48674 research outputs found
Sort by
Spatial characterization of human-puma interactions in social-ecological land systems of Argentina
No full textInteractions between humans and carnivores are widespread and shaped by diverse social-ecological factors. In Argentina, interactions with pumas (Puma concolor) primarily involve livestock depredation and hunting. Since both interaction types are threatening the coexistence between humans and pumas, understanding their spatial patterns is essential for promoting sustainable coexistence. We identified 90 interaction locations (51 depredation, 39 hunting) from online news and literature (2017–2022) and analyzed spatial social-ecological variables influencing these interactions using generalized linear models. Depredation interactions were associated with lower cattle density, higher human population density, and increased agriculture, while hunting interactions were linked to a higher number of settlements and greater small-livestock density. Within Social-Ecological Land Systems (SELS), these interactions occurred predominantly in “urbanized large-scale agricultural plains” and “low-diversity cold and temperate grassy rangelands”. These findings advance the understanding of interaction dynamics in socially and ecologically complex landscapes, highlighting the need for targeted policies for different regions
Epidemiological cut-off values for Vibrio parahaemolyticus calculated from disc diffusion inhibition zone size data generated at 35 and 28°C
No full textThis work was performed to generate the data needed to set epidemiological cut-off values for inhibition zone data of 8 antimicrobial agents against Vibrio parahaemolyticus determined using standardised disc diffusion protocols with incubation at 35 ± 2°C for 16 to 20 h and at 28 ± 2°C for 24 to 28 h. The zone diameter data aggregated from 3 to 5 laboratories were analysed by the normalised resistance interpretation algorithm. Cut-off values calculated from data obtained at 35 and 28°C were ≥23 and ≥24 mm for ceftazidime, ≥18 and ≥20 mm for enrofloxacin, ≥28 and ≥29 mm for florfenicol, ≥14 and ≥14 mm for gentamicin, ≥29 and ≥32 mm for meropenem, ≥17 and ≥18 mm for oxolinic acid, ≥22 and ≥24 mm for oxytetracycline, as well as ≥19 and ≥21 mm for trimethoprim/sulfamethoxazole. The influence of the incubation temperature on inhibition zone sizes was investigated by calculating the difference between the zones obtained at 35 and 28°C for a specific antimicrobial agent with a particular isolate by an individual laboratory. The mean of the differences calculated for 1314 paired observations was 0.38 mm with a standard deviation of 2.68 mm. The data generated in this work will be submitted to the Clinical and Laboratory Standards Institute and the European Committee on Antimicrobial Susceptibility Testing for consideration in their setting of internationally agreed-upon epidemiological cut-off values for V. parahaemolyticus that are essential for interpreting antimicrobial susceptibility testing data of this species
Automated detection and position control of therapeutic aids in the chest X-ray using deep learning
No full textDie vorliegende Arbeit befasst sich mit der automatisierten Erkennung und Positionskontrolle von vier spezifischen Arten von Therapiehilfen (Endotrachealtubus bzw. Trachealkanüle, verschiedene Arten von Kathetern, Thoraxdrainagen und Magensonden) in Röntgenthorax Bildern unter Verwendung von tiefen neuronalen Netzwerken. Durch den Einsatz von Segmentationsmodellen und dem Self-Aware-Mechanismus soll die röntgenbasierte Positionskontrolle dieser Therapiehilfen, die in der Intensiv- und Notfallmedizin eine zentrale Rolle spielen, optimiert und automatisiert werden. Dabei werden die Herausforderungen der Positionierung in der klinischen Praxis adressiert, die aufgrund der kritischen Bedeutung der korrekten Positionierung für die Patientensicherheit und Therapieeffizienz von hoher Relevanz sind.
Zunächst wurden drei verschiedene methodische Ansätze zur Segmentierung von Therapiehilfen untersucht, die sich in der Komplexität und Spezialisierung der verwendeten Modelle unterscheiden. Dabei zeigte sich, dass der Ansatz mit spezialisierten Modellen für jede Art von Therapiehilfe (Ansatz II) die besten Ergebnisse liefert. Die Evaluierung der Modelle basierte auf den Klassifikationsmetriken AUROC, Sensitivität, Spezifität, Genauigkeit und balancierte Genauigkeit sowie den Segmentierungsmetriken Dice und Dice+, letztere wurde speziell für diese Arbeit entwickelt. Insbesondere die SegResNet-Architektur zeichnete sich durch eine überlegene Segmentationsleistung aus.
