projection method for this knowledge-based VMAT delivery scheme for concave target volumes is an alternative technique for dose optimization. There are several properties, such as uniqueness of MLC motions and their continuous dependence on geometry and prescribed dose, that make this approach interesting to inverse planning. This method is still in an investigational stage, but promising results are presented. In future work it will be extended directly (without conceptual changes) in several directions to be more clinically applicable. Y1 - 2009 U6 - https://doi.org/10.1118/1.3173815 VL - 36 IS - 8 SP - 3764 EP - 3774 PB - American Association of Physicists in Medicine ER - TY - JOUR A1 - Dobler, Barbara A1 - Walter, Cornelia A1 - Knopf, Antje A1 - Fabri, Daniella A1 - Löschel, Rainer A1 - Polednik, Martin A1 - Schneider, Frank A1 - Wenz, Frederik A1 - Lohr, Frank T1 - Optimization of extracranial stereotactic radiation therapy of small lung lesions using accurate dose calculation algorithms JF - Radiation Oncology N2 - BACKGROUND The aim of this study was to compare and to validate different dose calculation algorithms for the use in radiation therapy of small lung lesions and to optimize the treatment planning using accurate dose calculation algorithms. METHODS A 9-field conformal treatment plan was generated on an inhomogeneous phantom with lung mimics and a soft tissue equivalent insert, mimicking a lung tumor. The dose distribution was calculated with the Pencil Beam and Collapsed Cone algorithms implemented in Masterplan (Nucletron) and the Monte Carlo system XVMC and validated using Gafchromic EBT films. Differences in dose distribution were evaluated. The plans were then optimized by adding segments to the outer shell of the target in order to increase the dose near the interface to the lung. RESULTS The Pencil Beam algorithm overestimated the dose by up to 15% compared to the measurements. Collapsed Cone and Monte Carlo predicted the dose more accurately with a maximum difference of -8% and -3% respectively compared to the film. Plan optimization by adding small segments to the peripheral parts of the target, creating a 2-step fluence modulation, allowed to increase target coverage and homogeneity as compared to the uncorrected 9 field plan. CONCLUSION The use of forward 2-step fluence modulation in radiotherapy of small lung lesions allows the improvement of tumor coverage and dose homogeneity as compared to non-modulated treatment plans and may thus help to increase the local tumor control probability. While the Collapsed Cone algorithm is closer to measurements than the Pencil Beam algorithm, both algorithms are limited at tissue/lung interfaces, leaving Monte-Carlo the most accurate algorithm for dose prediction. Y1 - 2006 U6 - https://doi.org/10.1186/1748-717X-1-45 VL - 1 PB - BMC ; Springer ER - TY - JOUR A1 - Dobler, Barbara A1 - Streck, Natalia A1 - Klein, Elisabeth A1 - Löschel, Rainer A1 - Haertl, Petra Maria A1 - Kölbl, Oliver T1 - Hybrid plan verification for intensity-modulated radiation therapy (IMRT) using the 2D ionization chamber array I'mRT MatriXX--a feasibility study JF - Physics in medicine and biology N2 - The 2D ionization chamber array I'mRT MatriXX (IBA, Schwarzenbruck, Germany) has been developed for absolute 2D dosimetry and verification of intensity-modulated radiation therapy (IMRT) for perpendicular beam incidence. The aim of this study is to evaluate the applicability of I'mRT MatriXX for oblique beam incidence and hybrid plan verification of IMRT with original gantry angles. For the assessment of angular dependence, open fields with gantry angles in steps of 10 degrees were calculated on a CT scan of I'mRT MatriXX. For hybrid plan verification, 17 clinical IMRT plans and one rotational plan were used. Calculations were performed with pencil beam (PB), collapsed cone (CC) and Monte Carlo (MC) methods, which had been previously validated. Measurements were conducted on an Elekta SynergyS linear accelerator. To assess the potential and limitations of the system, gamma evaluation was performed with different dose tolerances and distances to agreement. Hybrid plan verification passed the gamma test with 4% dose tolerance and 3 mm distance to agreement in all cases, in 82-88% of the cases for tolerances of 3%/3 mm, and in 59-76% of the cases if 3%/2 mm were used. Separate evaluation of the low dose and high dose regions showed that I'mRT MatriXX can be used for hybrid plan verification of IMRT plans within 3% dose tolerance and 3 mm distance to agreement with a relaxed dose tolerance of 4% in the low dose region outside the multileaf collimator (MLC). Y1 - 2010 U6 - https://doi.org/10.1088/0031-9155/55/2/N02 VL - 55 IS - 2 SP - N39 EP - N55 PB - IOP Publishing ER - TY - JOUR A1 - Zygmanski, Piotr A1 - Högele, Wolfgang A1 - Cormack, Robert A1 - Chin, Lee A1 - Löschel, Rainer T1 - A volumetric-modulated arc therapy using sub-conformal dynamic arc with a monotonic dynamic multileaf collimator modulation JF - Physics in medicine and biology N2 - We present a new dose delivery scheme utilizing sub-conformal dynamic arc (sub-CD-ARC) with a dynamic multileaf collimator (dMLC) modulation in a single rotation. In sub-conformal delivery, a MLC aperture conforms not to the whole target but to its portion, simultaneously completely avoiding organs at risk (OARs). In CD-ARC therapy, the dose deposition level depends on the target width and distance to the axis of rotation, and therefore the use of multiple arcs is necessary to achieve a uniform dose within the target in 3D. In our delivery scheme, the dose deposition variations symptomatic of non-modulating sub-CD-ARC are compensated for in 3D by a quasi-periodic monotonic dMLC modulation. For this reason, we call this scheme sub-conformal dynamic modulated arc (sub-CD-MARC) therapy. The advantage of using such a modulation is that MLC leaf motion has just a few control points, and that an inverse planning problem is reduced to a linear equation problem with a few unknown parameters, which have clear physical meaning. We show the general dosimetric properties of sub-CD-ARC