Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Authoring, deploying, and managing dynamic Virtual Patients in Virtual Clinical Environments

Authoring, deploying, and managing dynamic Virtual Patients in Virtual Clinical Environments Following their introduction at the beginning of the 21st century, interactive or dynamic Virtual Patients are beginning to be used more widely in clinical education. They can be seen as being at the end of a continuum of simulation technical complexity, having been earlier developed on a wide range of "media": human actors, paper, video, physical mannequins, etc. This paper focuses on the current emergent more complex Virtual Patients in three-dimensional (3D) immersive clinical environments. In these environments, in silico 3D patient avatars interact directly in response to virtual clinical interventions undertaken by avatars, each of which is controlled by one or more users. The paper explores the issues of authoring, deploying, and managing these real-time, dynamic Virtual Patients using as an example the immersive clinical environment CliniSpace. As clinician-accessible Virtual Patient authoring is now becoming available in immersive clinical environments, so these wider clinical and managerial non-technical issues are coming rapidly to the fore. Keywords: clinical simulation; clinical virtual worlds; interactive Virtual Patients; Virtual Patients; virtual worlds. Introduction The earliest dynamic avatar noted in Sean P. Egan's "History of Avatars" review was Ananova, "designed to *Corresponding author: Dick Davies, Exec. Producer for Ambient, Ambient Performance, Suite 336, 43 Bedford Street London WC2E 9HA, UK, E-mail: dick.davies@ambientperformance.com LeRoy Heinrichs: Department of Obstetrics and Gynecology, Emeritus, Stanford University School of Medicine, Stanford, CA, USA Parvati Dev: Innovation in Learning, Stanford, CA, USA deliver the latest news over the Internet and on mobile devices, the 28-year-old, green-haired, British, 3D avatar had a full range of facial expressions and could speak multiple languages. She was designed to be attractive, with a global appeal, and trustworthy and believable, based on the principle that face-to-face communication is one of the oldest and most trusted forms of obtaining information, a theory still adhered today by companies creating or deploying avatars" [1]. Since Ananova's appearance in 2000, many interactive avatars have been developed on a multitude of online platforms representing a variety of sectors in business, art, marketing, gaming, and education, including medical education. Applications in clinical medicine include VitalSims, CliniSpace, The Anatomy of Care, Virtual Heroes, OLIVE, and others referenced in an excellent, illustrated, online Medscape review by L. Stokowski, R.N., in 2013 focused on nursing [2]. Each of these platforms enables avatars to demonstrate basic physiologic responses that teach users about typical responses to different clinical situations. In these computer-generated environments, people enter simulated worlds in real time from their PC or Pad to interact and collaborate with others represented by avatars in real time and make real-time decisions as they would in the "real world." When virtual world technology is implemented in the medical context, it can replicate most clinical environments: the emergency department, the ward, or a chaotic pre-hospital experience. These virtual worlds when deployed in a clinical context have been termed "Virtual Clinical Worlds" [3]. Combining Virtual Patient (VP) technology with Virtual Clinical Worlds offers, in principle, a powerful platform for developing and deploying realistic clinical experiences. The barrier until recently, however, has been that for clinicians to deploy these kinds of integrated experiences, substantial input from IT professionals was necessary. The arrival of Commercial-off-the-Shelf (COTS) suppliers is changing the game. Why do COTS change the game? The emergence of a COTS-supplied Virtual Clinical Environment (VCE)/VP integrated platforms now means that questions that were 80Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments previously distant, due the fact that these technologies were emergent and so not ready for widespread deployment, are now coming to the fore as these technologies allow clinical professionals to configure and deploy them rather than IT professionals. Possibly, we could now argue for a "tipping point" in terms of clinical acceptance and possibly uptake. The issues therefore become increasingly clinical and non-IT technical, i.e., how to introduce and manage the use of these simulation technologies on a day-to-day basis for mainstream clinical use. Further, these virtualisation technologies offer some unique features that give rise to some wider questions around the construction of virtual experiences for professionals. One hope that does come through strongly in the literature is that, "Future progress must be made to develop rich, guided authoring tools that will allow medical educators, not just computer