Refine
Year of publication
Institute
- Fachbereich Medizintechnik und Technomathematik (1692) (remove)
Language
- English (1692) (remove)
Document Type
- Article (1359)
- Conference Proceeding (212)
- Book (43)
- Part of a Book (43)
- Doctoral Thesis (18)
- Other (6)
- Patent (4)
- Preprint (3)
- Lecture (2)
- Habilitation (1)
Keywords
- Biosensor (25)
- Finite-Elemente-Methode (12)
- Einspielen <Werkstoff> (10)
- CAD (8)
- civil engineering (8)
- Bauingenieurwesen (7)
- FEM (6)
- Limit analysis (6)
- Shakedown analysis (6)
- shakedown analysis (6)
Selected problems in the field of multivariate statistical analysis are treated. Thereby, one focus is on the paired sample case. Among other things, statistical testing problems of marginal homogeneity are under consideration. In detail, properties of Hotelling‘s T² test in a special parametric situation are obtained. Moreover, the nonparametric problem of marginal homogeneity is discussed on the basis of possibly incomplete data. In the bivariate data case, properties of the Hoeffding-Blum-Kiefer-Rosenblatt independence test statistic on the basis of partly not identically distributed data are investigated. Similar testing problems are treated within the scope of the application of a result for the empirical process of the concomitants for partly categorial data. Furthermore, testing changes in the modeled solvency capital requirement of an insurance company by means of a paired sample from an internal risk model is discussed. Beyond the paired sample case, a new asymptotic relative efficiency concept based on the expected volumes of multidimensional confidence regions is introduced. Besides, a new approach for the treatment of the multi-sample goodness-of-fit problem is presented. Finally, a consistent test for the treatment of the goodness-of-fit problem is developed for the background of huge or infinite dimensional data.
Messenger apps like WhatsApp and Telegram are frequently used for everyday communication, but they can also be utilized as a platform for illegal activity. Telegram allows public groups with up to 200.000 participants. Criminals use these public groups for trading illegal commodities and services, which becomes a concern for law enforcement agencies, who manually monitor suspicious activity in these chat rooms. This research demonstrates how natural language processing (NLP) can assist in analyzing these chat rooms, providing an explorative overview of the domain and facilitating purposeful analyses of user behavior. We provide a publicly available corpus of annotated text messages with entities and relations from four self-proclaimed black market chat rooms. Our pipeline approach aggregates the extracted product attributes from user messages to profiles and uses these with their sold products as features for clustering. The extracted structured information is the foundation for further data exploration, such as identifying the top vendors or fine-granular price analyses. Our evaluation shows that pretrained word vectors perform better for unsupervised clustering than state-of-the-art transformer models, while the latter is still superior for sequence labeling.
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.
This study evaluates neuromechanical control and muscle-tendon interaction during energy storage and dissipation tasks in hypergravity. During parabolic flights, while 17 subjects performed drop jumps (DJs) and drop landings (DLs), electromyography (EMG) of the lower limb muscles was combined with in vivo fascicle dynamics of the gastrocnemius medialis, two-dimensional (2D) kinematics, and kinetics to measure and analyze changes in energy management. Comparisons were made between movement modalities executed in hypergravity (1.8 G) and gravity on ground (1 G). In 1.8 G, ankle dorsiflexion, knee joint flexion, and vertical center of mass (COM) displacement are lower in DJs than in DLs; within each movement modality, joint flexion amplitudes and COM displacement demonstrate higher values in 1.8 G than in 1 G. Concomitantly, negative peak ankle joint power, vertical ground reaction forces, and leg stiffness are similar between both movement modalities (1.8 G). In DJs, EMG activity in 1.8 G is lower during the COM deceleration phase than in 1 G, thus impairing quasi-isometric fascicle behavior. In DLs, EMG activity before and during the COM deceleration phase is higher, and fascicles are stretched less in 1.8 G than in 1 G. Compared with the situation in 1 G, highly task-specific neuromuscular activity is diminished in 1.8 G, resulting in fascicle lengthening in both movement modalities. Specifically, in DJs, a high magnitude of neuromuscular activity is impaired, resulting in altered energy storage. In contrast, in DLs, linear stiffening of the system due to higher neuromuscular activity combined with lower fascicle stretch enhances the buffering function of the tendon, and thus the capacity to safely dissipate energy.
It has been shown that muscle fascicle curvature increases with increasing contraction level and decreasing muscle–tendon complex length. The analyses were done with limited examination windows concerning contraction level, muscle–tendon complex length, and/or intramuscular position of ultrasound imaging. With this study we aimed to investigate the correlation between fascicle arching and contraction, muscle–tendon complex length and their associated architectural parameters in gastrocnemius muscles to develop hypotheses concerning the fundamental mechanism of fascicle curving. Twelve participants were tested in five different positions (90°/105°*, 90°/90°*, 135°/90°*, 170°/90°*, and 170°/75°*; *knee/ankle angle). They performed isometric contractions at four different contraction levels (5%, 25%, 50%, and 75% of maximum voluntary contraction) in each position. Panoramic ultrasound images of gastrocnemius muscles were collected at rest and during constant contraction. Aponeuroses and fascicles were tracked in all ultrasound images and the parameters fascicle curvature, muscle–tendon complex strain, contraction level, pennation angle, fascicle length, fascicle strain, intramuscular position, sex and age group were analyzed by linear mixed effect models. Mean fascicle curvature of the medial gastrocnemius increased with contraction level (+5 m−1 from 0% to 100%; p = 0.006). Muscle–tendon complex length had no significant impact on mean fascicle curvature. Mean pennation angle (2.2 m−1 per 10°; p < 0.001), inverse mean fascicle length (20 m−1 per cm−1; p = 0.003), and mean fascicle strain (−0.07 m−1 per +10%; p = 0.004) correlated with mean fascicle curvature. Evidence has also been found for intermuscular, intramuscular, and sex-specific intramuscular differences of fascicle curving. Pennation angle and the inverse fascicle length show the highest predictive capacities for fascicle curving. Due to the strong correlations between pennation angle and fascicle curvature and the intramuscular pattern of curving we suggest for future studies to examine correlations between fascicle curvature and intramuscular fluid pressure.