and sub-CD-MARC for a specific geometry of the target and OARs (rotational symmetry with varying inner and outer radii along the axis of rotation). In addition, we present numerical results of sub-CD-MARC inverse planning optimization. Y1 - 2008 U6 - https://doi.org/10.1088/0031-9155/53/22/009 VL - 53 IS - 22 SP - 6395 EP - 6417 PB - IOP Publishing ER - TY - BOOK A1 - Schultz, David A1 - Jain, Brijnesh T1 - Nonsmooth Analysis and Subgradient Methods for Averaging in Dynamic Time Warping Spaces N2 - Time series averaging in dynamic time warping (DTW) spaces has been successfully applied to improve pattern recognition systems. This article proposes and analyzes subgradient methods for the problem of finding a sample mean in DTW spaces. The class of subgradient methods generalizes existing sample mean algorithms such as DTW Barycenter Averaging (DBA). We show that DBA is a majorize-minimize algorithm that converges to necessary conditions of optimality after finitely many iterations. Empirical results show that for increasing sample sizes the proposed stochastic subgradient (SSG) algorithm is more stable and finds better solutions in shorter time than the DBA algorithm on average. Therefore, SSG is useful in online settings and for non-small sample sizes. The theoretical and empirical results open new paths for devising sample mean algorithms: nonsmooth optimization methods and modified variants of pairwise averaging methods. Y1 - 2017 U6 - https://doi.org/10.48550/arXiv.1701.06393 ER - TY - THES A1 - Souza Jr., Luis Antonio de T1 - Computer-assisted diagnosis of Barrett's esophagus using machine learning techniques T1 - Auxílio ao diagnóstico automático do esôfago de Barrett utilizando aprendizado de máquina N2 - Esophageal adenocarcinoma is an illness that is usually hard to detect at the early stages in the presence of Barrett’s esohagus. The development of automatic evaluation systems of such illness may be very useful, thus assisting the experts in the neoplastic region detection. With the strong growth of machine learning techniques aiming to improve the effectivess of medical diagnosis, the use of such approaches characterizes a strong scenario to be explored for the early diagnosis of esophageal adenocarcinoma. Barrett’s esophagus as a predecessor of adenocarcinoma can be explained by some risk factors, such as obesity, smoking, and late medical diagnosis. This project proposes the development of new computer vision and machine learning techniques to assist the automatic diagnosis of the esophageal adenocarcinioma based on the evaluation of two kind of features: (i) handcrafted features, calculated by means of human knowledge using some image processing technique and; (ii) deeply-learnable features, calculated exclusively based on deep learning techniques. From the extensive application of global and local protocols for the models proposed in this work, the description of cancer-affected images and Barrett’s esophagus-affected samples were generalized and deeply evaluated using, for example, classifiers such as Support Vector Machines, ResNet-50 and the combination of descriptions by handcrafted and deeply-learnable features. Also, the behavior of the automatic definition of key-points within the evaluated techniques was observed, something of a paramount importance nowadays to guarantee transparency and reliability in the decisions made by computational techniques. Thus, this project contributes to both the computational and medical fields, introducing new classifiers, approaches and interpretation of the class generalization process, in addition to proposing fast and precise manners to define cancer, delivering important and novel results concerning the accurate identification of cancer in samples affected by Barrett’s esophagus, showing values around 95% of correct identification rates and arranged in a collection of scientific works developed by the author during the research period and submitted/published to date. KW - Machine Learning KW - Barrett's esophagus KW - Deep Learning KW - handcrafted features KW - deeply-learnable features KW - convolutional neural networks KW - interpretability Y1 - 2022 PB - Universidade Federal de São Carlos ER - TY - CHAP A1 - Ramsauer, Ralf A1 - Lohmann, Daniel A1 - Mauerer, Wolfgang T1 - Reliable Pre-Integration Tracking of Commits on Mailing Lists T2 - Invited Talk at Safety Summit at Embedded Linux Conference Europe (ELCE18) Y1 - 2018 CY - Edinburgh, Scotland ER - TY - CHAP A1 - Schröpfer, Johannes A1 - Schwägerl, Felix A1 - Westfechtel, Bernhard T1 - Consistency control for model versions in evolving model-driven software product lines T2 - 2019 ACM/IEEE 22nd International Conference on Model Driven Engineering Languages and Systems Companion (MODELS-C), Munich, Germany N2 - Model-driven software product lines evolve in both time and space. Consistency control for model versions constitutes a key challenge. We propose a novel approach to consistency control called well-formedness analysis and repair: Instead of attempting to guarantee consistency of each configurable version a priori (which is hard and restrictive), consistency is controlled only when a product version is actually configured. Conflicts, i.e., violation of well-formedness constraints, are detected and repaired, driven by configurable strategies. This approach is generic; it is instantiated for feature models (for the variability model) and EMF models (as domain artifacts). Y1 - 2019 U6 - https://doi.org/10.1109/MODELS-C.2019.00043 SP - 268 EP - 277 PB - IEEE ER - TY - THES A1 - Schedlbauer, Jürgen T1 - Anforderungsorientierte Optimierung, Evaluation und Validierung der praktischen Inhalte der Lehre in der Medizinischen Informatik in Hochschulen für Angewandte Wissenschaften (HAW) N2 - Die European University Association (EUA) definiert „Beschäftigungsfähigkeit“ als ein Hauptziel der Hochschulausbildung. Der zentrale Aspekt dieser Arbeit ist, kompetenzorientierte Lehre in der praktischen Ausbildung der medizinischen Informatik an Hochschulen für angewandte Wissenschaft(HAW) durch Analyse der Kompetenzanforderungen des Marktes und der