scientists, to develop virtual patients..." [4]. This wish is the focus of this paper. This paper will ­ Provide a briefing background on VPs integrated into VCEs including some of their possible applications in clinical settings; ­ Show that issues that arise from their use in the mainstream are in the main now clinical and non-technical rather than IT technical using the CliniSpace VCE as the case study; ­ Explore the wider non-technical issues around the design and deployment of these technologies in professional contexts. use the rules of a game; simulations attempt to model a system in a manner that is consistent with reality [6]. It is important that this conceptual distinction is held in mind for this paper. Dynamic (or interactive) VPs (DVPs) and VCEs are subgenres of simulation that arise from very distinct and different heritages. VPs very much arise from professional clinical practice and have only recently embraced computer technologies. Virtual Clinical Worlds, on the other hand, have arisen from the repurposing and exploitation of existing video gaming technology approaches in the 2000s to develop realistic simulated clinical environments [7]. These two strands of virtualisation whilst originally from different contexts are now being brought together due to the fact that computing power and fast networks are democratising access to these powerful tools. What are virtual worlds? Virtual worlds are now an accepted technology, deployed widely in consumer spaces and with which most of the millennial generation, i.e., those born post-1980, are familiar and most are comfortable [8]. Known also as immersive environments, virtual worlds can be defined as a "synchronous, persistent network of people, represented by avatars, facilitated by networked computers" [9]. There are a wide range of variations, but virtual worlds are computer simulations offering some or all of the following [10]: ­ Three-dimensional (3D) spaces; ­ People represented by avatars; ­ Objects in world are persistent and maybe interacted with, e.g., moveable chairs, drivable vehicles; ­ Communication is usually in real time via voice, text, and gesture. Dynamic Virtual Patients and Virtual Clinical Environments: into the mainstream? According to Sauvé et al. [5], games and simulations are distinctive concepts. A game is a fictitious or artificial situation governed by rules that structure their actions in view of an objective, which is to win or to overcome an obstacle. They are integrated into an educational context when the learning objectives are associated formally to the content and the game enhances learning in the cognitive, affective, and/or psychomotor domains. In that case, they are often termed "serious games." On the contrary, a simulation does not necessarily involve competition, and the users are not trying to win. Many educational simulations, unlike games, can function without human intervention. Games suspend the rules of reality in order to Immersive Clinical Environments or Virtual Clinical Worlds Where virtual worlds are used specifically for clinical purposes, they could be termed Immersive Clinical Environments or Virtual Clinical Worlds. Heinrichs et al. [3] went on to propose that in clinical practice, virtual worlds were deployed because they offer the following attributes: Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments81 ­ ­ ­ ­ ­ ­ Presence, Immersion, i.e., engagement, Team-based activities, Real workplace settings, Safe "play spaces," Relatively low cost (compared to custom health-care game development), within a clinical or, in certain cases where the focus is still on the clinical, pre-hospital settings. The paper now looks at VPs before moving on to examine how VPs can work within virtual worlds. Virtual Patients The genesis of VPs is the "human standardised patient" developed by Dr. Howard Burrows in 1963. Whilst computer simulations of patients started in the 1960s, it is only recently with the widespread uptake of computers in medical education that VPs, as they are now understood, have expanded. The term "Virtual Patients," however, is a portmanteau term for a very wide range of distinct approaches. "Such approaches include case presentations, interactive patient scenarios, virtual patient games, human standardised patients, high fidelity software simulations, high fidelity manikins and virtual conversational agents" [4]. The conceptualisation and building of VPs had its beginning in 1911 with a life-size "Mrs. Chase doll," an early mannequin specifically developed for nurses to practice their clinical skills [11]. Since then, an entire industry has developed responding to the market for physical mannequins. As computer technology and online learning has emerged, building avatars as VPs is offering to replace physical mannequins as the learning objects. A major difference only mentioned here are the actions of caregiver avatars that are able to move about in a 3D space, but appropriate patient avatar actions would be focused on movements onto and off a chair, couch or bed, or into or out of a