Ein weiterer Schwerpunkt war die Entwicklung und Evaluierung des Self-Aware-Mechanismus (SA), der in fast allen Szenarien zu einer Verbesserung der Segmentationsleistung führte. Dieser Mechanismus ermöglicht es dem Modell, seine eigenen Vorhersagen zu überprüfen und zu verfeinern, was insbesondere bei der Endpunktverfeinerung und dem Auffüllen oder Verbinden unvollständiger Segmente von Vorteil ist.
Die Untersuchung des Einflusses von Trainingsdatenmenge und Bildauflösung auf die Segmentationsleistung ergab, dass sowohl eine größere Datenmenge als auch eine höhere Bildauflösung zu besseren Ergebnissen führen, wobei die Verbesserungen mit zunehmender
Datenmenge bzw. Auflösung einem abnehmenden Grenznutzen folgen. Das Ergebnis der Arbeit ist, dass durch den Einsatz von Deep Learning und insbesondere des Self Aware-Mechanismus eine deutliche Leistungssteigerung in der automatisierten Detektion und Positionskontrolle von Therapiehilfen im Röntgenthorax erreicht werden kann. Diese Erkenntnisse bieten wertvolle Ansatzpunkte für die Weiterentwicklung automatisierter Systeme in der medizinischen Bildverarbeitung mit dem Potenzial, die Effizienz und Genauigkeit von automatisierten Kontrollsystemen in der klinischen Praxis signifikant zu verbessern.The present work addresses the automated detection and position control of four specific types of therapeutic aids (endotracheal tube or tracheal cannula, various types of catheters, chest drains and gastric tubes) in X-ray chest images using deep neural networks. By using segmentation models and the self-aware mechanism, the X-ray-based position control of these therapeutic aids, which play a central role in intensive care and emergency medicine, is to be optimized and automated. The challenges of positioning in clinical practice are addressed, which are highly relevant due to the critical importance of correct positioning for patient safety and therapy efficiency.
The study initially examined three different methodological approaches to the segmentation of therapeutic aids, which differ in the complexity and specialization of the models used. This showed that the approach with specialized models for each type of therapy aid therapy aid (approach II) delivers the best results. The evaluation of the models was based on the classification metrics AUROC, sensitivity, specificity, accuracy and balanced accuracy as well as the segmentation metrics Dice and Dice+, the latter developed specifically for this work. The SegResNet architecture in particular was characterized by superior segmentation performance.
Another focus was the development and evaluation of the self-aware (SA) mechanism, which led to an improvement in segmentation performance in almost all scenarios. This mechanism allows the model to check and refine its own predictions, which is particularly useful for endpoint refinement and filling or joining incomplete segments.
The investigation of the influence of training data volume and image resolution on segmentation performance showed that both a larger data quantity and a higher image resolution lead to better results, with the improvements following a decreasing marginal benefit with increasing data quantity or resolution.
The present work shows that the use of deep learning and in particular the self-aware mechanism can achieve a significant increase in performance in the automated detection and position control of therapeutic aids in X-ray chest radiographs. These findings offer valuable starting points for the further development of automated systems in medical image processing with the potential to significantly improve the efficiency and accuracy of automated control systems in clinical practice
Deceptive Ceropegia sandersonii uses an arabinogalactan for trapping its fly pollinators
No full textMany plant species have evolved surfaces that reduce insect attachment. Among such plants are deceptive trap flowers of Ceropegia. Their gliding zones consist of convex epidermal cells, each with a bristle-like central protuberance and a single small liquid droplet on its tip. So far, the molecular and physical mechanisms controlling the function of these droplets are unknown.
We analyzed the droplets of Ceropegia sandersonii flowers by microscopic approaches, studied how they behave when getting in contact with the feet of fly pollinators, and analyzed their chemical composition.
The droplets contaminate the insect feet, on which they solidify. As its main component, a negatively charged polysaccharide containing a β1,3-galactan backbone and Rha-α1,4-GlcA-β1,6-[Araf-α1,3-]Gal-β1,6 side chains or truncated versions of it was identified. The chemical structure represents a rudimentary version of an arabinogalactan, which is supported by its binding to β-d-glucosyl Yariv reagent. Candidates of arabinogalactan proteins were identified to which the polysaccharide might be connected.