Hochschulen zu optimieren. Als Erstes wird dazu ein Berufsbild der Medizinischen Informatik erarbeitet, indem mit Hilfe von Webcrawling, Webextraktion und Text Mining Begriffe von Stellenanzeigen, anhand einer Inhaltsanalyse, die Kompetenzen im Bereich Fachwissen und Soft Skills (Geschäftsprozessmanagement, Teamwork, Prozesse, Lern- und Problemlösungsfähigkeiten) ermittelt werden. Als Nächstes wird untersucht, inwiefern HAWs diese Kompetenzen in ihren Modulhandbüchern als Lernziele definieren. Um die Vermittlung der praktischen Inhalte der Ausbildung entsprechend dem Berufsbild zu optimieren wurde - basierend auf der Methodik des Design-based Research - die Lehrmethodik „INstall“ entwickelt und evaluiert um die Arbeitsmarktfähigkeit der Studierenden zu verbessern. Die Eignung der Lehrmethodik wurde anhand einer Fremd- und Selbsteinschätzung der Studierenden untersucht. Die Ergebnisse der Stellenanalyse zeigen, dass nur 45% der Begriffe der Kategorie Fachwissen und 55% der Kategorie Soft Skills zuzuordnen sind. Die vorherrschenden Begriffe waren Programmierung, Erfahrung, Projekt und Server. Die Auswertung der Modulhandbücher zur Kompetenzorientierung hat große Unterschiede offengelegt, da in zwei Dritteln der Modulhandbücher keine Soft Skills als Lernergebnisse formuliert wurden. Anhand eines Zyklus über drei Vorlesungen wurde nachgewiesen, dass die Erweiterung der Lernziele auf Soft Skills durch die Methode „INstall“, in der Fremd- und Selbsteinschätzung der Studierenden einen starken Zugewinn an Kompetenzen im fachlichen Wissen, bei der Teamarbeit, im Geschäftsprozessmanagement, beim Lernen, in der Kommunikation und der Problemlösung gebracht hat. Die Arbeit legt den Schluss nahe, dass Hard- und Soft Skills bei den Arbeitnehmeranforderungen zur Medizinischen Informatik gleich gewichtet sind. Daraus kann abgeleitet werden, dass die Vorgaben der Berufsverbände zum Kompetenzprofil überdacht werden sollten. Fraglich erscheint, ob ein großer Teil der Modulhandbücher den Vorgaben der EUA (Beschäftigungsfähigkeit) entsprechen, da die Lernziele ein Defizit an Soft Skills aufweisen. Insgesamt bleibt festzuhalten: Eine gezielte Ausrichtung der Lernziele von Übungen auf die Anforderungen des Arbeitsmarktes durch die neue Lehrmethode „INstall“ hat zu einer von den Studierenden selbst bestätigten spürbaren Steigerung der benötigten Kompetenzen geführt. N2 - The European University Association (EUA) defines employability as a primary goal of higher education. The central aspect of this work is to optimize competence-oriented teaching in the practical training of medical informatics at universities of applied science (HAW) by analyzing and comparing the competence requirements of the market and universities. First, a professional profile of medical informatics was developed by using web crawling, web extraction and text mining to determine the competencies in the field of specialist knowledge and soft skills (business process management, teamwork, processes, learning and problem-solving skills) on the basis of a content analysis . Following this, several HAWs were examined in order to ascertain to what extent they define these competencies as learning objectives in their module handbooks. In order to optimize both the communication of the training‘s practical content according to the job description and student employability, a teaching method – „INstall“ - based on the methodology of design-based research - was developed and evaluated . The suitability of this teaching methodology was examined on the basis of external and self-assessment by students taking part in a lecture tailored to make use of the new learning goals. The results of the job analysis show that only 45% of the terms found in the professional profile can be assigned to the specialist knowledge category and 55% to the soft skills category. The predominant terms were programming, experience, project and server. Evaluating the module handbooks on competence orientation revealed major differences, as two thirds failed to formulate them as learning outcomes . On the basis of a cycle of three lectures, it was demonstrated that – in the external and self-assessment of partaking students - the expansion of learning objectives with soft skills via the „INstall“ method led to a strong gain in technical knowledge, teamwork, business process management, learning, communication and problem-solving competencies. Another important finding is that hard and soft skills are equally weighted in the employee requirements for medical informatics. This observation may support the proposal that the requirements of the professional associations for the competence profile should be reconsidered. When considering the learning objectives‘ lack of soft skills , it is questionable whether a large part of the module handbooks meet the requirements of the EEA (employability). The findings clearly indicate that a targeted alignment of the learning objectives of exercises with the requirements of the job market through the new teaching method „INstall “ has led to a noticeable increase in the required competencies, which the students themselves confirmed. KW - Web crawling KW - Text mining KW - Competence-based education KW - Soft skills KW - Medical Informatics Graduate employability Y1 - 2021 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:355-epub-477509 SN - 978-3-88246-461-0 ER - TY - CHAP A1 - Greiner, Sandra A1 - Schwägerl, Felix A1 - Westfechtel, Bernhard T1 - Realizing multi-variant model transformations on top of reused ATL specifications T2 - Proceedings of the 5th International Conference on Model-Driven Engineering and Software Development MODELSWARD, Porto, Portugal N2 - Model transformations are crucial in model-driven software engineering (MDSE). While combining MDSE and software product line engineering (SPLE) techniques, summarized as model-driven product line engineering (MDPLE), promises increased productivity by relying on organized reuse, the benefits are impeded by transformation specifications designed exclusively for single-variant models. Applying single-variant model transformations to multi-variant