3,475 research outputs found
Designing Urban Inclusion
No full textThis book presents the productions of the first international MasterClass hosted by Metrolab in January and February of 2017, on the topic of inclusion in urban spaces and urban projects. The event is the first stage of a larger project conducted at Metrolab and involving collective and collaborative research. We would like to begin with a word about this project that is dear to us
L-R: Katie Lee; Leo Walters; Bruce Berger sitting on a boat on the Colorado River.
No full textPhoto of Photo of Arizona folk singer and author Katie Lee (far left), Leo Walters (center), and writer Bruce Berger (far right), sitting on a raft on the Colorado River, Glen Canyon, Uta
SMM891195 Supplemental Material - Supplemental material for Flexible modeling of ratio outcomes in clinical and epidemiological research
No full textSupplemental material, SMM891195 Supplemental Material for Flexible modeling of ratio outcomes in clinical and epidemiological research by Moritz Berger and Matthias Schmid in Statistical Methods in Medical Research</p
Pressure-Induced Upregulation of Preproendothelin-1 and Endothelin B Receptor Expression in Rabbit Jugular Vein In Situ
No full text
Abstract
—Upregulation of endothelin-1 (ET-1) synthesis in venous bypass grafts in response to arterial levels of blood pressure may play a major role in graft failure. To investigate this hypothesis, isolated segments of the rabbit jugular vein were perfused at physiological (0 to 5 mm Hg) and nonphysiological (20 mm Hg) levels of intraluminal pressure. As judged by reverse transcription–polymerase chain reaction analysis (mRNA level), neither endothelin-converting enzyme nor endothelin A receptor expression appeared to be pressure sensitive. In contrast, there was a profound and time-dependent increase in endothelial prepro-ET-1 mRNA and intravascular ET-1 abundance (by ELISA) as well as in smooth muscle endothelin B receptor mRNA and functional protein (by superfusion bioassay) on raising the perfusion pressure from 5 to 20 mm Hg, but not from 0 to 5 mm Hg, for up to 12 hours. Video microscopy analysis revealed that the segments were distended by 75% at 5 mm Hg and near maximally at 20 mm Hg compared with the resting diameter at 0 to 1 mm Hg. Treatment of the segments with actinomycin D (1 μmol/L), the specific protein kinase C inhibitor, Ro 31–8220 (0.1 μmol/L), or the c-Src family–specific tyrosine kinase inhibitor, herbimycin A (0.1 μmol/L), demonstrated that the pressure-induced expression of these gene products occurs at the level of transcription and requires activation of protein kinase C, but not c-Src. In venous bypass grafts such deformation-induced changes in gene expression may contribute not only to acute graft failure through ET-1–induced vasospasm but also to endothelin A receptor– and/or endothelin B receptor–mediated smooth muscle cell hyperplasia and graft occlusion.
</jats:p
Elevated Perfusion Pressure Upregulates Endothelin-1 and Endothelin B Receptor Expression in the Rabbit Carotid Artery
No full textTo investigate the hypothesis that high blood pressure activates the endothelin system in the vessel wall, isolated segments of the rabbit carotid artery were subjected to different levels of perfusion pressure. Both preproendothelin-l (ppET-1) mRNA abundance and intravascular ET-1 peptide content were strongly upregulated on raising the intraluminal pressure from 90 to 160 mm Hg for 3 to 12 hours, and this increase in ppET-1 mRNA occurred predominantly in the endothelial cells. Endothelin-converting enzyme-1 and endothelin A receptor (ETA-R) expression were pressure-insensitive, whereas that of the ETB-R in the smooth muscle cells was also significantly enhanced. Both the pressure-induced increase in ppET-1 and ETB-R expression required RNA synthesis because they were abolished by actinomycin D. The nuclear signaling mechanisms involved therein, however, appeared to be different. Thus, the pressure-induced expression of ppET-1 and activation of CCAAT-enhancer binding proteins beta and delta were blocked by the tyrosine kinase inhibitor herbimycin A, whereas ETB-R expression and the nuclear translocation of activator protein-1 were abolished by the protein kinase C inhibitor Ro 31-8220. One consequence of these presumably deformation-induced changes in gene expression was an increased rate of apoptosis of the smooth muscle cells in the media that if transferable to the situation in human blood vessels may contribute to hypertension-induced arterial remodeling
