https://libdoc.whz.de/opus4

Rule-based clinical decision support system for automated assessment of left ventricular diastolic function during stress echocardiography

Gulnora Rozikhodjaeva; Omonulla Juraev; H.-Christian Brauweiler; Tom Schaal — Mon, 20 Apr 2026 09:32:16 +0200

Heart failure with preserved ejection fraction (HFpEF) remains challenging to diagnose due to the complexity of diastolic function assessment during stress echocardiography, where multiple hemodynamic parameters must be evaluated under time pressure. Explainable artificial intelligence, specifically rule-based Clinical Decision Support Systems (CDSS), offers promising improvements in reproducibility and interpretability. Methods: A rule-based CDSS was developed and clinically validated to automate left ventricular diastolic function assessment during semi-supine bicycle stress echocardiography. A prospective cohort of 134 patients (mean age 61.3 ± 8.7 years) with exertional dyspnea and preserved left ventricular ejection fraction (LVEF >50%) was enrolled, excluding individuals with significant valvular pathologies, arrhythmias, or unstable ischemia. Echocardiographic and Doppler data were collected using Toshiba Aplio500 and Esaote MyLabSIGMA systems. The algorithm incorporated manual input of measurements, computed derived indices (e.g., diastolic reserve index, myocardial stiffness, vascular resistance), and applied rule-based logic in accordance with ASE/EACVI (2016/2022) guidelines and the ESC HFpEF consensus. Results: The CDSS generated diagnostic conclusions within 3 min per case, matching expert assessments in 93% of cases and correctly identifying stress-induced diastolic dysfunction in 85%. It demonstrated high diagnostic agreement (ICC > 0.94) and discrimination (AUC = 0.92). Rule-based outputs, such as “Impaired diastolic reserve” or “Right ventricular dysfunction under load,” were based on combinations of parameters (e.g., E/e′ > 15, Δe′ ≤ 0, TAPSE < 17 mm, PCWR > 12 mmHg). Conclusion: The explainable, guideline-compliant CDSS enables real-time, transparent analysis of diastolic function, supporting improved diagnostic consistency and augmented physician decision-making in cardiovascular care.

From Dialogue to Digital Memory: An Approach to Structuring Informal Organizational Knowledge with AI in Ambient Intelligence Contexts

Mon, 20 Apr 2026 07:49:37 +0200

An AI-Supported Approach Model for Personalizing Learning Processes

Mon, 20 Apr 2026 07:37:02 +0200

AI-supported qualitative analysis of free-text responses on home care burden and support needs in Saxony

Elisabeth Rau; Silke Geithner; Tom Schaal — Mon, 13 Apr 2026 09:19:14 +0200

The number of people in Germany requiring care has risen steadily, increasing the importance of informal care. This form of care is often associated with considerable psychological and physical strain. The aim of this study is to systematically categorize and qualitatively analyze the free-text responses from a survey on home care using an artificial intelligence-based approach in order to identify key challenges and support needs in home care from the perspective of informal caregivers and non-caregiving relatives. The study used data from a 2019 survey on home care in Saxony. Free-text responses were categorized and analyzed using GPT-4 Turbo within a hybrid human-AI workflow. All AI outputs were subsequently validated and corrected by researchers. Respondents reported substantial financial burdens for both care recipients and informal caregivers. They also highlighted structural barriers to accessing services and insufficient support from the care system. Improving home care requires structural measures, including the expansion of low-threshold counseling services, more flexible leave regulations, stronger financial security for informal caregivers, and the sustainable strengthening of care infrastructures. Given an AI error rate of 36.45%, the study emphasizes the need for human post-processing to ensure analytical accuracy.

