Hochschule Bielefeld University of Applied Sciences and Arts

Zusammenfassung Aufgrund des hohen Aufwands von Primärstudien bietet sich die Nutzung von bestehenden Daten, sogenannten Routinedaten, für die Beantwortung insbesondere von versorgungsnahen Forschungsfragen in der Rehabilitation an. Bisherige Übersichtsarbeiten zur Routinedatennutzung fokussieren auf die Akutversorgung im Zuständigkeitsbereich der Gesetzlichen Krankenversicherung (GKV), lassen jedoch die Rehabilitation und andere Reha-relevante Leistungsträger wie die Deutsche Rentenversicherung (DRV), die Deutsche Gesetzliche Unfallversicherung (DGUV) und die Bundesagentur für Arbeit (BA) weitgehend außen vor. Ziel dieses Beitrags ist daher, einen Überblick über Art, Zugang, Qualität und datenschutzrechtliche Aspekte zu Routinedaten im Kontext der Rehabilitationsforschung zu geben. Bestehende Empfehlungen, Ergebnisse einer selektiven Literaturrecherche und eigene Erfahrungen wurden herangezogen. Routinedaten zeichnen sich durch die sehr hohe Fallzahl, den oft großen Merkmalsumfang und die längsschnittliche Dokumentation über lange Zeiträume aus. Der Zugang zu Routinedaten der Deutschen Rentenversicherung und der Bundesagentur für Arbeit ist für Forschende vergleichsweise niedrigschwellig, während dies für Routinedaten anderer Leistungsträger und von Leistungserbringern bislang nicht gleichermaßen der Fall ist. Weiterhin können unter bestimmten Voraussetzungen Routinedatensätze untereinander und mit Primärdaten verknüpft werden, was die Anwendungsmöglichkeiten deutlich erweitern kann. Neben den Vorteilen von Routinedaten sind deren Limitationen zu beachten. Routinedaten wurden für andere Zwecke erhoben und enthalten meist nur Merkmale, die für die Administration erforderlich sind. Ein prospektiver Studienansatz mit Routinedaten ist aufgrund der kontinuierlichen Datenerhebung und -dokumentation grundsätzlich möglich, jedoch ist keine randomisierte Zuweisung zu Interventionen umsetzbar. Zudem sind Generalisierbarkeit und Qualität einschließlich psychometrischer Eigenschaften von Datensätzen und einzelner Variablen zu prüfen, ebenso die Verfügbarkeit von Routinedatensätzen. Das im Aufbau befindliche Forschungsdatenzentrum Gesundheit sieht bislang weder eine Integration von GKV-Rehabilitationsdaten noch die Verknüpfung von GKV-Daten mit Daten anderer Reha-relevanter Leistungsträger vor. Datenschutzrechtliche Aspekte sind ebenfalls bedeutsam. Bei Nutzung von pseudonymisierten Daten von Sozialversicherungsträgern ist durch die Datenhalter ein Antrag nach § 75 SGB X bei den zuständigen Aufsichtsbehörden zu stellen.

We present a comprehensive study of three-dimensional arrays of nanostructured Co3Fe tetrapods consisting of four pillars each with tetrahedral symmetry, prepared by focused electron beam-induced deposition and placed in two distinct orientations with respect to the direction of an external magnetic field. Using ultra-sensitive micro-Hall magnetometry, we obtain angular-dependent magnetic stray field hysteresis loops, which reveal a characteristic yet complex magnetization reversal behavior of the structures during a field sweep. We find that micromagnetic simulations significantly deviate from the experimental findings due to inherent limitations of the method. As an alternative approach, we derive a macrospin model that enables us to elucidate the observed hysteresis curves through the cascading switching dynamics of dipolar-coupled magnetic grains.