The high amount of GlcA in the polysaccharide helps to explain the unusual physical characteristics of the droplets, like viscoelasticity and hygroscopy. We add a new function to arabinogalactans and discuss why the identified polymer is well suited for catching and temporarily trapping pollinators
Stone and speech in The Winter's Tale: Patronage, oral transmission, and ecology
No full textThis article nuances the familiar binary of art and nature in The Winter's Tale by considering contemporary trends in manufacturing and art patronage that complicated both fields: trends that rendered nature a site for the circulation of cheap commodities, and incentivised wealthy patrons to signal their purchasing power by commissioning pieces so realistic that they seemed to merge with the natural sphere. Building on scholarship on oral culture and ecocriticism, this article suggests that Hermione's ‘awakening’ speech might be read as an extension of the movement towards hyperrealism, redeeming orality from contemporary prejudices and presenting it, instead, as an aesthetic artefact of its own.Cet article nuance la dichotomie familière entre l’art et la nature dans The Winter's Tale en examinant les tendances contemporaines dans les domaines de la fabrication et du mécénat artistique qui ont compliqué les deux : les tendances qui ont fait de la nature un lieu de circulation de marchandises bon marché et incité les riches mécènes à afficher leur pouvoir d’achat en commandant des œuvres si réalistes qu’elles semblaient se fondre dans la sphère naturelle. S’appuyant sur des études sur la culture orale et l’écocritique, cet article suggère que le discours d’éveil d’Hermione pourrait être interprété comme une extension du mouvement vers l’hyperréalisme, réhabilitant l’oralité face aux préjugés contemporains et la présentant plutôt comme un artefact esthétique à part entière
Imperial systems and local landscapes of Buldan Yayla in Western Anatolia (Türkiye) during the last 4000 years: An integrated palynological, historical, and archaeological approach
No full textThis study investigates long-term impacts of empires on local socio-ecosystems in western Anatolia (modern western Türkiye) over the past four millennia. We focus on Buldan Yayla Lake, located in a small mountain basin north of the Büyük Menderes (Great Meander) River valley. By examining palynological data alongside historical and archaeological records, we show how four major empires—Hittite, Roman, Byzantine, and Ottoman—shaped land use, vegetation, and resource management practices in a specific locality within a wider regional and transregional economic network. The ebb and flow of empire resulted in cycles of land use intensification and rewilding, resembling broader patterns of regional integration and fragmentation. The different administrative and economic structures of each empire, however, left distinct ecological imprints, with evidence of shifts from extensive pastoralism to specialised crop cultivation, or from olive-focused agriculture to one dominated by mixed agriculture. These shifts underscore both the variation and the adaptability of local socio-ecological systems within broader imperial networks and highlight the interplay of transregional and local factors in landscape transformation
Identification of the Genetic Basis of Phage Resistance in Sequentially Generated Phage-Resistant Klebsiella pneumoniae Using an Established Phage Library
No full textObjectives: To explore the genetic basis of phage resistance in sequentially generated capsular mutants of phage-resistant Klebsiella pneumoniae using an established phage library. Methods: Sequential induction strategies were employed to obtain phage-resistant K. pneumoniae capsular mutants by exposing ST11-K64 K. pneumoniae Kp2325 to different single phages. Whole genome sequencing and bioinformatic analysis were used to elucidate the capsular-related genetic changes in phage-resistant mutants. Phenotypic changes were assessed through gene complementation, growth assays, phage cleavage spectrum analysis, TEM for phage morphology, CPS analysis, biofilm formation, and virulence assays. Results: Three sequentially generated phage-resistant K. pneumoniae capsular mutants were obtained, designated R1, R2 and R3. The narrowing of the phage cleavage spectrum and the evolutionary trade-offs of biological phenotypes were observed. Key genetic changes included: (1) ISKpn26 insertion disrupting wcaJ in R1; (2) combined wcaJ insertion and 9-bp deletion in waaH in R2; and (3) CPS gene cluster deletion in R3 were identified as key mechanisms of phage resistance in K. pneumoniae mutants R1, R2 and R3, respectively. Conclusions: Sequential exposure to different single phages led to rapid evolution of phage resistance in K. pneumoniae via genetic mutations that disrupt capsular synthesis. These findings highlight the critical role of bacterial capsule in phage–host interactions and emphasize the need to use phage cocktails targeting different types of receptors to counteract the evolution of bacterial defense mechanisms in phage therapy
Ein Beitrag zur Kenntnis der Stomatitis pustulosa contagiosa mit bösartigem Verlauf beim Pferd
No full textDigitalisat der Ausgabe von 1921, erschienen 202
Characterizing Nonuniform Hyperbolicity by Mather-Type Admissibility
No full textWe consider linear cocycles acting on Banach spaces which satisfy the assumptions of the multiplicative ergodic theorem. A cocycle is nonuniformly hyperbolic if all Lyapunov exponents are non-zero, which is equivalent to the existence of a tempered exponential dichotomy. We provide an equivalent characterization of nonuniform hyperbolicity in terms of a Mather-type admissibility of a pair of weighted function spaces. As an application we give a short proof of the robustness of tempered exponential dichotomies under small linear perturbation
Der röntgenologische Nachweis der Fisteln beim Pferde
No full textDigitalisat der Ausgabe von 1917, erschienen 202