No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Towards a Unified Understanding of Data-Driven Support for Emergency Medical Service Logistics
Abstract
Time-critical medical emergencies challenge emergency medical service (EMS) systems worldwide every day. In order to respond to these incidents as soon as possible, EMS logistics' approaches can help locating and dispatching ambulances. Many of these approaches use estimates for the demand as well as the driving, service and turnaround times. In order to determine useful solutions and make informed decisions, reliable forecasts are necessary that take the characteristics and constraints of the planning problems at different levels into account. While many different approaches have been presented and tested in literature, a common understanding is still missing. This paper therefore proposes a taxonomy on EMS forecasting that distinguishes between medical emergencies and patient transports, demand and time intervals in the response process, as well as the three planning levels strategic, tactical and operational. In addition, an illustrative example and a research agenda are presented based on the findings for the taxonomy.
... Given the rapid and successful development and emergence of new DL models and NLP applications (e.g., for argumentation mining [27]), it is vital to understand the similarities and differences among different models. As mentioned in the introduction, taxonomies provide a common way in Information Systems research to summarise the current state of a technology [13,11,28,29,30,31]. ...
Despite a large number of available techniques around Deep Learning in Natural Language Processing (NLP), no holistic framework exists which supports researchers and practitioners to organise knowledge when designing, comparing and evaluating NLP applications. This paper addresses this lack of a holistic framework by developing a taxonomy for Deep Learning in Natural Language Processing. Based on a systematic literature review as proposed by Webster and Watson [1] and vom Brocke et al. [2] and the iterative taxonomy development process of Nickerson et al. [3] we derived five novel dimensions and 38 characteristics based on a sample of 205 papers. Our research suggests, that a Deep Learning NLP approach can be distinguished by five dimensions which were partly derived from the CRISP-DM methodology: application understanding, data preparation, modeling, learning technique and evaluation. We, therefore, hope to provide guidance and support for researchers and practitioners when using Deep Learning for NLP to design, compare and evaluate NLP applications.
... Arrival rates of patients or emergency calls, together with their necessary treatments and treatment times are the main uncertainties for ED managers. Different forecasting techniques including machine learning have been proposed to estimate the parameters (Afilal et al., 2016;Menke et al., 2014;Reuter-Oppermann & Wolff, 2020;Xu et al., 2013). Information systems: The design of DSS is a well-established area in IS research. ...
Emergency services in countries like Germany are defined by a complex service network and a multitude of planning problems and necessary decisions. In case of an emergency, emergency medical service (EMS) providers send an ambulance to the patient, perform first treatment and then transport most patients to a hospital. Often, more than one hospital is available. Therefore, an EMS provider must decide to which hospital, more specifically, to which emergency department (ED), the patient is taken, without knowing the current workload of the EDs. Applying the Design Science Research paradigm, we propose a concept for a decision support system that connects the EMS provider with the EDs and allows for an informed decision making that not only considers the patient’s, but also the EMS provider’s and the EDs’ objectives. Therefore, by applying system thinking, the decision support system enables customer-centric care. A first prototype of the system in form of a mock-up was developed and evaluated with emergency services providers who would mainly use the system. A research outline for fully designing and developing the system is presented.
In addition to fires and various other emergencies, the Fire Department of New York City (FDNY) responds to medical emergencies throughout New York City. When a patient requires hospital care, the FDNY uses a hospital suggestion pattern to recommend the closest hospital that can appropriately treat the patient, given the geographic location of the incident. In a city like New York with a high density of hospitals, determining the closest hospital (in terms of travel time) is not trivial. This paper presents the results of a collaboration between the FDNY and Columbia University which integrates data analytics in the derivation, implementation and monitoring of this hospital suggestion pattern. Both telematics data of city-owned vehicles and historical ambulance travel times to hospitals are utilized to estimate the order of closest hospitals for a given incident location. Implementation constraints (coarse spatial and temporal resolutions of the pattern) and sparsity of the historical ambulance dataset generate large aleatory and epistemic uncertainties that are accounted for using a probabilistic data fusion procedure. The result of this analysis, hospital suggestion pattern N, was implemented in December 2020. A statistical analysis shows an overall decrease in travel times to hospitals as an effect of this pattern.
Melanie Reuter-Oppermann, Luisa Pumplun, Helena Müller und Peter Buxmann untersuchen in ihrem Aufsatz „Künstliche Intelligenz im Rettungsdienst – Wege in die Zukunft“ zum einen die Möglichkeiten, die der Einsatz von KI im Rettungsdienst eröffnen könnte . Sie arbeiten aber ebenfalls sehr ausführlich heraus, dass dabei zahlreiche Anforderungen zu erfüllen sind, die sich sowohl aus der spezifischen Anwendungssituation ergeben als auch aus normativen Überlegungen . Damit zeigen die Autor*innen eine Verwendungsweise von KI auf, die in der einschlägigen Literatur nicht so oft in den Blick genommen wird, und lassen zudem erneut sichtbar werden, dass KI einem komplexen Geflecht aus Anforderungen gerecht werden muss
Der Einsatz von künstlicher Intelligenz im Gesundheitsbereich verspricht besonders großen Nutzen durch eine bessere Versorgung sowie effizientere Abläufe und bietet damit letztlich auch ökonomische Vorteile. Dem stehen unter anderem Befürchtungen entgegen, dass sich durch den Einsatz von künstlicher Intelligenz das Arzt-Patienten-Verhältnis verändern könnte, Arbeitsplätze gefährdet seien oder die Ökonomisierung des Gesundheitswesens einen weiteren Schub erfahren könnte. Zuweilen wird die Debatte um diese Technologie, zumal in der Öffentlichkeit, emotional und fern sachlicher Argumente geführt. Die Autorinnen und Autoren untersuchen die Geschichte des KI-Einsatzes in der Medizin, deren öffentliche Wahrnehmung, Governance der KI, die Möglichkeiten und Grenzen der Technik sowie Einsatzgebiete, die bisher noch nicht oder nur wenig im Fokus der Aufmerksamkeit waren. Dabei erweist sich die KI als leistungsfähiges Werkzeug, das zahlreiche ethische und soziale Fragen aufwirft, die bei der Einführung anderer Technologien bereits gestellt wurden; allerdings gibt es auch neue Herausforderungen, denen sich Professionen, Politik und Gesellschaft stellen müssen.
ResearchGate has not been able to resolve any references for this publication.