TY - JOUR A1 - Schoenrock, Britt A1 - Muckelt, Paul E. A1 - Hastermann, Maria A1 - Albracht, Kirsten A1 - MacGregor, Robert A1 - Martin, David A1 - Gunga, Hans-Christian A1 - Salanova, Michele A1 - Stokes, Maria J. A1 - Warner, Martin B. A1 - Blottner, Dieter T1 - Muscle stiffness indicating mission crew health in space JF - Scientific Reports N2 - Muscle function is compromised by gravitational unloading in space affecting overall musculoskeletal health. Astronauts perform daily exercise programmes to mitigate these effects but knowing which muscles to target would optimise effectiveness. Accurate inflight assessment to inform exercise programmes is critical due to lack of technologies suitable for spaceflight. Changes in mechanical properties indicate muscle health status and can be measured rapidly and non-invasively using novel technology. A hand-held MyotonPRO device enabled monitoring of muscle health for the first time in spaceflight (> 180 days). Greater/maintained stiffness indicated countermeasures were effective. Tissue stiffness was preserved in the majority of muscles (neck, shoulder, back, thigh) but Tibialis Anterior (foot lever muscle) stiffness decreased inflight vs. preflight (p < 0.0001; mean difference 149 N/m) in all 12 crewmembers. The calf muscles showed opposing effects, Gastrocnemius increasing in stiffness Soleus decreasing. Selective stiffness decrements indicate lack of preservation despite daily inflight countermeasures. This calls for more targeted exercises for lower leg muscles with vital roles as ankle joint stabilizers and in gait. Muscle stiffness is a digital biomarker for risk monitoring during future planetary explorations (Moon, Mars), for healthcare management in challenging environments or clinical disorders in people on Earth, to enable effective tailored exercise programmes. KW - Ageing KW - Anatomy KW - Muscle KW - Musculoskeletal system KW - Physiology Y1 - 2024 U6 - http://dx.doi.org/10.1038/s41598-024-54759-6 SN - 2045-2322 N1 - Corresponding author: Dieter Blottner VL - 14 IS - Article number: 4196 PB - Springer Nature CY - London ER - TY - JOUR A1 - Bornheim, Tobias A1 - Grieger, Niklas A1 - Blaneck, Patrick Gustav A1 - Bialonski, Stephan T1 - Speaker Attribution in German Parliamentary Debates with QLoRA-adapted Large Language Models JF - Journal for language technology and computational linguistics : JLCL N2 - The growing body of political texts opens up new opportunities for rich insights into political dynamics and ideologies but also increases the workload for manual analysis. Automated speaker attribution, which detects who said what to whom in a speech event and is closely related to semantic role labeling, is an important processing step for computational text analysis. We study the potential of the large language model family Llama 2 to automate speaker attribution in German parliamentary debates from 2017-2021. We fine-tune Llama 2 with QLoRA, an efficient training strategy, and observe our approach to achieve competitive performance in the GermEval 2023 Shared Task On Speaker Attribution in German News Articles and Parliamentary Debates. Our results shed light on the capabilities of large language models in automating speaker attribution, revealing a promising avenue for computational analysis of political discourse and the development of semantic role labeling systems. KW - large language models KW - German KW - speaker attribution KW - semantic role labeling Y1 - 2024 U6 - http://dx.doi.org/10.21248/jlcl.37.2024.244 SN - 2190-6858 VL - 37 IS - 1 PB - Gesellschaft für Sprachtechnologie und Computerlinguistik CY - Regensburg ER - TY - BOOK A1 - Staat, Manfred A1 - Digel, Ilya A1 - Trzewik, Jürgen A1 - Sielemann, Stefanie A1 - Erni, Daniel A1 - Zylka, Waldemar T1 - Symposium Proceedings; 4th YRA MedTech Symposium 2024 : February 1 / 2024 / FH Aachen Y1 - 2024 SN - 978-3-940402-65-3 U6 - http://dx.doi.org/10.17185/duepublico/81475 PB - Universität Duisburg-Essen CY - Duisburg ER - TY - CHAP A1 - Schmitz, Annika A1 - Apandi, Shah Eiman Amzar Shah A1 - Spillner, Jan A1 - Hima, Flutura A1 - Behbahani, Mehdi ED - Digel, Ilya ED - Staat, Manfred ED - Trzewik, Jürgen ED - Sielemann, Stefanie ED - Erni, Daniel ED - Zylka, Waldemar T1 - Effect of different cannula positions in the pulmonary artery on blood flow and gas exchange using computational fluid dynamics analysis T2 - 4th YRA MedTech Symposium 2024 : February 1 / 2024 / FH Aachen N2 - Pulmonary arterial cannulation is a common and effective method for percutaneous mechanical circulatory support for concurrent right heart and respiratory failure [1]. However, limited data exists to what effect the positioning of the cannula has on the oxygen perfusion throughout the pulmonary artery (PA). This study aims to evaluate, using computational fluid dynamics (CFD), the effect of different cannula positions in the PA with respect to the oxygenation of the different branching vessels in order for an optimal cannula position to be determined. The four chosen different positions (see Fig. 1) of the cannulas are, in the lower part of the main pulmonary artery (MPA), in the MPA at the junction between the right pulmonary artery (RPA) and the left pulmonary artery (LPA), in the RPA at the first branch of the RPA and in the LPA at the first branch of the LPA. Y1 - 2024 SN - 978-3-940402-65-3 U6 - http://dx.doi.org/10.17185/duepublico/81475 SP - 29 EP - 30 PB - Universität Duisburg-Essen CY - Duisburg ER - TY - INPR A1 - Grieger, Niklas A1 - Mehrkanoon, Siamak A1 - Bialonski, Stephan T1 - Preprint: Data-efficient sleep staging with synthetic time series pretraining T2 - arXiv N2 - Analyzing electroencephalographic (EEG) time series can be challenging, especially with deep neural networks, due to the large variability among human subjects and often small datasets. To address these challenges, various strategies, such as self-supervised learning, have been suggested, but they typically rely on extensive empirical datasets. Inspired by recent advances in computer vision, we propose a pretraining task termed "frequency pretraining" to pretrain a neural network for sleep staging by predicting the frequency content of randomly generated synthetic time series. Our experiments demonstrate that our method surpasses fully supervised learning in scenarios with limited data and few subjects, and matches its performance in regimes with many subjects. Furthermore, our results underline the relevance of frequency information for sleep stage scoring, while also demonstrating that deep neural networks utilize information beyond frequencies to enhance sleep staging performance, which is consistent with previous research. We anticipate that our approach will be advantageous across a broad spectrum of applications where EEG data is limited or derived from a small number of subjects, including the domain of brain-computer interfaces. Y1 - 2024 ER - TY - JOUR A1 - Engelmann, Ulrich M. A1 - Simsek, Beril A1 - Shalaby, Ahmed A1 - Krause, Hans-Joachim T1 - Key contributors to signal generation in frequency mixing magnetic detection (FMMD): an in silico study JF - Sensors N2 - Frequency mixing magnetic detection (FMMD) is a sensitive and selective technique to detect magnetic nanoparticles (MNPs) serving as probes for binding biological targets. Its principle relies on the nonlinear magnetic relaxation dynamics of a particle ensemble interacting with a dual frequency external magnetic field. In order to increase its sensitivity, lower its limit of detection and overall improve its applicability in biosensing, matching combinations of external field parameters and internal particle properties are being sought to advance FMMD. In this study, we systematically probe the aforementioned interaction with coupled Néel–Brownian dynamic relaxation simulations to examine how key MNP properties as well as applied field parameters affect the frequency mixing signal generation. It is found that the core size of MNPs dominates their nonlinear magnetic response, with the strongest contributions from the largest particles. The drive field amplitude dominates the shape of the field-dependent response, whereas effective anisotropy and hydrodynamic size of the particles only weakly influence the signal generation in FMMD. For tailoring the MNP properties and parameters of the setup towards optimal FMMD signal generation, our findings suggest choosing large particles of core sizes dc > 25 nm nm with narrow size distributions (σ < 0.1) to minimize the required drive field amplitude. This allows potential improvements of FMMD as a stand-alone application, as well as advances in magnetic particle imaging, hyperthermia and magnetic immunoassays. KW - key performance indicators KW - magnetic biosensing KW - coupled Néel–Brownian relaxation dynamics KW - frequency mixing magnetic detection KW - magnetic