Background/Objectives: This study analyzed the stability of individual preferences for the allocation of expenditure in the healthcare system using an experimental setting. Understanding these preferences can support policy decisions aimed at achieving a more needs-based allocation of scarce resources in healthcare systems. Stability in preferences might be essential in order to avoid frequent legislative changes and can potentially enhance public satisfaction with the healthcare system. Methods: Individual preferences were assessed through two questionnaire-based experimental studies conducted before and after the COVID-19 pandemic, each with about 160 participants, in the context of a healthcare seminar in the MaxLab of the Otto-von-Guericke-University Magdeburg, Germany. This study was intended as a preliminary study for a larger follow-up panel study. In particular, the questionnaire contained questions regarding satisfaction with the healthcare system, optimization options, possible maximum contributions, and preferences for the allocation of notional healthcare budget and research funds in order to provide initial evidence regarding the stability of such preferences. As the data were collected both before and after the COVID-19 pandemic, this significant change in the situation helps to provide clear indications of stability. The preferences collected were compared to the actual allocation of expenditure derived from official statistics in order to identify potential areas for policy adjustment. Results: Preferences for the allocation of healthcare expenditure appear to be relatively stable despite the effects of the pandemic. However, noticeable discrepancies exist between individual preferences and actual healthcare spending. Satisfaction with the healthcare system also remains relatively stable at a high level. Conclusions: Overall, the scientific measurement of public preferences could support more informed political decision-making and contribute to sustained satisfaction with the healthcare system. In particular, the distribution of funds to different disease categories should be adjusted on the basis of such preferences, taking into account the respective medical indications after representative regular surveys have been carried out.

Active Learning (AL) is a powerful method to efficiently gather data for the training of machine learning models. Particularly effective are AL methods that consider both the representativeness and informativeness of the data, allowing the methods to balance the exploration and exploitation. However, when AL is used in real applications, such as modelling an industrial process or a product, the resulting models often cannot be validated sufficiently due to a lack of independent test data. In this case, Decision Trees (DTs) offer a major advantage over other more established models in the field of AL. The intrinsic interpretability of DTs allows for validation without the use of independent test data. In this work, we present a novel AL method based on DTs. In addition to an intrinsically interpretable classification model, our new method utilizes different exploration and exploitation properties of the learned tree structure to gather training data. The behavior of our method is illustrated and the performance is compared to other established AL methods in an experimental study using third-party data sets. Our comparison demonstrates that the DTs trained with our method converge to a lower test error at a faster rate than DTs trained with the competing methods.

The research area of active learning (AL) has been a subject of continuous attention over the past few decades. However, the application of AL in real-world scenarios remains considerably less popular than might be expected, given the success and advances of the research in this area. As this discrepancy has already been identified by previous researchers, an overview of the literature concerning this issue is given. The identified challenges are grouped and categorized, and existing approaches and solutions that address them are presented. Based on the identified challenges, we propose the hypothesis that a lack of trust in the success of AL is a key factor contributing to the disparity between the advancements of applications and research. In particular, the difficulty of evaluating the performance of an AL process, due to the absence of independent test data, is a major contributor to this trust issue. Finally, we identify research areas and questions that should receive greater attention in order to enhance trust in AL methods and encourage their adoption.

Background: Muscle fatigue affects motor neuron function and, in turn, the electromyography\,(EMG) signals used to control wearable technologies. By accounting for this, we can more accurately regulate movement assistance. Muscle fatigue has been widely studied in the fields of sports science, rehabilitation, and occupational health. However, there are contradictory findings in the literature, and some aspects of muscle fatigue mechanisms are not yet well understood. Furthermore, none of the literature found focuses on the detection of early-stage muscle fatigue. Results: In this paper we specifically focus on early-stage muscle fatigue and show that the spectral variance ($(S_{var})$) of the EMG signal is a more reliable measure of fatigue across participants and contraction levels compared to other widely used features. We then make recommendations on how these findings can be used in EMG-controlled assist-as-needed technologies. Experiments were carried out with 16 participants who performed intermittent isometric contractions of the biceps brachii at 20\,%, 50\,% and 75\,% MVC. The reliability of the motor unit action potential conduction velocity (CV) and the median/mean power frequencies (MDPF/MNPF) as measures of fatigue was tested. CV exhibited some linear correlation with perceived fatigue only at the higher contraction levels, for the majority of participants (R2=$(0.18\pm0.23)$). Conclusions: CV was found to be an unreliable measure for muscle fatigue and no relationship was found between MDPF/MNPF and fatigue. However, for all participants and contraction levels, the fatigue-$(S_{var})$ relationship could be characterized using a Gaussian model\,(R2=$(0.87\pm0.12)$).

