TY - JOUR A1 - Knobe, M. A1 - Giesen, M. A1 - Plate, S. A1 - Gradl-Dietsch, G. A1 - Buecking, B. A1 - Eschbach, D. A1 - Laack, Walter van A1 - Pape, H.-C. T1 - The Aachen mobility and balance index to measure physiological falls risk: a comparison with the Tinetti POMA scale JF - European Journal Of Trauma And Emergency Surgery N2 - Purpose The most commonly used mobility assessments for screening risk of falls among older adults are rating scales such as the Tinetti performance oriented mobility assessment (POMA). However, its correlation with falls is not always predictable and disadvantages of the scale include difficulty to assess many of the items on a 3-point scale and poor specificity. The purpose of this study was to describe the ability of the new Aachen Mobility and Balance Index (AMBI) to discriminate between subjects with a fall history and subjects without such events in comparison to the Tinetti POMA Scale. Methods For this prospective cohort study, 24 participants in the study group and 10 in the control group were selected from a population of patients in our hospital who had met the stringent inclusion criteria. Both groups completed the Tinetti POMA Scale (gait and balance component) and the AMBI (tandem stance, tandem walk, ten-meter-walk-test, sit-to-stand with five repetitions, 360° turns, timed-up-and-go-test and measurement of the dominant hand grip strength). A history of falls and hospitalization in the past year were evaluated retrospectively. The relationships among the mobility tests were examined with Bland–Altmananalysis. Receiver-operated characteristics curves, sensitivity and specificity were calculated. Results The study showed a strong negative correlation between the AMBI (17 points max., highest fall risk) and Tinetti POMA Scale (28 points max., lowest fall risk; r = −0.78, p < 0.001) with an excellent discrimination between community-dwelling older people and a younger control group. However, there were no differences in any of the mobility and balance measurements between participants with and without a fall history with equal characteristics in test comparison (AMBI vs. Tinetti POMA Scale: AUC 0.570 vs. 0.598; p = 0.762). The Tinetti POMA Scale (cut-off <20 points) showed a sensitivity of 0.45 and a specificity of 0.69, the AMBI a sensitivity of 0.64 and a specificity of 0.46 (cut-off >5 points). Conclusion The AMBI comprises mobility and balance tasks with increasing difficulty as well as a measurement of the dominant hand-grip strength. Its ability to identify fallers was comparable to the Tinetti POMA Scale. However, both measurement sets showed shortcomings in discrimination between fallers and non-fallers based on a self-reported retrospective falls-status. KW - Tinetti test KW - Mobility KW - Mobility tests KW - Ground-level falls KW - Elderly KW - Fall prevention KW - Co-managed care KW - Balance Y1 - 2016 U6 - https://doi.org/10.1007/s00068-016-0693-2 SN - 1863-9941 VL - 42 IS - 5 SP - 537 EP - 545 PB - Springer CY - Berlin ER - TY - CHAP A1 - Klöser, Lars A1 - Kohl, Philipp A1 - Kraft, Bodo A1 - Zündorf, Albert T1 - Multi-attribute relation extraction (MARE): simplifying the application of relation extraction T2 - Proceedings of the 2nd International Conference on Deep Learning Theory and Applications DeLTA - Volume 1 N2 - Natural language understanding’s relation extraction makes innovative and encouraging novel business concepts possible and facilitates new digitilized decision-making processes. Current approaches allow the extraction of relations with a fixed number of entities as attributes. Extracting relations with an arbitrary amount of attributes requires complex systems and costly relation-trigger annotations to assist these systems. We introduce multi-attribute relation extraction (MARE) as an assumption-less problem formulation with two approaches, facilitating an explicit mapping from business use cases to the data annotations. Avoiding elaborated annotation constraints simplifies the application of relation extraction approaches. The evaluation compares our models to current state-of-the-art event extraction and binary relation extraction methods. Our approaches show improvement compared to these on the extraction of general multi-attribute relations. Y1 - 2021 SN - 978-989-758-526-5 U6 - https://doi.org/10.5220/0010559201480156 N1 - 2nd International Conference on Deep Learning Theory and Applications, DeLTA2021, July 7-9, 2021 SP - 148 EP - 156 PB - SciTePress CY - Setúbal 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 - 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 - https://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 - TY - CHAP A1 - Kloock, Joachim P. A1 - Schöning, Michael Josef T1 - Heavy metal detection with semiconductor devices based on PLD-prepared chalcogenide glass thin films T2 - Armenian Journal of Physics Y1 - 2007 SN - 1829-1171 SP - 95 EP - 98 ER - TY - CHAP A1 - Kloock, Joachim P. A1 - Schubert, J. A1 - Ermelenko, Y. A1 - Vlasov, Y. G. A1 - Bratov, A. A1 - Schöning, Michael Josef T1 - Thin-film sensors with chalcogenide glass materials – a general survey T2 - Biochemical sensing utilisation of micro- and nanotechnologies : Warsaw, [23rd - 26th] November 2005 / ed. by M. Mascini ... Y1 - 2006 SP - 92 EP - 97 CY - Warsaw ER - TY - JOUR A1 - Kloock, Joachim P. A1 - Mourzina, Y.G. A1 - Ermolenko, Y. A1 - Doll, T. A1 - Schubert, J. A1 - Schöning, Michael Josef T1 - Inorganic thin-film sensor membranes with PLD-prepared chalcogenide glasses: Challenges and implementation JF - Sensors. 4 (2004), H. 10 Y1 - 2004 SN - 1424-8220 SP - 156 EP - 162 ER - TY - JOUR A1 - Kloock, Joachim P. A1 - Mourzina, Y. A1 - Schubert, J. A1 - Ermelenko, Y. A1 - Schöning, Michael Josef T1 - Pulsed laser deposition: A tool for fabricating thin-film microsensors JF - Biomedizinische Technik. 49 (2004), H. 2 Y1 - 2004 SN - 0932-4666 SP - 1032 EP - 1033 ER - TY - JOUR A1 - Kloock, Joachim P. A1 - Moreno, Lia A1 - Bratov, A. A1 - Huachupoma, S. A1 - Xu, J. A1 - Wagner, Torsten A1 - Yoshinobu, T. A1 - Ermolenko, Y. A1 - Vlasov, Y. G. A1 - Schöning, Michael Josef T1 - PLD-prepared cadmium sensors based on chalcogenide glasses —ISFET, LAPS and μISE semiconductor structures JF - Sensors and Actuators B: Chemical. 118 (2006), H. 1-2 Y1 - 2006 SN - 0925-4005 N1 - Eurosensors XIX - Eurosensors XIX - The 19th European Conference on Solid-State Transducers SP - 149 EP - 155 ER - TY - JOUR A1 - Kleefeld, Andreas A1 - Zimmermann, M. ED - Constanda, Christian ED - Bodmann, Bardo E.J. ED - Harris, Paul J. T1 - Computing Elastic Interior Transmission Eigenvalues JF - Integral Methods in Science and Engineering N2 - An alternative method is presented to numerically compute interior elastic transmission eigenvalues for various domains in two dimensions. This is achieved by discretizing the resulting system of boundary integral equations in combination with a nonlinear eigenvalue solver. Numerical results are given to show that this new approach can provide better results than the finite element method when dealing with general domains. Y1 - 2022 SN - 978-3-031-07171-3 U6 - https://doi.org/10.1007/978-3-031-07171-3_10 N1 - Corresponding author: Andreas Kleefeld SP - 139 EP - 155 PB - Birkhäuser CY - Cham ER - TY - JOUR A1 - Kleefeld, Andreas A1 - Reißel, Martin T1 - The Levenberg–Marquardt method applied to a parameter estimation problem arising from electrical resistivity tomography JF - Applied Mathematics and Computation Y1 - 2011 SN - 0096-3003 VL - 217 IS - 9 SP - 4490 EP - 4501 PB - Elsevier CY - Amsterdam ER -