relaxation KW - micromagnetic simulation KW - magnetic nanoparticles Y1 - 2024 U6 - http://dx.doi.org/10.3390/s24061945 SN - 1424-8220 N1 - This article belongs to the Special Issue "Advances in Magnetic Sensors and Their Applications" VL - 24 IS - 6 PB - MDPI CY - Basel ER - TY - CHAP A1 - Simsek, Beril A1 - Krause, Hans-Joachim A1 - Engelmann, Ulrich M. ED - Digel, Ilya ED - Staat, Manfred ED - Trzewik, Jürgen ED - Sielemann, Stefanie ED - Erni, Daniel ED - Zylka, Waldemar T1 - Magnetic biosensing with magnetic nanoparticles: Simulative approach to predict signal intensity in frequency mixing magnetic detection T2 - 4th YRA MedTech Symposium 2024 : February 1 / 2024 / FH Aachen N2 - Magnetic nanoparticles (MNP) are investigated with great interest for biomedical applications in diagnostics (e.g. imaging: magnetic particle imaging (MPI)), therapeutics (e.g. hyperthermia: magnetic fluid hyperthermia (MFH)) and multi-purpose biosensing (e.g. magnetic immunoassays (MIA)). What all of these applications have in common is that they are based on the unique magnetic relaxation mechanisms of MNP in an alternating magnetic field (AMF). While MFH and MPI are currently the most prominent examples of biomedical applications, here we present results on the relatively new biosensing application of frequency mixing magnetic detection (FMMD) from a simulation perspective. In general, we ask how the key parameters of MNP (core size and magnetic anisotropy) affect the FMMD signal: by varying the core size, we investigate the effect of the magnetic volume per MNP; and by changing the effective magnetic anisotropy, we study the MNPs’ flexibility to leave its preferred magnetization direction. From this, we predict the most effective combination of MNP core size and magnetic anisotropy for maximum signal generation. Y1 - 2024 SN - 978-3-940402-65-3 U6 - http://dx.doi.org/10.17185/duepublico/81475 SP - 27 EP - 28 PB - Universität Duisburg-Essen CY - Duisburg ER - TY - JOUR A1 - Zhen, Manghao A1 - Liang, Yunpei A1 - Staat, Manfred A1 - Li, Quanqui A1 - Li, Jianbo T1 - Discontinuous fracture behaviors and constitutive model of sandstone specimens containing non-parallel prefabricated fissures under uniaxial compression JF - Theoretical and Applied Fracture Mechanics N2 - The deformation and damage laws of non-homogeneous irregular structural planes in rocks are the basis for studying the stability of rock engineering. To investigate the damage characteristics of rock containing non-parallel fissures, uniaxial compression tests and numerical simulations were conducted on sandstone specimens containing three non-parallel fissures inclined at 0°, 45° and 90° in this study. The characteristics of crack initiation and crack evolution of fissures with different inclinations were analyzed. A constitutive model for the discontinuous fractures of fissured sandstone was proposed. The results show that the fracture behaviors of fissured sandstone specimens are discontinuous. The stress–strain curves are non-smooth and can be divided into nonlinear crack closure stage, linear elastic stage, plastic stage and brittle failure stage, of which the plastic stage contains discontinuous stress drops. During the uniaxial compression test, the middle or ends of 0° fissures were the first to crack compared to 45° and 90° fissures. The end with small distance between 0° and 45° fissures cracked first, and the end with large distance cracked later. After the final failure, 0° fissures in all specimens were fractured, while 45° and 90° fissures were not necessarily fractured. Numerical simulation results show that the concentration of compressive stress at the tips of 0°, 45° and 90° fissures, as well as the concentration of tensile stress on both sides, decreased with the increase of the inclination angle. A constitutive model for the discontinuous fractures of fissured sandstone specimens was derived by combining the logistic model and damage mechanic theory. This model can well describe the discontinuous drops of stress and agrees well with the whole processes of the stress–strain curves of the fissured sandstone specimens. KW - Constitutive model KW - Damage mechanics theory KW - Discontinuous fractures KW - Uniaxial compression test KW - Non-parallel fissures Y1 - 2024 U6 - http://dx.doi.org/10.1016/j.tafmec.2024.104373 SN - 0167-8442 VL - 131 IS - Art. No. 104373 PB - Elsevier CY - Amsterdam ER - TY - CHAP A1 - Kohl, Philipp