TY - INPR A1 - Bornheim, Tobias A1 - Niklas, Grieger A1 - Blaneck, Patrick Gustav A1 - Bialonski, Stephan T1 - Preprint: Speaker attribution in German parliamentary debates with QLoRA-adapted large language models T2 - Journal for Language Technology and Computational Linguistics 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. Y1 - 2023 U6 - http://dx.doi.org/10.48550/arXiv.2309.09902 N1 - Veröffentlichte Version verfügbar unter: https://doi.org/10.21248/jlcl.37.2024.244 ER - TY - INPR A1 - Ringers, Christa A1 - Bialonski, Stephan A1 - Solovev, Anton A1 - Hansen, Jan N. A1 - Ege, Mert A1 - Friedrich, Benjamin M. A1 - Jurisch-Yaksi, Nathalie T1 - Preprint: Local synchronization of cilia and tissue-scale cilia alignment are sufficient for global metachronal waves T2 - bioRxiv N2 - Motile cilia are hair-like cell extensions present in multiple organs of the body. How cilia coordinate their regular beat in multiciliated epithelia to move fluids remains insufficiently understood, particularly due to lack of rigorous quantification. We combine here experiments, novel analysis tools, and theory to address this knowledge gap. We investigate collective dynamics of cilia in the zebrafish nose, due to its conserved properties with other ciliated tissues and its superior accessibility for non-invasive imaging. We revealed that cilia are synchronized only locally and that the size of local synchronization domains increases with the viscosity of the surrounding medium. Despite the fact that synchronization is local only, we observed global patterns of traveling metachronal waves across the multiciliated epithelium. Intriguingly, these global wave direction patterns are conserved across individual fish, but different for left and right nose, unveiling a chiral asymmetry of metachronal coordination. To understand the implications of synchronization for fluid pumping, we used a computational model of a regular array of cilia. We found that local metachronal synchronization prevents steric collisions and improves fluid pumping in dense cilia carpets, but hardly affects the direction of fluid flow. In conclusion, we show that local synchronization together with tissue-scale cilia alignment are sufficient to generate metachronal wave patterns in multiciliated epithelia, which enhance their physiological function of fluid pumping. Y1 - 2021 U6 - http://dx.doi.org/10.1101/2021.11.23.469646 N1 - Veröffentlicht in eLife 12:e77701 (https://doi.org/10.7554/eLife.77701). 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 - Khodaverdi, M. A1 - Pauly, F. A1 - Schroder, G. A1 - Ziemons, Karl A1 - Sievering, R. A1 - Halling, H. T1 - Preliminary studies of a micro-CT for a combined small animal PET/CT scanner JF - 2001 IEEE Nuclear Science Symposium Conference Record, Vol. 3 N2 - We are developing an X-ray computed tomography (CT) system which will be combined with a high resolution animal PET system. This permits acquisition of both molecular and anatomical images in a single machine. In particular the CT will also be utilized for the quantification of the animal PET data by providing accurate data for attenuation correction. A first prototype has been built using a commercially available plane silicon diode detector. A cone-beam reconstruction provides the images using the Feldkamp algorithm. First measurements with this system have been performed on a mouse. It could be shown that the CT setup fulfils all demands for a high quality image of the skeleton of the mouse. It is also suited for soft tissue measurements. To improve contrast and resolution and to acquire the X-ray energy further development of the system, especially the use of semiconductor detectors and iterative reconstruction algorithms are planned. Y1 - 2002 SN - 1082-3654 SP - 1605 EP - 1606 ER - TY - CHAP A1 - Knott, Thomas C. A1 - Sofronia, Raluca E. A1 - Gerressen, Marcus A1 - Law, Yuen A1 - Davidescu, Arjana A1 - Savii, George G. A1 - Gatzweiler, Karl-Heinz A1 - Staat, Manfred A1 - Kuhlen, Torsten W. T1 - Preliminary bone sawing model for a virtual reality-based training simulator of bilateral sagittal split osteotomy T2 - Biomedical simulation : 6th International Symposium, ISBMS 2014, Strasbourg, France, October 16-17, 2014 : proceedings (Lecture notes in computer science : vol. 8789) N2 - Successful bone sawing requires a high level of skill and experience, which could be gained by the use of Virtual Reality-based simulators. A key aspect of these medical simulators is realistic force feedback. The aim of this paper is to model the bone sawing process in order to develop a valid training simulator for the bilateral sagittal split osteotomy, the most often applied corrective surgery in case of a malposition of the mandible. Bone samples from a human cadaveric mandible were tested using a designed experimental system. Image processing and statistical analysis were used for the selection of four models for the bone sawing process. The results revealed a polynomial dependency between the material removal rate and the applied force. Differences between the three segments of the osteotomy line and between the cortical and cancellous bone were highlighted. KW - Bone sawing KW - virtual reality KW - training simulator Y1 - 2014 SN - 978-3-319-12057-7 (Online) SN - 978-3-319-12056-0 (Print) U6 - http://dx.doi.org/10.1007/978-3-319-12057-7_1 SP - 1 EP - 10 PB - Springer CY - Cham ER - TY - CHAP A1 - Wagner, Torsten A1 - Schöning, Michael Josef T1 - Preface of the Special Issue of I3S 2005 in Jülich (Germany) N2 - International Symposium on Sensor Science, I3S 2005 <3; 2005; Juelich, Germany> In: Sensors 2006, 6, 260-261 ISSN 1424-8220 KW - Biosensor KW - I3S 2005 KW - International Symposium on Sensor Science Y1 - 2006 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:hbz:a96-opus-1365 ER - TY - JOUR A1 - Schöning, Michael Josef A1 - Doll, Theodor A1 - Wagner, Patrick T1 - Preface / Doll, Theodor ; Wagner, Patrick ; Schöning, Michael J. JF - physica status solidi (a) . 206 (2009), H. 3 Y1 - 2009 SN - 1862-6319 N1 - Special Issue: Engineering of Functional Interfaces (EnFI 08) SP - 389 EP - 390 ER - TY - JOUR A1 - Engelmann, Ulrich M. A1 - Shasha, Carolyn A1 - Teeman, Eric A1 - Slabu, Iona A1 - Krishnan, Kannan M. T1 - Predicting size-dependent heating efficiency of magnetic nanoparticles from experiment and stochastic Néel-Brown Langevin simulation JF - Journal of Magnetism and Magnetic Materials Y1 - 2019 U6 - http://dx.doi.org/10.1016/j.jmmm.2018.09.041 SN - 0304-8853 VL - 471 IS - 1 SP - 450 EP - 456 PB - Elsevier CY - Amsterdam ER - TY - JOUR A1 - Schelthoff, Christof A1 - Basermann, Achim A1 - Reichel, Björn T1 - Preconditioned CG methods for sparse matrices on massively parallel machines / Basermann, A. ; Reichel, B. ; Schelthoff, C. JF - Parallel Computing. 23 (1997), H. 3 Y1 - 1997 SN - 0167-8191 SP - 381 EP - 398 ER - TY - JOUR A1 - Yoshinobu, T. A1 - Ui, Y. A1 - Iwasaki, H. A1 - Näther, Niko A1 - Koudelka-Hep, M. A1 - Schöning, Michael Josef T1 - Potentiometric imaging in a microfluidic channel JF - Biomedizinische Technik. 49 (2004), H. 2 Y1 - 2004 SN - 0932-4666 SP - 998 EP - 999 ER -