Advanced calibration methods for nonlinear multisensor systems
No full textThis thesis reports on advanced calibration methods for nonlinear multisensor systems (MSSs). The application of half-blind calibration (HBC) is ex- tended to nonlinear MSSs, and a probabilistic calibration method based on Bayesian inference is investigated. The effectiveness of the methods is evaluated using a complementary metal oxide semiconductor (CMOS)-integrated Hall-sensor system designed to measure the out-of-plane component of the magnetic induction.The method of HBC is extended to sensor systems with a nonlinear response. It directly addresses the inverse calibration problem by treating the sensor signals as independent variables, and the measurands as dependent variables. If a MSS is used for inference of not all physical quantities from the sensor signals, then the HBC enables an efficient calibration, since only the quantitative values of the measurand are needed for a successful calibration. During HBC, the MSS is exposed to both the measurands and disturbances while ensuring that for both quantities the range of interest is sufficiently covered. However, only the measurand values and the output signals of the MSSs need to be recorded. For each such measurand, the extracted polynomial function of the sensor signals predicts cross-sensitivity-compensated values. The method is demonstrated using a Hall-stress-temperature sensor system, where both temperature and mechanical stress are sources of cross-sensitivities. For a polynomial regression model of degree four, the calibration of 20 MSSs in the ranges of ±21mT, −40 to 125 °C, and up to −80 MPa of in-plane stress, achieves an accuracy of 136 μT.Building upon the HBC, the Bayesian sensor calibration is investigated with a focus on keeping potentially high calibration costs low in terms of time and resources. In contrast to the HBC based on ordinary least squares (OLS) calibration, the Bayesian sensor calibration allows a calibration with fewer measurements than model parameters. This is achieved using Bayesian inference by combining calibration data of a sensor with prior information about the ensemble to which it belongs. A Bayesian experimental design under I-optimality and G-optimality allows to identify optimal calibration conditions. The method is first investigated with a Hall sensor system that has a single cross-sensitivity to temperature, and secondly with a similar MSS possessing two cross-sensitivities, namely temperature and mechanical stress. In both frameworks, the nonlinear sensor response curves are modeled by polynomial basis functions to predict the out-of-plane magnetic induction B.In the first study, the Hall-temperature sensor system requires seven parameters in the temperature range between −30 and 150 °C and for magnetic field values B between −25 and 25 mT. For the prior, a multivariate normal distribution of the model parameters is acquired using 14 specimens of the sensor ensemble. I-optimal calibration at one, two, and three temperatures reduces the root mean square (rms) standard deviation of B, inferred from sensor output signals, from 203μT before calibration down to 78, 41, and 34 μT, respectively.In the second study, the packaged Hall sensor system comprises a Hall sensor, a stress sensor, and a temperature sensor. They are combined in a polynomial sensor response model with 11 parameters to infer B. For the calibration, sensors are exposed to mechanical stress values between 0 and −68MPa, temperatures between −40 and 100°C, and B values between −25 and 25mT. A sample of 35 sensors serves to extract the prior model parameter distribution, and 15 sensors serve for validation. The Bayesian experimental design is applied to identify sets of 2–8 optimal calibration conditions. In the case of I-optimality, the median rms σ values of the ±1σ confidence intervals for the extracted B values were found to be 113–71 μT after near-I-optimal calibrations based on 2–8 measurements, respectively. The experimentally determined medians of the rms deviations between predicted and applied B values were found to be 89–71 μT over the full range of applied