This research, comprising three experiments with a total of 1718 population‐representative participants, investigates the strategies Muslim organizations can utilize to sustain trust and positive perceptions in the direct aftermath of terrorist attacks. It evaluates the effectiveness of different crisis communication strategies as outlined by the Situational Crisis Communication Theory. Additionally, it examines the effects of a positive pre‐crisis reputation, statement framing and the publishing source on attitudes towards Muslim organizations, Muslims in general and Islam. Three experiments with several reference groups were conducted. Multivariate analyses underscore the critical importance of active crisis communication in cultivating positive attitudes and trust in Muslim organizations. Across experiments, the findings indicate that the act of issuing a statement itself holds more substantial influence than the specific crisis response strategy employed. In addition, the source of publication played a notable role in shaping perceptions; statements released through personal channels resulted in more positive reactions compared to statements released by a public source.

Weather plays a broad and decisive role in many areas. Its volatility can disrupt traffic and endanger lives. It is therefore imperative to accurately predict its impact. Improved forecasting accuracy can aid multi-industry decision making. Traffic authorities can control traffic in advance; Agriculture can adjust its strategy in time; Resource allocation can be optimized in the energy sector. The rise of machine learning and deep learning technologies has opened up new prospects for weather-related forecasting. This article takes a methodical look at the application of machine learning and deep learning to traffic and weather forecasting, dissecting the details of model construction, data processing processes, and performance evaluation metrics. By comparing the advantages and disadvantages of each model, it provides ideas for model improvement, powerfully guides future research direction, and lays the foundation for building a more accurate prediction framework. The research direction of this thesis has far-reaching theoretical and practical value.

Microalgal biotechnology is gaining attention due to its potential to produce pigments, lipids, biofuels, and value-added products. However, challenges persist in terms of the economic viability of microalgal lipid production in photobioreactors due to slow growth rates, expensive media, complex downstream processing, limited product yields, and contamination risks. Recent studies suggest that co-cultivating microalgae with bacteria can enhance the profitability of microalgal bioprocesses. Immobilizing bacteria offers advantages such as protection against shear forces, the prevention of overgrowth, and continuous product secretion. Previous work has shown that biopolymeric immobilization of Paenibacillus polymyxa enhances 2,3-butanediol production. In this study, a novel co-fermentation process was developed by exploiting the chemical crosstalk between a freshwater microalga Scenedesmus obliquus, also known as Tetradesmus obliquus, and an immobilized plant-growth-promoting bacterium, Paenibacillus polymyxa. This co-cultivation resulted in increased metabolite production, with a 1.5-fold increase in the bacterial 2,3-butanediol concentration and a 3-fold increase in the microalgal growth rates compared to these values in free-cell co-cultivation. Moreover, the co-culture with the immobilized bacterium exhibited a 5-fold increase in the photosynthetic pigments and a 3-fold increase in the microalgal lipid concentration compared to these values in free-cell co-cultivation. A fixed bed photobioreactor was further constructed, and the co-cultivation bioprocess was implemented to improve the bacterial 2,3-butanediol and microalgal lipid production. In conclusion, this study provides conclusive evidence for the potential of co-cultivation and biopolymeric immobilization techniques to enhance 2,3-butanediol and lipid production.

This paper analyses diversity and intersectionality aspects in the R&D of wearable assistive and rehabilitation technologies (WEARTechs). We advocate for inclusive, innovative research that we hope will help bridge the gap between laboratories and the real world and reduce disparities in healthcare and technology development. We performed a systematic literature review of the intersections between assistive technologies and diversity and conducted a thematic analysis of the diversity factors identified in the literature. In addition, we carried out a supplementary literature search on WEARTechs to discover which, if any, diversity aspects are currently being reported on. Our findings indicate that diversity has not been addressed in the field of WEARTechs. There is not sufficient knowledge to determine, which diversity-related aspects researchers must consider when evaluating the performance of any specific WEARTech device. Nor about how these can be properly addressed in the R&D process. We, therefore, provide actionable recommendations on how to integrate diversity-relevant aspects at different R&D stages. We hope that our review will help scientists rethink and reformulate approaches to the R&D of WEARTechs and build the way towards more inclusive solutions. It is our belief that this will spark innovation and enhance discovery potential in the field.