wheelchair, or rotation side-to-side or sitting up in bed. These are actions that mannequins are unable to perform, except as users position them manually. In contrast, virtual avatars can be directed to run or walk and change direction as the keyboard's pointer keys are activated. Torso twisting or flexion is programmed to be displayed upon compressing a () key, or another chosen by the coder. Arm, head, and facial movements are guided in coordination with spoken words. For example, auscultation of the chest or heart integrates the appropriate sounds with a stethoscope positioned and held by the user. Abnormal sounds are encoded as appropriate for the scenario. The mention of auscultation sounds leads to focusing on the intrinsic physiologic system that resides in all living organisms, and in realistic VPs. The current generation of VPs will usually be designed to encompass either a basic or perhaps a more sophisticated system, and include physiologic sounds of the cardiovascular and gastrointestinal systems. Such sounds can be programmed to represent either normal or pathological conditions. The physiology model selected may be extremely sophisticated based on established physiological principles such as those incorporated in HuMod [12] or upon reports of big data derived from clinical practice/ experience. The latter "pragmatic model" continues to be selected for CliniSpace's VPs [13]. The fundamental parameter or "driver" for the pragmatic model(s) is blood volume upon which an individual's vital signs are based. Representations of population-based data reflect the height, weight, and blood volume differences among men vs. women and the body mass index (BMI) of healthy people and of those overweight and obese individuals [14]. Similarly, the increasing blood volume of pregnant women reflects the changing BMI of that physiological condition [15]. In the CliniSpace Virtualsimcenter, the patient monitor displays the vital signs of avatars based on this fundamental principle. One thus recognizes the core value of using blood volume in the architecture of VPs that can be designed to accurately reflect and teach this feature of human physiology. The term "Virtual Patients," however, is a portmanteau term for a range of distinct approaches. "Such approaches include case presentations, interactive patient scenarios, Virtual Patient games, human standardised patients, high fidelity software simulations, high fidelity manikins and virtual conversational agents" [4]. Given that the VPs developed in CliniSpace cover a number of these approaches, then the term "dynamic Virtual Patient" is proposed. The characteristics of the DVP are as follows: ­ 3D clinical environment, ­ A human physiology engine, ­ 3D avatars, ­ Data displays, ­ Medical procedure capability, ­ Reporting or assessment capabilities. In CliniSpace, each DVP (DynaPatient) has an underlying pathophysiological model consisting of 82Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments 22 y/o pregnant female, auto injury/traumatic hemorrhage, EOIS=5; GASHD 1.0 250 TR-005-traumatic hemorrhagic shock-untreated 200 Response of model Dysrrhythmias, bradycardia, or PEA, (pulseless electrical activity) PR SaO2 BP-S RR 50 11 min to death 0 8 9 7 Time, min CliniSpaceTM Figure 1:Graphical representation of a pathophysiological model for a patient with traumatic haemorrhage without intervention. (Copyright IIL.) ­ ­ ­ ­ ­ ­ ­ A set of states (healthy, deteriorating, fibrillation, sepsis, severe sepsis, shock, asystole, death). Mathematical equations that define the smooth changes within that state. Conditions that define transitions between states. A physiology model that interacts within the environment. Each model has a "driver" or independent variable. Some of the variables are "observable," for example, as the vital signs appear on a monitor. The AI of the VP is linked to the AI of the environment. Each object in the world has its own rules: e.g., administering fluids at the IV stand temporarily decreases the heart rate and increases the blood pressure. An important lesson to be learnt is that acute blood loss must be stopped before administration of fluids will become helpful (see Figure 1). Authoring, deploying, and managing Virtual Clinical Experiences In the introduction to this paper, it is claimed that these new VCE/DVP integrated platforms now allow clinical professionals to configure and deploy them without the requirement for IT professionals. The consequences of the arrival of these modifiable platforms is that the issues are now becoming clinical and non-IT technical, i.e., how to introduce and manage the use of these simulation technologies on a day-to-day basis for mainstream clinical use with clinician educators. This section investigates, using CliniSpace as a case study example, how to author and deploy/manage these virtual simulation technologies. The focus of this paper is on DVPs embedded in Virtual Clinical Worlds, and what is clear is that at the centre of these new VCEs is the underlying human physiology engine. Poulton [16] has argued that, "It is