input models results in output models lacking the variability information. Multi-variant model transformations (MVMT), which preserve variability information, have only recently been understood as an explicit research problem. In this paper, we propose an a posteriori approach towards MVMT. Following the paradigm of organized reuse, we propose to employ single-variant model transformations without modifications in a first step, and to transfer variability information afterwards based on the artifacts provided by the single-variant transformation specification. In particular, we implemented this approach for the well-known model-to-model transformation language ATL. To deduce variability information, the execution artifacts (trace and execution model) are analyzed. Then, variability annotations are transfered to the target model automatically. The implementation is evaluated based on a practically example of a Graph product line. Results exhibit that our approach outperforms the conventional solution with respect to user effort, accuracy and performance. Y1 - 2017 U6 - https://doi.org/10.5220/0006137803620373 VL - 1 SP - 362 EP - 373 PB - SCITEPRESS - Science and Technology Publications ER - TY - THES A1 - Schmid, Isabel T1 - Weiterentwicklung von Methoden und Ansätzen zur automatisierten Informationsextraktion aus Social Networks N2 - Ob Werbebotschaften effektiv in einem sozialen Netzwerk verbreitet werden können, ist abhängig von der Person, die dabei als Multiplikator eingesetzt wird (Deng et al., 2016). Der Einsatz eines einflussreichen Nutzers kann nicht nur die Reichweite der Marketingbotschaft deutlich erhöhen, sondern auch in den Bereichen Innovations- und Produktmanagement wirken. Neben der großen praktischen Relevanz des Themengebiets hat dieses auch in der Wissenschaft eine Vielzahl an Publikationen hervorgebracht. Eine klare Definition und Abgrenzung einflussreicher Nutzertypen wie z.B. Influencer, Key User oder Lead User fehlt bislang und soll mit Hilfe eines ausführlichen Literaturüberblicks angefertigt werden (Veröffentlichung 1). Außerdem wird im Rahmen der Dissertation die Identifikation von Lead User weiterentwickelt, indem verschiedene Analyseformen wie z.B. die SNA, die Sentimentanalyse und das Topic Modeling miteinander kombiniert werden. Das Thema der Sentimentanalyse wird zudem in einer weiteren Veröffentlichung näher betrachtet, insbesondere im Zuge der Analyse langer und unstrukturierter Texte. Jedoch lassen sich Vorteile von sozialen Netzwerken nicht nur in Bezug auf die Verbindung zwischen Kunde und Unternehmen feststellen, sondern auch im Bereich ESN. ESN werden mit dem Ziel angewandt, die organisationale Effektivität sowie die Effizienz im Unternehmen zu verbessern (Stei et al., 2016). Zentral dabei ist, dass die Mitarbeiter dazu motiviert werden, aktiv Inhalte zu erzeugen und ihr Wissen mit anderen zu teilen (Günther and Spath, 2010). Hierbei lassen sich auch unter den Mitarbeitern unterschiedliche Nutzergruppen erkennen. So können u.a. aufgrund von Aktivität und Zentralität im Netzwerk Nutzertypen voneinander unterschieden werden. Zentrale Personen im Netzwerk müssen gefördert werden, um Kommunikation und Kollaboration zu sichern. Deshalb stehen die sog. „value-adding user“ im Zentrum der Analyse einer weiteren Publikation. N2 - The dissertation "Development of methods and approaches for automated information extraction from social networks" deals with the analysis of social media network data. Besides the distinction and characterization of different influential user types in social media networks, found in the research literature, the focus is on the (automated) identification of different influential user types applying various methods for analyzing the structured and unstructured social media data. In addition to the characterization and identification of influential users, this dissertation also concentrates on the analysis of textual social media network data. Thus, a comparison between different topic modeling techniques is performed in order to provide recommendations for several application scenarios. Furthermore, requirements for a trend analysis tool are collected. Based on these, software tools for the automated analysis of textual social media content (considering aspect-based sentiment analysis) were created and evaluated. Further research activities aim at deriving design approaches for the social media strategy regarding socially relevant topics (such as sustainability) by analyzing social media network data. KW - Social Media Network KW - Nutzertypen KW - Social Media Analyse Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:355-epub-534918 ER - TY - JOUR A1 - Bürkner, Leonhard A1 - Westner, Markus T1 - KI erfolgreich einführen: Status quo und Erfolgsfaktoren in deutschen Großunternehmen JF - Wirtschaftsinformatik & Management N2 - Viele deutsche Großunternehmen experimentieren derzeit intensiv mit künstlicher Intelligenz (KI), stehen aber vor der Frage, wie sich erste Pilotprojekte in einen nachhaltigen, wirtschaftlich wirksamen Einsatz überführen lassen. Eine empirische Studie mit 34 Chief Information Officers (CIO) und IT-Entscheidern in deutschen Großunternehmen zeigt: 112 identifizierte KI-Use-Cases, ein klar erkennbarer Reifezuwachs – aber auch deutliche Hürden bei Daten, Kompetenzen und Akzeptanz. Der Beitrag fasst den Status quo zusammen, validiert zentrale Erfolgsfaktoren aus der Forschung und leitet konkrete Empfehlungen für die Praxis von IT- und Fachbereichsverantwortlichen ab. KW - Künstliche Intelligenz KW - Generative KI KW - Wirtschaftsinformatik KW - Digitalisierung KW - Management KW - KI-Adoption KW - KI-Einführung KW - KI-Use-Cases KW - KI-Reifegrad KW - Großunternehmen Y1 - 2026 U6 - https://doi.org/10.1365/s35764-026-00593-6 N1 - Corresponding author der OTH Regensburg: Markus Westner PB - Springer ER - TY - JOUR A1 - Jain, Brijnesh A1 - Schultz, David T1 - Asymmetric learning vector quantization for efficient nearest neighbor classification in dynamic time warping spaces JF - Pattern Recognition N2 - The nearest neighbor method together with the dynamic time warping (DTW) distance is one of the most popular approaches in time series classification. This method suffers from high storage and computation requirements for large training sets. As a solution to both drawbacks, this article extends learning vector quantization (LVQ) from Euclidean spaces to DTW spaces. The proposed generic LVQ scheme uses asymmetric weighted averaging as update rule. We theoretically justify the asymmetric LVQ scheme via subgradient techniques and by the margin-growth principle. In addition, we show that the decision boundary of two prototypes from different classes is piecewise quadratic. Empirical results exhibited superior performance of asymmetric generalized LVQ (GLVQ) over other state-of-the-art prototype generation methods for nearest neighbor classification. Y1 - 2018 U6 - https://doi.org/10.1016/j.patcog.2017.10.029 VL - 76 SP - 349 EP - 366 PB - Elsevier ER - TY - JOUR A1 - Klingl, F. A1 - Waas, Thomas A1 - Weise, Wolfram T1 - Nuclear bound states of omega mesons JF - Nuclear Physics A N2 - Models based on chiral SU(3)_L \otimes SU(3)_R symmetry and vector meson dominance suggest an attractive potential for the $\omega$-meson in a nuclear medium. We discuss the feasibility of producing nuclear bound states of $\omega$-mesons using $(d,^3He)$ and pion induced reactions on selected nuclear targets. Y1 - 1999 U6 - https://doi.org/10.1016/S0375-9474(99)00114-1 VL - 650 IS - 3 SP - 299 EP - 312 PB - Elsevier ER - TY - GEN A1 - Stadler, Sebastian A1 - Cornet, Henriette A1 - Huang, Danqing A1 - Frenkler, Fritz ED - Jung, Timothy ED - tom Dieck, M. Claudia ED - Rauschnabel, Philipp A. T1 - Designing Tomorrow’s Human-Machine Interfaces in Autonomous Vehicles: an Exploratory Study in Virtual Reality T2 - Augmented Reality and Virtual Reality: Changing Realities in a Dynamic World N2 - Technical advances in the automotive industry strive in the direction of full automation. However, besides advantages like improving traffic and fuel efficiency, people do not always trust Autonomous Vehicles (AVs) to make critical decisions. With the ultimate goal of reducing anxiety of passengers of AVs, this explorative study (i) proposes possible design concepts and variants for Human-Machine Interfaces (HMI) for passengers inside the AV using a requirements catalogue, (ii) evaluates the HMI concepts and variants thanks to an experience simulation in Virtual Reality (VR), and (iii) derives the most suitable HMI concept and refines it based on observations of participants’ behaviours during the experience simulation in VR, as well as questionnaires and interviews. The results show that the HMI concepts help passengers to reduce anxiety in the AV. Overall, VR turned out to be a suitable tool for this exploratory study. Further work will focus on testing HMI concepts in a variety of more complex scenarios to ensure user acceptance. Y1 - 2020 SN - 978-3-030-37869-1 U6 - https://doi.org/10.1007/978-3-030-37869-1_13 SP - 151 EP - 160 PB - Springer ER - TY - INPR A1 - Waas, Thomas A1 - Brockmann, R. A1 - Weise, Wolfram T1 - Deeply bound pionic states and the effective pion mass in nuclear systems N2 - We show that the s-wave pion-nuclear potential which reproduces the deeply bound pionic states in Pb, recently discovered at GSI, is remarkably close to the one constructed directly from low energy theorems based on chiral symmetry. Converting this information into an effective pion mass we find $m_\pi^\star/m_\pi\simeq 1.13$ in the center of the Pb nucleus, and $m_\pi^\star/m_\pi\simeq 1.07$ in symmetric nuclear matter. Y1 - 1997 U6 - https://doi.org/10.48550/arXiv.hep-ph/9704397 N1 - Der Aufsatz wurde peer-reviewed veröffentlicht und ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/8893 ER - TY - JOUR A1 - Waas, Thomas A1 - Brockmann, R. A1 - Weise, Wolfram T1 - Deeply bound pionic states and the effective pion mass in nuclear systems JF - Physics Letters B N2 - We show that the s-wave pion-nuclear potential which reproduces the deeply bound pionic states in Pb, recently discovered at GSI, is remarkably close to the one constructed directly from low energy theorems based on chiral symmetry. Converting this information into an effective pion mass we find m:/m,, N 1.13 in the center of the Pb nucleus, and mi/m,, N 1.07 in symmetric nuclear matter. Y1 - 1997 U6 - https://doi.org/10.1016/S0370-2693(97)00643-6 N1 - Die Preprint-Version ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/7914 VL - 405 IS - 3;4 SP - 215 EP - 218 PB - Elsevier ER - TY - GEN ED - Handels, Heinz ED - Breininger, Katharina ED - Deserno, Thomas M. ED - Maier, Andreas ED - Maier-Hein, Klaus H. ED - Palm, Christoph ED - Tolxdorff, Thomas T1 - Bildverarbeitung für die Medizin 2026 : Proceedings, German Conference on Medical Image Computing, Lübeck March 15–17, 2026 N2 - Die Konferenz "BVM – Bildverarbeitung für die Medizin" ist seit vielen Jahren als die nationale Plattform für den Austausch von Ideen und die Diskussion der neuesten Forschungsergebnisse im Bereich der Medizinischen Bildverarbeitung und der Künstlichen Intelligenz (KI) etabliert. Auch 2026 haben (junge) Wissenschaftler*innen, Industrie und Anwender*innen diesen Austausch vertieft. Die Beiträge dieses Bandes – die meisten davon in englischer Sprache – umfassen alle Bereiche der medizinischen Bildverarbeitung, insbesondere die Bildgebung und -akquisition, Segmentierung und Analyse, Registrierung, Visualisierung und Animation, computerunterstützte Diagnose sowie bildgestützte Therapieplanung und Therapie. Hierbei kommen Methoden des maschinellen Lernens, der biomechanischen Modellierung sowie der Validierung und Qualitätssicherung zum Einsatz. KW - Bildverarbeitung KW - Computerunterstützte Medizin KW - Bildgebendes Verfahren KW - Bildanalyse KW - Deep Learning Y1 - 2026 SN - 978-3-658-51099-2 U6 - https://doi.org/10.1007/978-3-658-51100-5 SN - 1431-472X PB - Springer Vieweg CY - Wiesbaden ER - TY - INPR A1 - Eichenseher, Vincent A1 - Franz, Maja A1 - Wolff, Christian A1 - Mauerer, Wolfgang T1 - Pattern or Not? QAOA Parameter Heuristics and Potentials of Parsimony N2 - Structured variational quantum algorithms such as