مدل سازی جریان کاری مدیریت هشدار در مراقبت های بهداشتی با توجه به چارچوب IHE با استفاده از شبکه های پتری رنگی

Ensuring patient safety in medical device networks by managing alarms and related clinical data is a life-critical issue. The Integrating the Healthcare Enterprise (IHE) initiative emerged to discuss and solve the interoperability and integration problems among medical information systems, vendors and users in order to improve patient care and healthcare system dependability. This paper models and analyzes the IHE Alarm Communication Management by using the Unified Modelling Language and Coloured Timed Petri nets. Aiming at generality, the model does not refer to a specific healthcare context but it is based on a general scheme where the message transactions are integrated with the nurse responses. In order to show the potentialities of the model, a real case study is simulated and different scenarios with different levels of workload are analyzed. The results illustrate that the model is able to provide support for structured and comprehensive analysis of the healthcare system management.

Many hospital enterprises are rapidly moving to extensive integration of both patient care process and information system, thanks to the advances in information technology and related fields. In addition, major medical device companies have designed patient monitoring systems that automatically reconfigure themselves for different hospital applications. Moreover, managing alarms of medical devices and related clinical data is a life-critical issue in order to avoid exposing patients to serious injuries (Gehlot and Sloane, 2006). Consequently, an emergent and challenging subject inherent to the healthcare system dependability is the management of the medical alarm devices integrated with the hospital workflow planning. Integrating the Healthcare Enterprise (IHE) (IHE.net Home. Available at http://www.ihe.net/) is an initiative that has the goal of establishing a reliable integrated environment for medical devices and clinical information systems. Indeed, IHE identifies the key transactions required to automate processes, reduce errors and increase efficiency. The IHE is organized across a number of clinical and operational domains that produce a set of technical framework documents. In particular, the domain of the IHE Patient Care Device (IHE PCD) (IHE Patient Care Device Technical Committee, 2008) addresses the integration of medical devices with healthcare computer systems and intends to improve the flow of information between the point-of-care and the electronic healthcare record (Ettinger et al., 2009; Carr and Moore, 2003). Moreover, the IHE PCD technical framework is composed by a set of profiles that address specific integration scenarios including at least one regulated medical device. In particular, in the IHE PCD domain, the Alarm Communication Management (ACM) profile (IHE Patient Care Device (PCD) Technical Committee, 2008 and Booch et al., 2005) establishes interoperability between systems of different manufacturers, and results in a communication standard for alarm messages. It defines the communication of alarms from alarm source systems to alarm manager systems and from alarm manager systems to alarm archiver systems. Hence, ACM “is meant to improve clinical efficiency by using technology to deliver the right alarms, with the right priority, to the right individuals via devices and transactions with the right content” (IHE Patient Care Device Technical Committee, 2008). The objective of this paper is a better understanding the IHE PCD ACM profile for more effective and integrated implementation. The related literature deals with the problem of analyzing and modelling the IHE scheduled workflow integration profile (Hussein and Winter, 2008) and testing the standards in e-health domain (Vega et al., 2010) in order to facilitate the adoption of the IHE concepts and to study the interoperability scenarios. In particular, Ettinger et al. (2009) developed an extensible open source toolkit based on the ACM profile and interfaced the patient monitoring equipments by means of suitable export interfaces. Moreover, some contributions deal with the workflow management (Nie et al., 2009 and Dallien et al., 2008), and translate the IHE integration profile into Petri Net (PN) workflow models to bridge the gap between traditional applications and process-oriented healthcare solutions. Sloane and Gehlot (2005) use Coloured Timed Petri Nets (CTPNs) to model a medical device alarm management system. However, the model leads just to describe the system behaviour and does not accurately follow the corresponding ACM profile. The aim of the paper is analyzing, modelling and simulating the healthcare information system related to the ACM profile integrated with the wireless medical device network and the heathcare operators. The modelling procedure follows two steps. First, the ACM profile is described and analyzed by means of the Unified Modelling Language (UML) (Miles and Hamilton, 2006), a graphic and textual modelling language intended to understand and describe systems from various viewpoints. Second, the behaviour of the ACM profile inside healthcare enterprises is modelled by regarding it as a Discrete Event System (DES), whose dynamics depends on the interaction of discrete events, such as the sending/reception of messages and the actions of the involved actors (e.g., the nurses). Among the available DES models, CTPNs (Jensen, 1992) are selected as a graphical and mathematical technique to describe concurrency, conflicts and synchronization. In particular, CTPNs are an enhancement of PNs in which the tokens have “colour” or types, and time is introduced to evaluate performances. Moreover, CTPNs are able to combine the strengths of PN with the strengths of a high-level programming language, and are suitable to model healthcare systems (Jørgensen et al., 2008). Indeed, the structure of the CTPN provides the primitives for process interaction, while the programming language provides the primitives for the definition of data types and the manipulations of data values. In this paper, the proposed CTPN describes in a unified modelling framework the heterogeneous system, comprising medical devices, information system and human resources. Hence, the paper describes the alarm management system exploiting the two main peculiarities of the CTPNs: the graphical characteristic of the CTPN, which enables the description of the system, easy to build and to verify; the mathematical and software translation of the CTPN, which allows simulating the system and evaluating the performances. In order to show the effectiveness of the presented modelling technique, a real case study is analyzed and simulated considering a simple IHE PCD ACM profile. The obtained results show how the CTPN model and the performance analysis help in the verification and validation of such a profile. The paper is organized as follows. Section 2 illustrates the Alarm Communication Management profile, and Section 3 describes the corresponding CTPN representation. Moreover, Section 4 specifies the simulation experiments, and reports the simulation results. Finally, Section 5 discusses the conclusions.

The analysis of the IHE integration profiles can provide a useful support for verifying and assessing healthcare processes, with the related information, workflow and workload systems. Aiming at studying and evaluating healthcare processes in which a large number of different medical devices is used, this paper takes into consideration the most recent IHE Patient Care Device domain and in particular the Alarm Communication Management (ACM) system. To this aim, we employ the UML communication and sequence diagrams to describe the ACM profile transactions. Moreover, the ACM profile integrated with the nurses’ actions is modelled as a discrete event system and its dynamics is described and simulated in a Coloured Timed Petri Nets (CTPNs) framework. A real case study analyzing a hospital department alarm system shows that CTPNs are an effective tool to predict and evaluate life-threatening data delays. Indeed, the calculated performance indices can be used to depict various configurations of patient devices, wireless networks, information systems, and human resources along with accurate priorities. Hence, the analysis and evaluation of the impact of the ACM profile provide a basic approach to deal with the healthcare system dependability. Moreover, the simplicity of the resulting CTPN describing the system shows how the used framework has high potentiality for describing large and complex healthcare systems. Finally, in future research the presented modelling approach will be applied to analyze complex networks, including the detailed model of the clinical staff workflows, in order to handle and evaluate the clinical risks properly.