Moderner Übersetzungsunterricht unter Nutzung digitaler Technologien

Anja Lange; Iryna Gaman — Tue, 07 Apr 2026 13:22:03 +0200

Fachdeutsch vermitteln mit KI: Didaktisierung von Lehrmaterialien und neue didaktische Konzepte durch den Einsatz von Künstlicher Intelligenz

Anja Lange; Guldastan Ismailova — Tue, 07 Apr 2026 13:12:44 +0200

Auf dem Weg zu einem kritisch-engagierten Ansatz und einem KI-integrierenden Curriculum im BA-Studiengang „Wirtschaftskommunikation Deutsch“

Janina M. Vernal Schmidt — Mon, 09 Mar 2026 12:54:31 +0100

KI-Berichterstattung zwischen Euphorie und Realität – KI-Newsletter als Informationsmedium

Ines-Andrea Busch-Lauer — Mon, 09 Mar 2026 12:53:53 +0100

Präsentationen im Zeitalter Künstlicher Intelligenz. Einsatz, Reflexion und Beurteilung im fremd- und fachsprachlichen Unterricht an Hochschulen

Sandra Braun — Mon, 09 Mar 2026 12:53:46 +0100

Wenn KI auf Fach und Sprache trifft

Mon, 09 Mar 2026 12:53:41 +0100

Was passiert, wenn Künstliche Intelligenz (KI) auf die Herausforderungen der Fachkommunikation trifft? Wie tragfähig sind die aktuellen Modelle und Konzepte für die Arbeit mit KI im fachlichen und fachsprachlichen Kontext? Expertinnen und Experten für Fachkommunikationsforschung aus neun Ländern stellen in diesem Band ihre aktuellen KI-Projekte in Forschung und Lehre vor. Im Mittelpunkt stehen dabei Modellbildung, KI-Kompetenz, Terminologie, Fachübersetzen und Dolmetschen sowie die Vermittlung von Fachsprache im Hochschulkontext. Mit Blick auf die Qualität fachkommunikativer Forschung und Lehre der Zukunft thematisieren sie Potenziale und Risiken der Nutzung von KI.

Tracking und Ausgabeverfahren

Mon, 23 Feb 2026 07:02:31 +0100

VR/AR-Geräte für die Ein- und Ausgabe

Paul Grimm; Wolfgang Broll; Rigo Herold; Dirk Reiners; Carolina Cruz-Neira — Mon, 09 Feb 2026 14:42:07 +0100

AI-driven risk estimation: a GPT-based approach to news monitoring for manufacturing resilience

Adrian Jacob; Anas Ben Achour; Uwe Teicher — Mon, 19 Jan 2026 07:31:29 +0100

In today’s rapidly evolving commercial landscape, manufacturing enterprises face significant challenges in maintaining resilience amid disruptions such as pandemics, natural disasters, and geopolitical conflicts. To address these challenges, we introduce a novel GPT-based early detection tool designed for real-time supply chain risk assessment. This system integrates proprietary company data, including supply chain portfolios, with publicly available information, such as news articles, to estimate risk scores for respective supply chains, thereby enhancing decision-making processes. Leveraging advanced machine learning techniques–Generative Pretrained Transformers (GPT), zero-shot learning, and structured outputs–the tool operates locally to ensure data privacy and minimize information leakage. Utilizing the "news-please" crawler and the "Llama 3.1" GPT model, the system continuously monitors selected media sources, providing timely risk assessments. Our research demonstrates the tool’s potential to enhance proactive risk management in supply chains, validated through testing on both real and augmented datasets. By evaluating four exemplary supply chains, we characterize the tool’s capability to support decision-making in unpredictable global environments. The results indicate that, while the system occasionally exhibits oversensitivity, it consistently aids in identifying critical events that may impact supply chain operations. Future developments will focus on refining the tool’s accuracy and expanding its applications, particularly in monitoring regulatory changes.