validation conditions.Diese Arbeit befasst sich mit Methoden für die Kalibrierung nichtlinearer Multisensorsysteme (MSS). In diesem Rahmen wurde die halbblinde Kalibrierung (HBC) auf nichtlineare MSS erweitert und ein probabilistischer Ansatz zur Kalibrierung nichtlinearer MSS, basierend auf der Bayes’schen Inferenz, untersucht. Für die Validierung der Methoden dient ein Halbleiter- Hall-Sensorsystem, welches dafür ausgelegt ist, die Komponente des Magnetfelds B senkrecht zur Chipebene zu messen.Die HBC dient der Kalibrierung von Sensorsystemen mit nichtlinearem Verhalten, wobei die Sensorsignale als unabhängige und die Messgrößen als abhänge Variablen betrachtet werden. Dient ein Sensorsystem lediglich der Vorhersage einer Teilmenge der physikalischen Eingangsgrößen, ermöglicht die HBC eine zeit- und kosteneffiziente Kalibrierung, da das quantitative Wissen der Eingangsgrößen, welche nicht von Interesse sind, nicht erforderlich ist. Diese Störgrößen müssen lediglich ausreichend variiert und deren Sensorsignale aufgezeichnet werden, um eine erfolgreiche Kompensation der Querempfindlichkeiten zu gewährleisten. Die Methode wird mit einem Hall-Sensorsystem demonstriert, welches querempfindlich gegenüber mechanischen Spannungen und der Temperatur ist. Nach der Kalibrierung, weisen 20 MSS in einem Magnetfeldbereich von ±21mT, einem Temperaturbereich von −40 bis 125°C und mechanischen Spannungen von bis zu −80 MPa eine Genauigkeit von 136 μT auf.Aufbauend auf der HBC, wird die Bayes’sche Sensorkalibrierung untersucht, unter anderem mit dem Ziel, potentiell hohe Kalibrierkosten hinsichtlich Zeit- und Ressourcenaufwand gering zu halten. Im Gegensatz zur HBC mit der Methode der kleinsten Quadrate, erlaubt die Bayes’sche Sensorkalibrierung eine Kalibrierung mit weniger Messungen als Modellparametern. Das wird durch die Kombination von Kalibrierdaten mit der A-priori-Verteilung der Modellparameter des vorliegenden Ensembles erreicht. Mithilfe eines Bayes’schen Design of Experiment werden opti- male Kalibrierbedingungen identifiziert, wobei sowohl die I- als auch die G-Optimalität untersucht werden. Die Methode wird zuerst mit einem Hall- Sensorsystem, welches eine Querempfindlichkeit gegenüber der Temperatur aufweist, und anschließend mit einem ähnlichem System untersucht, welches zusätzlich querempfindlich gegenüber mechanischen Spannungen ist. In bei- den Anwendungsfällen wird die nichtlineare Sensorantwort mithilfe von polynomialen Basisfunktionen beschrieben.Im ersten Anwendungsfall sind für die Kalibrierung des Hall-Temperatur- Sensorsystems sieben Modellparameter erforderlich in einem Temperaturbereich von −30 bis 150°C und für Magnetfeldwerte B zwischen −25 und 25mT. Für die Prior-Generierung wird eine multivariate Normalverteilung der Modellparameter angenommen, welche auf der Basis von 14 Sensorsystemen berechnet wird. Eine I-optimale Kalibrierung bei einer Temperatur reduziert den rms der Standardabweichungen von B von 203 μT vor der Kalibrierung auf 78 μT. Nach einer Kalibrierung bei zwei und drei Temperaturen wird der Wert weiter auf 41 und 34 μT reduziert.Im zweiten Anwendungsfall wird ein ähnliches Hall-Sensorsystem mit einem zusätzlichen Spannungssensor untersucht. Das verwendete polynomiale Modell mit den drei Sensorsignalen beinhaltet 11 Parameter, um B weitestgehend bereinigt von Quereinflüssen vorhersagen zu können. Während der Kalibrierung werden die Sensoren mechanischen Spannungen von etwa 0 bis −68MPa, Temperaturen zwischen −40 und 100°C, sowie B Werten zwischen −25 und 25 mT ausgesetzt. Eine Stichprobe von 35 Sensoren dient der Prior-Generierung und weitere 15 Sensoren des- selben Ensembles dienen der Validierung. Auch mit diesen komplexeren Rahmenbedingungen erlaubt ein Bayes’sches Design of Experiment die Identifikation von 2–8 optimalen Kalibrierbedingungen. Für den Fall der I-Optimalität ergeben sich Medianwerte der rms σ Werte des ±1σ Konfidenzintervalls für die extrahierten B-Werte von 113–71 μT. Dabei erfolgten die Kalibrierungen an nahezu I-optimalen Bedingungen mit 2–8 Messungen. Die entsprechenden, experimentell ermittelten Mediane der rms-Werte der Residuen betrugen 89–71 μT
Geschichte der Königlichen Berger-Oberrealschule (früher Realschule und Realgymnasium) zu Posen während ihres fünfzigjährigen Bestehens : 1853 - 1903
No full textvon Moritz FriebeIn Fraktur[Progr.-Nr. 208
- …
