Search

ZusammenfassungDie Feinstaubbelastung steht immer wieder im Mittelpunkt umwelt- und gesundheitspolitischer Diskussionen. Empirische Analysen, auch solche in diesem Bereich, die nicht auf individuellen Experimentaldaten beruhen, können als ökologische „real life“-Studien kleine Bausteine zum Verständnis der Zusammenhänge von Feinstaubbelastung und deren Gesundheitsfolgen zur Verfügung stellen. In diesem Beitrag wird deskriptiv untersucht, ob in Deutschland eine Korrelation von Feinstaubbelastung und Lebenserwartung zu finden ist. Dazu werden auf Kreisebene nach den Messbereichen Hintergrund und Verkehr differenzierte Daten von Feinstaubmessstationen der Jahre 2002 bis 2016 verwendet. Als weitere Kovariate der Lebenserwartung werden auf Kreisebene das verfügbare Einkommen pro Kopf und die Abiturientenquote berücksichtigt. Darüber hinaus werden Siedlungseigenschaften (städtische bzw. ländliche Gebiete) sowie ein nichtlinearer Zeittrend in die Untersuchung einbezogen. Die Niveauschätzungen ergeben einen negativen Zusammenhang für die Feinstaubbelastung im westdeutschen Hintergrundbereich. Darüber hinaus zeigt sich ein negativer Zusammenhang zwischen der Veränderung von Feinstaubbelastung und der Veränderung der Lebenserwartung für den westdeutschen Verkehrsbereich, der jedoch aufgrund der Datenlage als unsicherer anzusehen ist. Letzteres ist aufgrund der Besonderheiten dieses Bereichs (tages- und jahreszeitlich bedingte Variationen des Verkehrsflusses; Änderung der Feinstaub-Exposition der Bevölkerung innerhalb kurzer Entfernungen u. v. m.) nicht überraschend. Für Ostdeutschland konnten keine statistisch signifikanten Zusammenhänge gefunden werden. Die für Gesamtdeutschland gefundene Korrelation zwischen Feinstaubbelastung im Hintergrundbereich und durchschnittliche Lebenserwartung liegt bei 0,037 Jahre pro 1 $$\upmu$$ μ g/m$${}^{3}$$ 3 PM$${}_{10}$$ 10 .

Reinforcement Learning (RL) is a fast-growing field of research that is mostly applied in the realm of video games due to the compatibility of RL and game tasks. AI Gym has established itself as the gold standard toolkit for Reinforcement Learning research. Unfortunately, toolkits like AI Gym are very optimized for benchmark purposes and may not always be suitable for real world type problems. Additionally, fixed wing flight simulation has specific requirements and may need other solutions. In this paper, we propose QPlane as an alternative toolkit for RL training of fixed wing aircraft. QPlane was developed in an effort to create a RL toolkit for fixed wing aircraft simulation that is easily modifiable for different scenarios. QPlane is replicable and flexible for ease of implementation to high performance computing, and is modular for quick environment and algorithm replacement. In this paper we present and discuss details of QPlane, as well as proof of concept results.

Background COVID-19 is a lung disease, and there is medical evidence that air pollution is one of the external causes of lung diseases. Fine particulate matter is one of the air pollutants that damages pulmonary tissue. The combination of the coronavirus and fine particulate matter air pollution may exacerbate the coronavirus’ effect on human health. Research question This paper considers whether the long-term concentration of fine particulate matter of different sizes changes the number of detected coronavirus infections and the number of COVID-19 fatalities in Germany. Study design Data from 400 German counties for fine particulate air pollution from 2002 to 2020 are used to measure the long-term impact of air pollution. Kriging interpolation is applied to complement data gaps. With an ecological study, the correlation between average particulate matter air pollution and COVID-19 cases, as well as fatalities, are estimated with OLS regressions. Thereby, socioeconomic and demographic covariates are included. Main findings An increase in the average long-term air pollution of 1 μg/m3 particulate matter PM2.5 is correlated with 199.46 (SD = 29.66) more COVID-19 cases per 100,000 inhabitants in Germany. For PM10 the respective increase is 52.38 (SD = 12.99) more cases per 100,000 inhabitants. The number of COVID-19 deaths were also positively correlated with PM2.5 and PM10 (6.18, SD = 1.44, respectively 2.11, SD = 0.71, additional COVID-19 deaths per 100,000 inhabitants). Conclusion Long-term fine particulate air pollution is suspected as causing higher numbers of COVID-19 cases. Higher long-term air pollution may even increase COVID-19 death rates. We find that the results of the correlation analysis without controls are retained in a regression analysis with controls for relevant confounding factors. Nevertheless, additional epidemiological investigations are required to test the causality of particulate matter air pollution for COVID-19 cases and the severity.

Giardia trophozoites attach to the intestinal microvilli (or inert surfaces) using an undefined “suction-based” mechanism, and remain attached during cell division to avoid peristalsis. Flagellar motility is a key factor in Giardia's pathogenesis and colonization of the host small intestine. Specifically, the beating of the ventral flagella, one of four pairs of motile flagella, has been proposed to generate a hydrodynamic force that results in suction-based attachment via the adjacent ventral disc. We aimed to test this prevailing “hydrodynamic model” of attachment mediated by flagellar motility. We defined four distinct stages of attachment by assessing surface contacts of the trophozoite with the substrate during attachment using TIRF microscopy (TIRFM). The lateral crest of the ventral disc forms a continuous perimeter seal with the substrate, a cytological indication that trophozoites are fully attached. Using trophozoites with two types of molecularly engineered defects in flagellar beating, we determined that neither ventral flagellar beating, nor any flagellar beating, is necessary for the maintenance of attachment. Following a morpholino-based knockdown of PF16, a central pair protein, both the beating and morphology of flagella were defective, but trophozoites could still initiate proper surface contacts as seen using TIRFM and could maintain attachment in several biophysical assays. Trophozoites with impaired motility were able to attach as well as motile cells. We also generated a strain with defects in the ventral flagellar waveform by overexpressing a dominant negative form of alpha2-annexin::GFP (D122A, D275A). This dominant negative alpha2-annexin strain could initiate attachment and had only a slight decrease in the ability to withstand normal and shear forces. The time needed for attachment did increase in trophozoites with overall defective flagellar beating, however. Thus while not directly required for attachment, flagellar motility is important for positioning and orienting trophozoites prior to attachment. Drugs affecting flagellar motility may result in lower levels of attachment by indirectly limiting the number of parasites that can position the ventral disc properly against a surface and against peristaltic flow., Author Summary Giardia is a widespread, single-celled, intestinal parasite that infects millions of people and animals each year. Colonization of the small intestine is a critical part of Giardia's life cycle in any host. This colonization is initiated when cells attach to the intestinal wall via a specialized suction cup-like structure, the ventral disc. In the host, Giardia moves by beating four pairs of flagella; movement of the ventral pair has been implicated in attachment. This study shows that the beating of the flagella is not important for attachment, but rather for positioning Giardia close to the intestinal wall prior to attachment, and thus disproves the commonly held model of giardial attachment. This work implies that drugs targeting Giardia motility could prevent or slow attachment, leading to lower rates of infection.