A1 - Freyer, Nils A1 - Krämer, Yoka A1 - Werth, Henri A1 - Wolf, Steffen A1 - Kraft, Bodo A1 - Meinecke, Matthias A1 - Zündorf, Albert ED - Conte, Donatello ED - Fred, Ana ED - Gusikhin, Oleg ED - Sansone, Carlo T1 - ALE: a simulation-based active learning evaluation framework for the parameter-driven comparison of query strategies for NLP T2 - Deep Learning Theory and Applications. DeLTA 2023. Communications in Computer and Information Science N2 - Supervised machine learning and deep learning require a large amount of labeled data, which data scientists obtain in a manual, and time-consuming annotation process. To mitigate this challenge, Active Learning (AL) proposes promising data points to annotators they annotate next instead of a subsequent or random sample. This method is supposed to save annotation effort while maintaining model performance. However, practitioners face many AL strategies for different tasks and need an empirical basis to choose between them. Surveys categorize AL strategies into taxonomies without performance indications. Presentations of novel AL strategies compare the performance to a small subset of strategies. Our contribution addresses the empirical basis by introducing a reproducible active learning evaluation (ALE) framework for the comparative evaluation of AL strategies in NLP. The framework allows the implementation of AL strategies with low effort and a fair data-driven comparison through defining and tracking experiment parameters (e.g., initial dataset size, number of data points per query step, and the budget). ALE helps practitioners to make more informed decisions, and researchers can focus on developing new, effective AL strategies and deriving best practices for specific use cases. With best practices, practitioners can lower their annotation costs. We present a case study to illustrate how to use the framework. KW - Active learning KW - Query learning KW - Natural language processing KW - Deep learning KW - Reproducible research Y1 - 2023 SN - 978-3-031-39058-6 (Print) SN - 978-3-031-39059-3 (Online) U6 - http://dx.doi.org/978-3-031-39059-3 N1 - 4th International Conference, DeLTA 2023, Rome, Italy, July 13–14, 2023. SP - 235 EP - 253 PB - Springer CY - Cham ER - TY - CHAP A1 - Klöser, Lars A1 - Büsgen, André A1 - Kohl, Philipp A1 - Kraft, Bodo A1 - Zündorf, Albert ED - Conte, Donatello ED - Fred, Ana ED - Gusikhin, Oleg ED - Sansone, Carlo T1 - Explaining relation classification models with semantic extents T2 - DeLTA 2023: Deep Learning Theory and Applications N2 - In recent years, the development of large pretrained language models, such as BERT and GPT, significantly improved information extraction systems on various tasks, including relation classification. State-of-the-art systems are highly accurate on scientific benchmarks. A lack of explainability is currently a complicating factor in many real-world applications. Comprehensible systems are necessary to prevent biased, counterintuitive, or harmful decisions. We introduce semantic extents, a concept to analyze decision patterns for the relation classification task. Semantic extents are the most influential parts of texts concerning classification decisions. Our definition allows similar procedures to determine semantic extents for humans and models. We provide an annotation tool and a software framework to determine semantic extents for humans and models conveniently and reproducibly. Comparing both reveals that models tend to learn shortcut patterns from data. These patterns are hard to detect with current interpretability methods, such as input reductions. Our approach can help detect and eliminate spurious decision patterns during model development. Semantic extents can increase the reliability and security of natural language processing systems. Semantic extents are an essential step in enabling applications in critical areas like healthcare or finance. Moreover, our work opens new research directions for developing methods to explain deep learning models. KW - Relation classification KW - Natural language processing KW - Natural language understanding KW - Information extraction KW - Trustworthy artificial intelligence Y1 - 2023 SN - 978-3-031-39058-6 (Print) SN - 978-3-031-39059-3 (Online) U6 - http://dx.doi.org/10.1007/978-3-031-39059-3_13 N1 - 4th International Conference, DeLTA 2023, Rome, Italy, July 13–14, 2023. SP - 189 EP - 208 PB - Springer CY - Cham ER -