Suchergebnisse

ZusammenfassungMethoden der künstlichen Intelligenz haben sich in den letzten Jahren zu essenziellen Bestandteilen fast aller Bereiche von Wissenschaft und Technik entwickelt. Dies gilt auch für die Hydrologie: Vielschichtige neuronale Netzwerke – auch bekannt als Modelle des Deep Learning – ermöglichen hier Vorhersagen von Niederschlagsabflussmengen in zuvor unerreichter Präzision.Dieser Beitrag beleuchtet das Potenzial von Deep Learning für wasserwirtschaftliche Anwendungen. Der erste Teil des Artikels zeigt, wie sogenannte Long Short-Term Memory-Netzwerke – eine spezifisch für Zeitreihen entwickelte Methode des Deep Learnings – für die Niederschlags-Abfluss-Modellierung verwendet werden, und wie diese für eine Reihe hydrologischer Probleme bessere Ergebnisse als jedes andere bekannte hydrologische Modell erzielen. Der zweite Teil demonstriert wesentliche Eigenschaften der Long Short-Term Memory-Netzwerke. Zum einen zeigen wir, dass diese Netzwerke beliebige Daten verarbeiten können. Dies erlaubt es, mögliche synergetische Effekte aus unterschiedlichen meteorologischen Datensätzen zu extrahieren und damit die Modellgüte zu verbessern. Zum anderen stellen wir dar, wie relevante hydrologische Prozesse (wie z. B. das Akkumulieren und Schmelzen von Schnee) innerhalb der Modelle abgebildet werden, ohne dass diese spezifisch darauf trainiert wurden.

Background Inhomogeneous lung ventilation and pulmonary air leaks are common adverse effects of mechanical ventilation in preterm infants suffering from respiratory distress syndrome. We hypothesized that shortening of inspiratory times can improve the homogeneity of lung aeration. We assumed that ultrashort inspiratory times lead to a full build‐up of pressure in regions with low compliance while maintaining incomplete filling of regions with high compliance. Methods We connected a two‐compartment model of the lung with different compliances of the two compartments to a neonatal ventilator. Pressures and flow rates were measured separately for each compartment at inspiratory times ranging from 0.1 to 0.8 s and various combinations of tidal volumes. Results An inspiratory time of 0.8 s resulted in near total pressure equalization between the airway pressure and the two compartments, as 97% of the maximum applied peak pressure was reached in the low compliance compartment and 95% in the high compliance compartment. The distribution of the tidal volume was proportional to the compartment compliance. Ultrashort inspiratory times lowered the peak pressure and tidal volume in the high compliance compartment but maintained higher pressure and volume in the low compliance compartment. An inspiratory time of 0.2 s resulted in a peak pressure of 80% of peak airway pressure in the low compliance compartment and in 61% of peak airway pressure in the high compliance compartment (p < .001). Conclusion Ventilation with ultrashort inspiratory times may improve homogeneity of air distribution in inhomogeneous lungs of ventilated preterm or term infants.

ZusammenfassungDas Schätzen von räumlich verteilten Parametern hydrologischer Modelle ist ein bereits lang erforschtes und anspruchsvolles Problem. Parameter-Transferfunktionen, die einen funktionellen Zusammenhang zwischen Modellparametern und geophysikalischen Gebietseigenschaften herstellen, sind eine potenzielle Möglichkeit, Parameter ohne Kalibrierung zu schätzen. Function Space Optimization (FSO) ist eine symbolische Regressionsmethode, die automatisiert Transferfunktionen aus Daten schätzen kann. Sie basiert auf einem textgenerierenden neuronalen Netzwerk, das die Suche nach einer optimalen Funktion in ein kontinuierliches Optimierungsproblem umwandelt.In diesem Beitrag beschreiben wir die Funktionsweise von FSO und geben ein Beispiel der Anwendung mit dem mesoscale Hydrological Model (mHM). Ziel der Anwendung ist die Schätzung zweier Transferfunktionen für die Parameter KSat (gesättigte hydraulische Leitfähigkeit) und FieldCap (Feldkapazität). Dafür verwenden wir Daten 7 großer deutscher Einzugsgebieten über einen Zeitraum von 5 Jahren zum Schätzen der Transferfunktionen und weiterer numerischer Parameter. Die resultierenden Funktionen und Parameter werden ohne weitere Kalibrierung auf 222 Validierungsgebiete über eine Validierungsperiode von 35 Jahren angewendet. Mit der Anwendung in diesen "unbeobachteten" Gebieten können wir die Übertragbarkeit und die zumindest regionale Gültigkeit der Transferfunktionen überprüfen.Die Ergebnisse zeigen, dass bei einer Anwendung in unbeobachteten Gebieten die Modellgüte in einem ähnlichen Wertebereich wie in den Trainingsgebieten liegt und somit weiterhin akzeptabel ist. Die Nash-Sutcliffe Efficiency (NSE) in den Trainingsgebieten über den Validierungszeitraum unterscheidet sich mit einem medianen Wert von 0,73 nicht nennenswert von dem der Validierungsgebiete mit einem medianen NSE von 0,65.Zusammengefasst haben Transferfunktionen das Potenzial, die Vorhersagefähigkeiten, Übertragbarkeit auf andere Gebiete sowie physikalische Interpretierbarkeit bestehender hydrologischer Modelle zu verbessern. Mit FSO wurde zum ersten Mal eine objektive, datengetriebene Methode entwickelt, mit der Transferfunktionen geschätzt werden können.

International audience ; Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross-sectional survey using a semistructured online questionnaire addressing 29 national European milk-banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.

International audience ; Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross-sectional survey using a semistructured online questionnaire addressing 29 national European milk-banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.

International audience ; Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross-sectional survey using a semistructured online questionnaire addressing 29 national European milk-banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.

International audience ; Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross-sectional survey using a semistructured online questionnaire addressing 29 national European milk-banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.

International audience ; Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross-sectional survey using a semistructured online questionnaire addressing 29 national European milk-banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.

Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross-sectional survey using a semistructured online questionnaire addressing 29 national European milk-banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.

Data about the regulatory approaches to donor human milk (DHM) in European countries are lacking. The aim of this study is to describe the various regulations of DHM within European countries, to assess its legislative context and its impact in relation to donor milk banking. We performed a cross‐sectional survey using a semistructured online questionnaire addressing 29 national European milk‐banking representatives from June 2020 to February 2021. Representatives of 26 national DHM services participated in this study. The legal classification and regulatory status of DHM were defined in 9 out of 26 areas of jurisdiction (35%) as either food product (n = 6), product of human origin according to a blood, tissue, cell regulation (n = 2), or medicinal product (n = 1). In the remainder, DHM remains unclassified. Most legislations did not provide a comprehensive framework concerning DHM and costs to cover milk bank operations were rarely reimbursed. In general, the lack of national legislative governance and the actual legislative regulations in place do not support the use of DHM in European countries. National medical guidelines for the use of DHM have been issued in only 11 countries. The current number and distribution of milk banks (n = 239) within participating countries may not provide an equitable access to DHM for eligible infants. These findings could guide stakeholders aiming to establish a regulatory framework for DHM.