Lage und Zukunft des wissenschaftlichen Nachwuchses Eine Stellungnahme des Beirats des Wissenschaftlichen Nachwuchses (WiN) der Gesellschaft für Informatik (GI e. V.
Recently, health insurance claims have regained the attention of the scientific community as a source of real-world evidence in health care research and quality improvement. To date, very few studies are available which investigate the validity of health insurance claims; these may be affected by bias from several sources, such as possible upcoding of co-morbidities and complications for reimbursement advantages. The IDOMENEO study investigates the inpatient treatment of peripheral arterial disease (PAD) comprehensively using various data sources with a consortium involving experts from health care research and data privacy, a large health insurance fund, biostatisticians, jurists, and computer scientists. Prospective registry data were collected from 30–40 vascular centres in Germany using the GermanVasc registry. In addition, health insurance claims data were prospectively collected from BARMER, the second largest health insurance fund in Germany. The consortium is currently developing a data privacy compliant method of health insurance claims data validation, the methodological foundations of which are described here.
A bench scale flue gas desulphurization spray dry scrubbing unit was employed to study the effect of fly ash on the removal of SO,. The equipment consisted of a spray dryer with an ultrasonic nozzle for atomization and a pulse jet baghouse. The flue gas rate was 1500 I,/h (dry gas). Four fly ashes, originating from four different countries were investigated. The alkalinity and reactivity of the fly ashes were determined in a pH-stat equipment. Pure fly ash removed SO, in both the spray dryer and in the baghouse. An increase of humidity divided the fly ashes into two groups. The high calcium fly ash gave a considerably higher SO, removal than the medium and low calcium fly ashes which showed similar SO, removals. Fly ash did not enhance the removal of SO, when added to a lime slurry because lime suppresses the dissolution of the alkali in the fly ashes. The pressure drop build-up in the fabric filter showed a strong dependence on material properties.
ZusammenfassungVerfahren des maschinellen Lernens (ML) beruhen auf dem Prinzip, dass ein Algorithmus Muster und statistische Zusammenhänge in Datensätzen erkennt, diese in einem Modell abbildet und das Modell anschließend auf andere Datensätze anwenden kann. Neben den großen Chancen, die maschinelle Lernverfahren mit sich bringen, birgt diese Technologie allerdings auch Risiken für die Privatsphäre, die in diesem Artikel in Form von Privatsphäreangriffen beleuchtet werden.Angriffe wie Model Inversion zielen auf oftmals sensible Informationen ab, die sich während der Trainingsphase eines ML-Algorithmus ungewollt in einem Modell etabliert haben. Wenn Trainingsdaten Personenbezug aufweisen, insbesondere wenn es sich etwa um vertrauliche medizinische Daten handelt, kann dies problematisch für Betroffene sein.Demgegenüber stehen Techniken des privatsphärefreundlichen maschinellen Lernens wie Federated Learning, die eine Risikominimierung für ein breites Spektrum an Privatsphäreverletzungen ermöglichen. Ausgewählte Techniken aus diesem Bereich werden in diesem Artikel ausführlich dargestellt.Dies ist der zweite Teil einer zweiteiligen Artikelserie, deren Auftakt unter dem Titel Grundlagen und Verfahren bereits in der letzten Ausgabe des Informatik Spektrums erschienen ist.
ZusammenfassungMaschinelle Lernverfahren finden seit einigen Jahren in immer mehr Bereichen vielfältige Anwendung, wodurch die Relevanz der dabei verwendeten Techniken deutlich wird. Unter dem Begriff des maschinellen Lernens (ML, oft auch „künstliche Intelligenz“) existieren zahlreiche Algorithmen, die unterschiedliche Komplexität und verschiedene Eigenschaften mit sich bringen. Für das Training dieser Algorithmen sind meist große Mengen an Daten notwendig. Insbesondere bei der Verwendung von personenbezogenen Daten stellen sich hierbei Fragen rund um den Datenschutz und die Privatsphäre von Betroffenen.Dies ist der erste Teil eines zweiteiligen Artikels zum Thema privatsphärefreundliches ML. Dieser erste Teil bietet einen leicht verständlichen Einstieg in das Thema des ML und geht dabei auf die wichtigsten Grundbegriffe ein. Außerdem werden einige der meistverwendeten ML-Verfahren, wie Entscheidungsbäume und neuronale Netze, vorgestellt. Im zweiten Teil, der in der kommenden Ausgabe des Informatik Spektrums erscheint, werden Privatsphäreangriffe und datenschutzfördernde Maßnahmen im Kontext von ML behandelt.
BackgroundMultidisciplinary rehabilitation is recommended to reduce sickness absence and disability in patients with subacute or chronic LBP. This RCT aimed to investigate whether a 12-week coordinated work oriented multidisciplinary rehabilitation intervention was effective on return to work and number of days off work during one-year follow-up when compared to usual care. MethodsThis study is a randomized controlled trial comparing a 12-week multidisciplinary vocational rehabilitation program with usual treatment. 770 patients with LBP, who were sick-listed, or at risk of being sick-listed were included in the study. The primary outcome was number of days off work due to LBP. The secondary outcomes were disability, health-related quality of life, pain, psychological distress and fear avoidance behavior.Data were collected at baseline, at the end of treatment, and at 6- and 12-months follow-up. Analyses were carried out according to the “intention-to-treat” principles.Results A significant decrease in the number of days off work was found in both groups at the end of treatment and at 6- and 12-months follow-up. Additionally, disability, pain, health related quality of life, psychological distress, and fear avoidance beliefs improved in both groups. No statistically significant differences were found between the groups on any of the outcomes.Conclusions The coordinated multidisciplinary intervention had no additional effect on sickness absence, disability, pain, or health related quality of life as compared with that of usual care. Trial registrationThis study was registered in ClinicalTrials.gov (registration ID: NCT01690234). The study was approved by The Danish Regional Ethics Committee (file no: H-C-2008-112) as well as registered at and approved by the Danish Data Protection Agency.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.