now possible to consider the extension of the current relatively lightweight Virtual Patient into a truly interactive patient simulation, an "e-human" or "digital avatar." At this stage, the simulation takes on new capabilities, offering authentic patient management, clinical and communication skills training; the potential capability to mimic the health or disease of any citizen." This paper explores these emergent "new capabilities" for simulation. Authoring Once the overarching structure and plan for a clinical virtual world is established, one must consider "What type of Virtual Patient(s) would one like to design and develop? The answer to this question relates directly to the teaching/learning objective(s). Choosing an avatars' appearance, whether youthful or aged, male or female, thin or obese, etc., is guided by that objective. For example, one would be unlikely to select a young thinly built individual to exhibit hypertension, but this clinical condition would not be a surprising finding in a young obese avatar. In contrast, both young and older, thin and Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments83 Figure 2:Students in CliniSpace cooperating in treating a VP. (Copyright IIL and Charles Drew University.) obese individuals are subject to systemic infections that may present with flushed appearances or skin rashes [17]. In a study among 60 students at Charles Drew University School of Medicine, an affiliate of the University of California, Los Angeles, a large majority, 90%, reported ease of use, and 70% reported effective learning among team members about antibiotic overdose because the avatars in an infection scenario developed a skin rash unique to the excessive drugs use. The virtual rash subsided and the skin colour of the avatar returned to normal after the excessive rate of administration was corrected. This example illustrates the dynamic and innovative design of CliniSpace's DynaPatientsTM as learning objects (Figure 2). Authoring of VPs is an "art" in which experience, reflection, and learning through the opinions of users supports engagement, adoption, and better learning scores. A Carnegie Mellon University webpage [18] offers a cogent and relevant discussion of learning objectives. Objectives, assessments, and instructional strategies are key elements in authoring. As one uses the intrinsic CliniSpace, or any other authoring tool, the most appropriate directive is, "At the end of this to-be-authored scenario, what do I want/expect the learner to be able to learn and/ or do?" The answer to this seemingly simple question must consider the learner's prior knowledge and experience and comfort level with computers, online e-learning, video games play, and adoption of new methods and technologies. A common strategy is starting simply, and increasing complexity as one proceeds, thus allowing for multiple entry points for users. This design permits a progressive scale for entry into the set of scenarios, or entry at selected points. This feature enables instructors to use the scenarios for testing. Most [19, 20] of the few VP models available for deployment in virtual environments are runtime, i.e., they run from start to finish and cannot be started or stopped, or even most cases parameterised simply. Two types of VP designs can be distinguished: a "narrative" or passive structure and a "problem-solving" or active structure. In the narrative/passive cases, the simulation represents a single medical state, often in considerable detail, and with relevant graphics, audio, and visual media displaying the patient's medical condition. Fewer simulations support the evolution of the "problem solving"/active patient's state, both with and without medical intervention. In the problem-solving/active model, one specifies both gradual changes in physiologic variables as well as a number of discrete important "states," with the patient moving from state to state based on the VP's condition and on the actions taken by the learner. From this the question arises as to why "active" should be preferred to "passive" VPs? Two reasons are advanced. First, "passive" VPs are experienced as pale imitations of real-world patients so medical professionals find it difficult to take them seriously and, second, they can be "learnt" and then "gamed" by the end user. To expand 84Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments on this, the problem with passive models, in addition to the highlighted poor end-user experience already mentioned above, is that most professional educators are in organisational contexts that have distinctive clinical and other associated procedures that are followed for specific clinical situations. They, therefore, need the clinical environment and VPs to be customisable in order to be able to train for local requirements. Offering VP models that "do not quite fit with our procedures" undermines credibility. A further issue is "gaming the game" where simple runtime models can be learnt and then "gamed" by the user over a period of time. Their passivity has a repetitive and thus predictable