Rapid review on GenAI in nursing education

Laura Hinsche; Martina Hasseler; Tim Tischendorf; Tom Schaal — Mon, 19 Jan 2026 06:57:38 +0100

Background: The use of generative AI, as represented by ChatGPT, holds promising potential for nursing education. This manifests itself in various areas, including personalized learning, simulation training and teaching process support. However, its integration requires careful consideration of ethical implications, adaptation of curricula and a high level of digital competence on the part of teachers. Only in this way can potential risks, such as the distortion of knowledge, bias and educational inequalities, be avoided. Methods: Relevant publications were identified between 2019 and 2025 as part of a comprehensive literature search in the specialist databases PubMed, Embase, CINAHL and Scopus. The search was conducted using combined search terms that included the terms “generative AI”, “ChatGPT” and “nursing”. After removing duplicates and screening (PRISMA-guided), 140 full texts were analysed and divided into two publications. This rapid overview focuses on the topic of generative AI in nursing education. Results: As part of the analysis of the included studies, five thematic areas were identified, which were divided into the categories of nursing education, competence development and nursing skills, implementation possibilities, examination quality and ethical considerations, and evaluated. A key theme is the dual potential of this technology: it can enrich learning through features such as virtual tutors and improved exam preparation, but it also requires critical consideration of ethical issues such as plagiarism, data bias and the need for human oversight. Outlook: In this context, the conclusion emphasises the urgent need to adapt curricula and provide targeted further training for teachers so that GenAI can be used responsibly and effectively—rather than, as is often the case at present, by banning it altogether.

A Resource of German Light Verb Constructions Along with Possible Alternative Formulations

Ralf Laue — Thu, 15 Jan 2026 14:35:37 +0100

In this paper, the authors present a freely accessible resource covering German light verb constructions from the domain of administrative and business language (such as “Bericht erstatten” (to report)) together with their associated verb frames (in particular objects and governing prepositions). These constructions, which in many cases represent overly complex and difficult-to-read language, are joined within the same resource by simple verbs, also together with their respective verb frames. This not only provides a type of thesaurus for light verb constructions that can assist in text simplification, but the documented verb frames can also support the automatic generation of grammatically correct suggestions for simpler textual alternatives. (Abstract ist formal geringfügig verändert gegenüber dem Original.)

Integrating Class Relation Knowledge in Probabilistic Learning Vector Quantization

Marika Kaden; Ronny Schubert; Tina Geweniger; Wieland Hermann; Thomas Villmann — Thu, 15 Jan 2026 11:59:23 +0100

Buck, Isabella: Wissenschaftliches Schreiben mit KI

Ines Busch-Lauer — Wed, 07 Jan 2026 13:48:05 +0100

Rezension zu: Buck, Isabella: Wissenschaftliches Schreiben mit KI. Tübingen: Narr Francke Attempto Verlag, 2025, 242 Seiten. ISBN: 978-3-8385-6365-7.

AI in Forecasting: A Missing Component in ERP Systems?

Sebastian Junghans; Tim Neumann; Tobias Teich — Mon, 05 Jan 2026 14:05:49 +0100

KI und die Zukunft des Lehrberufs - Wo stehen wir? Was passiert mit uns?

Anja Lange — Mon, 05 Jan 2026 13:48:53 +0100

How to Rob an Empty Safe With ChatGPT

Ralf Laue — Thu, 09 Oct 2025 11:01:36 +0200

This article investigates, for the first time, how quickly curious individuals can uncover the secrets behind popular magic tricks through internet research requiring minimal effort. To do this, 20 well-known magic tricks were selected, and the underlying trick secrets were sought using the AI assistant ChatGPT. It was found that using language assistants and the large language models behind them makes the search significantly more promising compared to recent keyword-based searches. For 18 of the 20 magic tricks, an explanation in the form of a descriptive text or an exposure video was found with little effort. The significance of this observation for the art of magic was not the main focus of this contribution, but the concluding section at least contains the author's initial thoughts on the matter.

