Design of meta-DSS to support the decision-making process: supporting activities on a clinical pathway in the emergency department

在临床决策中，有许多情况下，最终结果是多个相互交织的任务的结果。在急诊科（ED）管理病人是我的想法。为了促进患者管理和遵循循证医学，医院引入临床路径作为工具，以确保患者得到最适当的护理。这些途径是从临床实践指南发展而来的，它们概述了一系列暂时的临床评估和干预措施。在这个意义上，路径是一个决策过程的规定性模型，其中个别任务对应于不同的、依赖时间的活动。虽然创建临床决策支持系统是为了一次支持一项活动，但没有一种设计方法可以为一组活动提供全面的支持。本提案中描述的研究目的是开发一种设计元决策支持系统的方法，该系统将不同的支持需求和能力结合在一起，从而支持整个决策过程。我们将使用基于组织的多智能体系统工程（O-MaSE）方法创建一个元决策支持系统（meta-DSS）设计，该方法允许由于智能体之间的相互作用而对问题结构进行建模更改（患者管理的常见情况是，根据最新的评估和先前的行动改变治疗）。我们将把包含路径的独立代理（实体）的单个任务视为独立的代理（实体），每个代理（实体）都有自己的需求和功能。我们将对这些代理进行建模，将它们作为“服务”组合在一起以实现一个共同的目标。我们将使用面向服务的体系结构（SOA）范例来连接来自临床路径的随需应变的松散耦合任务，以及独立临床DSS提供的支持功能。SOA的实现将涉及到任务（服务请求者）的识别和描述，这些任务（服务请求者）调用与某些特定临床DSS（服务提供者）相关的个人支持需求。我们正计划评估meta-DSS设计在儿科头部创伤和哮喘的ED管理中的临床应用能力。

In clinical decision-making there are many situations where a final outcome is a result of multiple intertwined tasks. Managing a patient in the Emergency Department (ED) comes to mind. In order to facilitate patient management and to follow evidence-based medicine, hospitals introduce clinical pathways as tools to ensure that patients receive the most appropriate care. These pathways are developed from clinical practice guidelines and they outline a sequence of temporal clinical evaluations and interventions. In that sense a pathway is a prescriptive model of a decision-making process with individual tasks corresponding to different, time-dependent activities. While there are clinical DSS created to support one activity at a time, there is no design methodology that provides comprehensive support for a set of activities. The purpose of the research described in this proposal is to develop a methodology for designing a meta-DSS that brings together different support requirements and capabilities so that entire decision-making process is supported. We will create a meta-DSS design using the Organizational-based Multiagent System Engineering (O-MaSE) methodology that allows for modeling changes to the problem structure as a result of interactions between agents (common situation for patient management where treatment changes depending on latest assessment and previous actions). We will be looking at individual tasks comprising a pathway as independent agents (entities), each with its own requirements and capabilities. We will model these agents, as "services" that should be brought together in order to accomplish a common goal. We will use the Service-Oriented Architecture (SOA) paradigm for connecting on demand loosely coupled tasks from a clinical pathway as well as support capabilities provided by stand-alone clinical DSS. The implementation of the SOA will involve identification and description of tasks (service requestors) that invoke individual support requirements associated with some specific clinical DSS (service providers). We are planning to evaluate the capabilities of a meta-DSS design using clinical pathways for ED management of pediatric head trauma and asthma.