the Quantum Approximate Optimisation Algorithm (QAOA) have emerged as leading candidates for exploiting advantages of near-term quantum hardware. They interlace classical computation, in particular optimisation of variational parameters, with quantum-specific routines, and combine problem-specific advantages -- sometimes even provable -- with adaptability to the constraints of noisy, intermediate-scale quantum (NISQ) devices. While circuit depth can be parametrically increased and is known to improve performance in an ideal (noiseless) setting, on realistic hardware greater depth exacerbates noise: The overall quality of results depends critically on both, variational parameters and circuit depth. Although identifying optimal parameters is NP-hard, prior work has suggested that they may exhibit regular, predictable patterns for increasingly deep circuits and depending on the studied class of problems. In this work, we systematically investigate the role of classical parameters in QAOA performance through extensive numerical simulations and suggest a simple, yet effective heuristic scheme to find good parameters for low-depth circuits. Our results demonstrate that: (i) optimal parameters often deviate substantially from expected patterns; (ii) QAOA performance becomes progressively less sensitive to specific parameter choices as depth increases; and (iii) iterative component-wise fixing performs on par with, and at shallow depth may even outperform, several established parameter-selection strategies. We identify conditions under which structured parameter patterns emerge, and when deviations from the patterns warrant further consideration. These insights for low-depth circuits may inform more robust pathways to harnessing QAOA in realistic quantum compute scenarios. Y1 - 2026 U6 - https://doi.org/10.48550/arXiv.2510.08153 ER - TY - INPR A1 - Gierisch, Vincent A1 - Mauerer, Wolfgang T1 - QEF: Reproducible and Exploratory Quantum Software Experiments N2 - Commercially available Noisy Intermediate-Scale Quantum (NISQ) devices now make small hybrid quantum-classical experiments practical, but many tools hide configuration or demand ad-hoc scripting. We introduce the Quantum Experiment Framework (QEF): A lightweight framework designed to support the systematic, hypothesis-driven study of quantum algorithms. Unlike many existing approaches, QEF emphasises iterative, exploratory analysis of evolving experimental strategies rather than exhaustive empirical evaluation of fixed algorithms using predefined quality metrics. The framework's design is informed by a comprehensive review of the literature, identifying principal parameters and measurement practices currently reported in the field. QEF captures all key aspects of quantum software and algorithm experiments through a concise specification that expands into a Cartesian product of variants for controlled large-scale parameter sweeps. This design enables rigorous and systematic evaluation, as well as precise reproducibility. Large sweeps are automatically partitioned into asynchronous jobs across simulators or cloud hardware, and ascertain full hyper-parameter traceability. QEF supports parameter reuse to improve overall experiment runtimes, and collects all metrics and metadata into a form that can be conveniently explored with standard statistical and visualisation software. By combining reproducibility and scalability while avoiding the complexities of full workflow engines, QEF seeks to lower the practical barriers to empirical research on quantum algorithms, whether these are designed for current NISQ devices or future error-corrected quantum systems. Y1 - 2026 U6 - https://doi.org/10.48550/arXiv.2511.04563 ER - TY - CHAP A1 - Schönberger, Manuel A1 - Trummer, Immanuel A1 - Mauerer, Wolfgang T1 - Hybrid Mixed Integer Linear Programming for Large-Scale Join Order Optimisation T2 - Proceedings of the VLDB Endowment N2 - Finding optimal join orders is among the most crucial steps to be performed by query optimisers. Though extensively studied in data management research, the problem remains far from solved: While query optimisers rely on exhaustive search methods to determine ideal solutions for small problems, such methods reach their limits once queries grow in size. Yet, large queries become increasingly common in real-world scenarios, and require suitable methods to generate efficient execution plans. While a variety of heuristics have been proposed for large-scale query optimisation, they suffer from degrading solution quality as queries grow in size, or feature highly sub-optimal worst-case behavior, as we will show. We propose a novel method based on the paradigm of mixed integer linear programming (MILP): By deriving a novel MILP model capable of optimising arbitrary bushy tree structures, we address the limitations of existing MILP methods for join ordering, and can rely on highly optimised MILP solvers to derive efficient tree structures that elude competing methods. To ensure optimisation efficiency, we embed our MILP method into a hybrid framework, which applies MILP solvers precisely where they provide the greatest advantage over competitors, while relying on more efficient methods for less complex optimisation steps. Thereby, our approach gracefully scales to extremely large query sizes joining up to 100 relations, and consistently achieves the most robust plan quality among a large variety of competing join ordering methods. Y1 - 2026 U6 - https://doi.org/10.14778/3778092.3778097 VL - 19 IS - 3 SP - 348 EP - 360 PB - VLDB Endowment ER - TY - CHAP A1 - Gutbrod, Max A1 - Rauber, David A1 - Palm, Christoph ED - Handels, Heinz ED - Breininger, Katharina ED - Deserno, Thomas M. ED - Maier, Andreas ED - Maier-Hein, Klaus H. ED - Palm, Christoph ED - Tolxdorff, Thomas T1 - Improving Generalization in Mitotic Cell Detection via Domain Transformations T2 - Bildverarbeitung für die Medizin 2025: Proceedings, German Conference on Medical Image Computing, Lübeck March 15-17, 2026 N2 - We address domain generalization (DG) in mitotic-cell (MC) detection by combining a β-variational autoencoder (VAE) for