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.
Background
Hip fractures are a common and costly health problem, resulting in significant morbidity and mortality, as well as high costs for healthcare systems, especially for the elderly. Implementing surgical preventive strategies has the potential to improve the quality of life and reduce the burden on healthcare resources, particularly in the long term. However, there are currently limited guidelines for standardizing hip fracture prophylaxis practices.
Methods
This study used a cost-effectiveness analysis with a finite-state Markov model and cohort simulation to evaluate the primary and secondary surgical prevention of hip fractures in the elderly. Patients aged 60 to 90 years were simulated in two different models (A and B) to assess prevention at different levels. Model A assumed prophylaxis was performed during the fracture operation on the contralateral side, while Model B included individuals with high fracture risk factors. Costs were obtained from the Centers for Medicare & Medicaid Services, and transition probabilities and health state utilities were derived from available literature. The baseline assumption was a 10% reduction in fracture risk after prophylaxis. A sensitivity analysis was also conducted to assess the reliability and variability of the results.
Results
With a 10% fracture risk reduction, model A costs between $8,850 and $46,940 per quality-adjusted life-year ($/QALY). Additionally, it proved most cost-effective in the age range between 61 and 81 years. The sensitivity analysis established that a reduction of ≥ 2.8% is needed for prophylaxis to be definitely cost-effective. The cost-effectiveness at the secondary prevention level was most sensitive to the cost of the contralateral side’s prophylaxis, the patient’s age, and fracture treatment cost. For high-risk patients with no fracture history, the cost-effectiveness of a preventive strategy depends on their risk profile. In the baseline analysis, the incremental cost-effectiveness ratio at the primary prevention level varied between $11,000/QALY and $74,000/QALY, which is below the defined willingness to pay threshold.
Conclusion
Due to the high cost of hip fracture treatment and its increased morbidity, surgical prophylaxis strategies have demonstrated that they can significantly relieve the healthcare system. Various key assumptions facilitated the modeling, allowing for adequate room for uncertainty. Further research is needed to evaluate health-state-associated risks.
Background
Post-COVID-19 syndrome (PCS) is a lingering disease with ongoing symptoms such as fatigue and cognitive impairment resulting in a high impact on the daily life of patients. Understanding the pathophysiology of PCS is a public health priority, as it still poses a diagnostic and treatment challenge for physicians.
Methods
In this prospective observational cohort study, we analyzed the retinal microcirculation using Retinal Vessel Analysis (RVA) in a cohort of patients with PCS and compared it to an age- and gender-matched healthy cohort (n = 41, matched out of n = 204).
Measurements and main results
PCS patients exhibit persistent endothelial dysfunction (ED), as indicated by significantly lower venular flicker-induced dilation (vFID; 3.42% ± 1.77% vs. 4.64% ± 2.59%; p = 0.02), narrower central retinal artery equivalent (CRAE; 178.1 [167.5–190.2] vs. 189.1 [179.4–197.2], p = 0.01) and lower arteriolar-venular ratio (AVR; (0.84 [0.8–0.9] vs. 0.88 [0.8–0.9], p = 0.007). When combining AVR and vFID, predicted scores reached good ability to discriminate groups (area under the curve: 0.75). Higher PCS severity scores correlated with lower AVR (R = − 0.37 p = 0.017). The association of microvascular changes with PCS severity were amplified in PCS patients exhibiting higher levels of inflammatory parameters.
Conclusion
Our results demonstrate that prolonged endothelial dysfunction is a hallmark of PCS, and impairments of the microcirculation seem to explain ongoing symptoms in patients. As potential therapies for PCS emerge, RVA parameters may become relevant as clinical biomarkers for diagnosis and therapy management.
We consider time-dependent portfolios and discuss the allocation of changes in the risk of a portfolio to changes in the portfolio’s components. For this purpose we adopt established allocation principles. We also use our approach to obtain forecasts for changes in the risk of the portfolio’s components. To put the approach into practice we present an implementation based on the output of a simulation. Allocation is illustrated with an example portfolio in the context of Solvency II. The quality of the forecasts is investigated with an empirical study.
On the applicability of several tests to models with not identically distributed random effects
(2023)
We consider Kolmogorov–Smirnov and Cramér–von-Mises type tests for testing central symmetry, exchangeability, and independence. In the standard case, the tests are intended for the application to independent and identically distributed data with unknown distribution. The tests are available for multivariate data and bootstrap procedures are suitable to obtain critical values. We discuss the applicability of the tests to random effects models, where the random effects are independent but not necessarily identically distributed and with possibly unknown distributions. Theoretical results show the adequacy of the tests in this situation. The quality of the tests in models with random effects is investigated by simulations. Empirical results obtained confirm the theoretical findings. A real data example illustrates the application.