A model of a “smart” composite based on a thermosensitive PNIPAM polymer deposited on a FeRh substrate with a modified periodic microstructure was proposed. The initial parameters of the model were determined from the properties of the actual composite sample and its components. Cooling of the sample using a magnetic field was shown by two independent methods, and at ~37 °C, it was −5.5 °C when a magnetic field of 1.8 T was applied. Based on experimental data, models of traditional and modified PNIPAM/FeRh composites were constructed. Calculations show that surface modification allows for an increase in the activation time for a polymer layer that is 20 µm thick from ~20 ms for a conventional composite to ~60 ms for a modified composite. Modification of the surface in the form of wells can be used to more effectively implement the idea of loading and releasing drugs for potential biomedical applications.

Data are crucial components of machine learning and deep learning in real-world applications. However, when collecting data from actual systems, we often encounter issues with missing information, which can harm accuracy and lead to biased results. In the context of video surveillance, missing data may arise due to obstructions, varying camera angles, or technical issues, resulting in incomplete information about the observed scene. This paper introduces a method for handling missing data in tabular formats, specifically focusing on video surveillance. The core idea is to fill in the missing values for a specific feature using values from other related features rather than relying on all available features, which we refer to as the imputation approach based on informative features. The paper presents three sets of experiments. The first set uses synthetic datasets to compare four optimization algorithms—Particle Swarm Optimization (PSO), Grey Wolf Optimization (GWO), Whale Optimization Algorithm (WOA), and the Sine–Cosine Algorithm (SCA)—to determine which one best identifies features related to the target feature. The second set works with real-world datasets, while the third focuses on video-surveillance datasets. Each experiment compares the proposed method, utilizing the best optimizer from the first set, against leading imputation methods. The experiments evaluate different types of data and various missing-data rates, ensuring that randomness does not introduce bias. In the first experiment, using only synthetic data, the results indicate that the WOA-based approach outperforms PSO, GWO, and SCA optimization algorithms. The second experiment used real datasets, while the third used tabular data extracted from a video-surveillance system. Both experiments show that our WOA-based imputation method produces promising results, outperforming other state-of-the-art imputation methods.

Biocompatible hydrogels are versatile platforms for encapsulating living cells in biotechnology due to their unique physical, structural and mechanical properties. The diffusion of dissolved carbon dioxide (dCO2) into the hydrogel matrix is of great importance for the growth of immobilised photosynthetic cells like microalgae and cyanobacteria. However, non-invasive analysis methods for measuring the diffusion of dCO2 in hydrogels are limited. In this article, we describe an indirect method for the non-invasive measurement of diffusion rates for dCO2 in hydrogels. We visually tracked the diffusion along the axial direction of pH indicator-doped hydrogel monoliths by recording the interface position over time. We calculated the interface velocity and the pseudo diffusion coefficients (Dpseudo) over time. The obtained Dpseudo values are in a realistic range compared to literature values. Therefore, this novel analysis method for dCO2 diffusion gained valuable insights into diffusion dynamics in different hydrogels and can aid in the design of better immobilisation matrices for photosynthetic cells.•Non-invasive, rapid method for estimation of dissolved CO2 (dCO2) diffusion in hydrogels •Automatic analysis of colour interface formation due to acidification of hydrogels by diffusing dCO2 •Agarose hydrogels exhibit an approximated 30x higher pseudo dCO2 diffusion coefficient than silica gel

In vertebrates and plants, dsRNA plays crucial roles as PAMP and as a mediator of RNAi. How higher fungi respond to dsRNA is not known. We demonstrate that Magnaporthe oryzae (Mo), a globally significant crop pathogen, internalizes dsRNA across a broad size range of 21 to about 3000 bp. Incubation of fungal conidia with 10 ng/µL dsRNA, regardless of size or sequence, induced aberrant germ tube elongation, revealing a strong sequence-unspecific effect of dsRNA in this fungus. Accordingly, the synthetic dsRNA analogue poly(I:C) and dsRNA of various sizes and sequences elicited canonical fungal stress pathways, including nuclear accumulation of the stress marker mitogen-activated protein kinase Hog1p and production of ROS. Leaf application of dsRNA to the cereal model species Brachypodium distachyon suppressed the progression of leaf blast disease. Notably, the sequence-unspecific effect of dsRNA depends on higher doses, while pure sequence-specific effects were observed at low concentrations of dsRNA ( < 0.03 ng/µL). The protective effects of dsRNA were further enhanced by maintaining a gap of at least seven days between dsRNA application and inoculation, and by stabilising the dsRNA in alginate-chitosan nanoparticles. Overall, our study opens up additional possibilities for the development and use of dsRNA pesticides in agriculture.