domain transformations with feature-space alignment together with an object detector. The β-VAE synthesizes domain-transformed images, and the detector is trained to map originals and their transformed counterparts to equal representations. On the MIDOG++ dataset, this approach improves out-of-domain detection F1 scores by 7 and 3 percentage points compared to the color-variation augmentation and stain-normalization baselines. Results further suggest that morphology shifts hinder generalization more than stain shifts. KW - Künstliche Intelligenz KW - Bildverarbeitung Y1 - 2026 U6 - https://doi.org/10.1007/978-3-658-51100-5_71 SP - 362 EP - 367 PB - Springer Vieweg CY - Wiesbaden ER - TY - JOUR A1 - Schmidbauer, Lukas A1 - Lobe, Elisabeth A1 - Schaefer, Ina A1 - Mauerer, Wolfgang T1 - It’s Quick to be Square: Fast Quadratisation for Quantum Toolchains JF - ACM Transactions on Quantum Computing N2 - Many of the envisioned use-cases for quantum computers involve optimisation processes. While there are many algorithmic primitives to perform the required calculations, all eventually lead to quantum gates operating on quantum bits, with an order as determined by the structure of the objective function and the properties of target hardware. When the structure of the problem representation is not aligned with structure and boundary conditions of the executing hardware, various overheads degrading the computation may arise, possibly negating any possible quantum advantage. Therefore, automatic transformations of problem representations play an important role in quantum computing when descriptions (semi-)targeted at humans must be cast into forms that can be “executed” on quantum computers. Mathematically equivalent formulations are known to result in substantially different non-functional properties depending on hardware, algorithm and detail properties of the problem. Given the current state of noisy intermediate-scale quantum (NISQ) hardware, these effects are considerably more pronounced than in classical computing. Likewise, efficiency of the transformation itself is relevant because possible quantum advantage may easily be eradicated by the overhead of transforming between representations. In this paper, we consider a specific class of higher-level representations, that is, PUBOs, and devise novel automatic transformation mechanisms into widely used QUBOs that substantially improve efficiency and versatility over the state of the art. In addition, we conduct a comprehensive investigation of industry-relevant problem formulations and their conversion into a quantum-specific representation, identifying significant obstacles in scaling behaviour and demonstrating how these can be circumvented. KW - Algorithmic optimisation KW - Graphs KW - Performance KW - Pseudo Boolean Function Y1 - 2026 U6 - https://doi.org/10.1145/3800943 N1 - Preprint ebenfalls auf OPUS veröffentlicht: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/7937 VL - 7 IS - 2 PB - ACM ER - TY - INPR A1 - Ramsauer, Ralf A1 - Mauerer, Wolfgang T1 - Towards System-Level Quantum-Accelerator Integration N2 - Quantum computers are often treated as experimental add-ons that are loosely coupled to classical infrastructure through high-level interpreted languages and cloud-like orchestration. However, future deployments in both, high-performance computing (HPC) and embedded environments, will demand tighter integration for lower latencies, stronger determinism, and architectural consistency, as well as to implement error correction and other tasks that require tight quantum-classical interaction as generically as possible. We propose a vertically integrated quantum systems architecture that treats quantum accelerators and processing units as peripheral system components. A central element is the Quantum Abstraction Layer (QAL) at operating system kernel level. It aims at real-time, low-latency, and high-throughput interaction between quantum and classical resources, as well as robust low-level quantum operations scheduling and generic resource management. It can serve as blueprint for orchestration of low-level computational components "around" a QPU (and inside a quantum computer), and across different modalities. We present first results towards such an integrated architecture, including a virtual QPU model based on QEMU. The architecture is validated through functional emulation on three base architectures (x86_64, ARM64, and RISC-V), and timing-accurate FPGA-based simulations. This allows for a realistic evaluation of hybrid system performance and quantum advantage scenarios. Our work lays the ground for a system-level co-design methodology tailored for the next generation of quantum-classical computing. Y1 - 2025 U6 - https://doi.org/10.48550/arXiv.2507.19212 ER - TY - CHAP A1 - Ramsauer, Ralf A1 - Bielmeier, Benno A1 - Mauerer, Wolfgang T1 - Towards Real-World System-Level Integration of Quantum Accelerators: A Hardware/Software Co-Design Approach T2 - INFORMATIK; Lecture Notes in Informatics (LNI); GI Quantum Computing Workshop. Potsdam. 16.–19. September 2025 N2 - This work-in-progress explores architectural and systemic foundations for integrating quantum accelerators into heterogeneous computing environments. We propose and implement a modular architecture where Quantum Processing Units (QPUs) operate as peripheral devices, supporting pulse-level control interfaces and high-level circuit execution offloading, while internally managing compilation, transpilation, and scheduling. To enable efficient quantum-classical orchestration, we introduce a Quantum Abstraction Layer (QAL) at the operating system level to enable seamless communication, resource management, and integration with existing software frameworks. Our two-step design approach begins with validating the architecture through simulations in virtualised environments. We then implement an FPGA-based surrogate supporting both result- and timing-accurate modes, enabling full-stack emulation and performance evaluation in the absence of physical quantum hardware. This platform supports extensible and rapid prototyping, Hardware/Software