Fast-Track Your Decisions: Leveraging Low Code to Speed Up Simulation-Driven Insights in Semiconductor Industry

Madlene Leißau; Adrian Rössl; Christoph Laroque — Fri, 27 Jun 2025 09:12:14 +0200

Low-code approaches can accelerate decision-making in the semiconductor industry by streamlining simulation-driven insights. This supports the paradigm shift to Industry 4.0 and Industry 5.0 by enabling rapid development and optimized workflows. However, existing simulation methods often require extensive coding expertise, limiting accessibility and slowing down model development. This paper presents a simulation template that streamlines the development of discrete event simulation models in semiconductor manufacturing. Thus, the simulation template implements reusable components to simplify model creation and reduce development time. The approach encourages collaboration between technical and nontechnical stakeholders. Combined with a low-code data farming framework, the simulation template increases agility, accelerates experimentation, and supports efficient, data-driven production planning decisions.

Neural network-based localization of spherical MRI markers

Christian Fiedler; Silke Kolbig — Thu, 19 Jun 2025 07:59:16 +0200

This work focuses on the detection and localization of small spherical fiducial markers in magnetic resonance imaging (MRI) using neural networks. Two image processing pipelines based on U-Net and YOLO architectures were developed and evaluated on a data set of T1- and T2-weighted MRI with voxel sizes ranging from 0.6 to 1.6 mm. Detection performance is evaluated using the F1-score, whereas localization is evaluated using two metrics that describe the deviation of the predicted position from the true position. Although the benchmark method, a conventional image processing pipeline based on connected component analysis achieved marginally lower positioning errors, the neural network approaches outperformed it in terms of detection performance, especially by reducing false negatives. The results show that both pipelines achieve accurate marker detection and localization, with U-Net slightly outperforming YOLO in terms of positioning accuracy. A key advantage of the neural network-based pipelines is their ability to handle markers with non-uniform or incomplete appearance, which enhances their robustness in real-world scenarios and provides flexibility by eliminating the need for manual parameter adjustments. While neural networks offer the advantage that they can be easily adapted to various imaging conditions, their dependence on training data can be a limitation. The results suggest that neural network based pipelines offer a robust alternative for fiducial marker detection and localization.

Data-driven eigenmode estimation of optical fibers in TMI-regime by exploitation of physically constrained “glass box” machine learning model

Alexander Kabardiadi-Virkovski; Leander Kläber; Peter Hartmann — Fri, 13 Jun 2025 14:22:08 +0200

Asking ChatGPT for Pattern Recommendations: EuroPLoP 2024 Focus Group Report

Ralf Laue; João José Maranhão; Eduardo Martins Guerra — Fri, 13 Jun 2025 13:43:27 +0200

GenAI in nursing and clinical practice: a rapid review of applications and challenges

Tim Tischendorf; Laura Hinsche; Martina Hasseler; Tom Schaal — Thu, 12 Jun 2025 14:13:35 +0200

Aim: The use of artificial intelligence in nursing has become increasingly important in recent years. In particular, generative artificial intelligence (GenAI) such as ChatGPT offers the potential to improve care processes, support decision-making, and reduce workload. The aim of this paper is to provide an overview of the current state of research on the use of GenAI in nursing and clinical practice. Subject and methods: A systematic literature search was conducted in the PubMed, Embase, CINAHL, and Scopus databases. Studies from the last 5 years (2019–2024) dealing with the use of GenAI in professional nursing and the improvement of nursing skills through AI were included. Studies on machine learning, deep learning, and specific disease contexts were excluded. A total of 13 studies were included in the analysis. Results: GenAI in nursing and clinical practice can increase the efficiency of tasks such as scheduling and care planning, but there are currently significant gaps in decision accuracy and reliability. Studies show potential to reduce workload, but also point to the need for further research and technical improvements. Conclusion: Although GenAI in nursing is promising, there are still significant limitations. Future developments and regulatory measures are needed to ensure the safe and effective use of GenAI in nursing practice.