Co-Design, and allows for investigations on the practical quantum advantage under realistic system-level constraints of various use cases. We aim for applicability by hardware vendors, facilitating early development even before physical quantum processors are available. Y1 - 2025 U6 - https://doi.org/10.18420/inf2025_158 SP - 1761 EP - 1765 PB - Gesellschaft für Informatik e.V. ER - TY - INPR A1 - Schmidbauer, Lukas A1 - Lobe, Elisabeth A1 - Schäfer, Ina A1 - Mauerer, Wolfgang T1 - It's Quick to be Square: Fast Quadratisation for Quantum Toolchains N2 - Many of the envisioned use-cases for quantum computers involve optimisation processes. While there are many algorithmic primitives to perform the required calculations, all eventually lead to quantum gates operating on quantum bits, with an order as determined by the structure of the objective function and the properties of target hardware. When the structure of the problem representation is not aligned with structure and boundary conditions of the executing hardware, various overheads degrading the computation may arise, possibly negating any possible quantum advantage. Therefore, automatic transformations of problem representations play an important role in quantum computing when descriptions (semi-)targeted at humans must be cast into forms that can be executed on quantum computers. Mathematically equivalent formulations are known to result in substantially different non-functional properties depending on hardware, algorithm and detail properties of the problem. Given the current state of noisy intermediate-scale quantum (NISQ) hardware, these effects are considerably more pronounced than in classical computing. Likewise, efficiency of the transformation itself is relevant because possible quantum advantage may easily be eradicated by the overhead of transforming between representations. In this paper, we consider a specific class of higher-level representations, i.e. polynomial unconstrained binary optimisation problems, and devise novel automatic transformation mechanisms into widely used quadratic unconstrained binary optimisation problems that substantially improve efficiency and versatility over the state of the art. We also identify what influence factors of lower-level details can be abstracted away in the transformation process, and which details must be made available to higher-level abstractions. Y1 - 2024 N1 - Der Aufsatz wurde peer-reviewed veröffentlicht und ist ebenfalls in diesem Repositorium verzeichnet unter: https://opus4.kobv.de/opus4-oth-regensburg/frontdoor/index/index/docId/9006 ER - TY - INPR A1 - Höß, Nicole A1 - Paradis, Carlos A1 - Kazman, Rick A1 - Mauerer, Wolfgang T1 - Oops!... I did it again. Conclusion (In-)Stability in Quantitative Empirical Software Engineering: A Large-Scale Analysis N2 - Context: Mining software repositories is a popular means to gain insights into a software project's evolution, monitor project health, support decisions and derive best practices. Tools supporting the mining process are commonly applied by researchers and practitioners, but their limitations and agreement are often not well understood. Objective: This study investigates some threats to validity in complex tool pipelines for evolutionary software analyses and evaluates the tools' agreement in terms of data, study outcomes and conclusions for the same research questions. Method: We conduct a lightweight literature review to select three studies on collaboration and coordination, software maintenance and software quality from high-ranked venues, which we formally replicate with four independent, systematically selected mining tools to quantitatively and qualitatively compare the extracted data, analysis results and conclusions. Results: We find that numerous technical details in tool design and implementation accumulate along the complex mining pipelines and can cause substantial differences in the extracted baseline data, its derivatives, subsequent results of statistical analyses and, under specific circumstances, conclusions. Conclusions: Users must carefully choose tools and evaluate their limitations to assess the scope of validity in an adequate way. Reusing tools is recommended. Researchers and tool authors can promote reusability and help reducing uncertainties by reproduction packages and comparative studies following our approach. Y1 - 2026 U6 - https://doi.org/10.48550/arXiv.2510.06844 ER - TY - CHAP A1 - Safi, Hila A1 - Niedermeier, Christoph A1 - Mauerer, Wolfgang T1 - TWiDDle: Twirling and Dynamical Decoupling, and Crosstalk Noise Modeling T2 - 2025 IEEE International Conference on Quantum Computing and Engineering (QCE), 30 August - 05 September 2025, Albuquerque N2 - Crosstalk remains a major source of correlated error in quantum systems, yet lacks a precise, community-wide definition - hindering systematic analysis and mitigation. This paper introduces a model-driven approach to crosstalk characterisation through three architecture-inspired noise models: (1) simultaneous two-qubit gate execution, (2) shared qubit interference, and (3) proximity-induced noise from shared control or readout hardware. These models act as both diagnostic tools and building blocks for crosstalk-aware quantum programming. We assess their impact across a broad benchmark suite - quantum simulation, Grover's algorithm, and fault-tolerant primitives like surface, Shor, and Steane codes - and evaluate two mitigation techniques: dynamical decoupling and Pauli twirling. While both are discussed in literature, only dynamical decoupling consistently enhances fidelity across noise types. Our work links low-level noise effects to high-level software engineering, underscoring the role of hardware-software co-design in scalable quantum computing. Model-based, hardware-aware design flows and composable noise abstractions improve error mitigation and program portability. Integrating such strategies into the toolchain is essential for building resilient quantum programs under realistic noise conditions. Y1 - 2025 U6 - https://doi.org/10.1109/QCE65121.2025.10313 SP - 162 EP - 168 PB - IEEE ER -