nature and performance. They are then viewed as not sufficiently realistic and dismissed as "simply games." Once played and their properties experienced, going back again and again is not compelling. In response to these issues, the approach taken in CliniSpace has been to build simple-to-use standalone authoring tools and enable the running VP model to be adjusted by the scenario facilitator in real time. In practice, this first enables the scenario controller to rapidly "tweak" the scenario prior to runtime, so changing various elements of the scenario, and second, by adjusting the running VP model in real time, the scenario controller is able to alter the original course of the scenario in response to real-time activities and happenings in the immersive environment. The sequence, severity, and resolution of the activities and happenings can be built dynamically into the scenarios [14]. CliniSpace offers authoring of the clinical environment and the in silico VP via a stand-alone set-up and authoring tool. This tool ­ ­ Parameterises and sets triggers in the default VP pathophysiological model in the CliniSpace VCE, Sets up various physical aspects of the CliniSpace environment, e.g., rooms, equipment, dialogue, imaging graphics, etc., and then loads them via a simple file upload into the live single or multiuser CliniSpace environment. Each CliniSpace Virtual Clinical Experience can therefore be individualised for the end user(s) by the author/facilitator to link to a specific educational requirement. Headers for each of 12 tabs of the "CliniSpace Authoring Tool" are Learning Objectives, Patient Introduction, General Observations, Patient History, Physical Exam, Differential Diagnosis, Recommended Case Protocol, Vital Signs, Imaging, Labs, Medications, and Intravenous Options. Each of these is represented in a mobile, interactive, electronic medical record computer sited nearby each patient's bed. The page for each topic has key words listed for which relevant and unique "answers" can be introduced. These pages are further organised as templates to enable authors to create six consecutive "states" for each VP over whatever time period is selected. These features are designed to facilitate efficient authoring. As an example (Figure 3), the Patient State interface sets up the parameters to the VP's initial clinical state. Further interfaces set up the patient hospital room with the objects to be visible at the beginning of a case, medications available to be administered, imaging to be ordered and viewed, intravenous options, etc. To add Figure 3:Patient State tab in CliniSpace. (Source: CliniSpace, IIL.) Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments85 Figure 4:Event Authoring Tab in CliniSpace. (Source: CliniSpace, IIL.) realism to the "patient," an Event Authoring function (Figure 4) offers a range of functionality that allows the author to set up and manage events in CliniSpace. Multiple conditions (Flags) can be added to a procedure to deliver a VP clinical response, e.g., the administration of a combination of drugs to lower a patient's heart rate. Event Authoring is a key and new component of the CliniSpace stand-alone set-up application and consists of the following elements: ­ Procedures ­ general, medications and intravenous options; ­ Attributes ­ vital signs, animations, sounds, texts, states; ­ Flags ­ conditions that are added to Procedures; ­ Transitions ­ Start and Stop conditions. The other key aspect of VP management is real-time facilitator intervention. In this, the running patient physiological model can be moved forward in time or stopped. This function is particularly useful, for example, after administering a drug such as an antibiotic, which, in the real world, would naturally take some time to have an effect. In this case, the scenario can be moved on a number of hours by the facilitator. Taken together, this dual active management functionality within CliniSpace ­ offline stand-alone authoring and real-time facilitator intervention ­ provides active VP management. Deployment and managing Once the clinical virtual environment has been developed and the DVP authored, then the simulation is ready for deployment. Once deployed, however, as it is a real-time simulation experience, it has to be managed in real time. This subsection explores the issues in deploying and managing these environments again using CliniSpace as the example. Figure 5 shows the interactions between the various elements of the CliniSpace Virtual Sim Centre. It places the physiology of real patients at the centre. In detail: ­ The pathophysiology of real patients is analysed and the key elements enumerated as a set of algorithmic models. ­ These are then modelled as a "DynaPatient" with a complex, dynamic pathophysiology, e.g., trauma. ­ The physiological variables of this DynaPatient are then modified by the medical educator using a simple-to-use authoring tool and at the same time the VCE is modified, e.g., medications made available, equipment made available in the ward, bed placed in a certain position, etc. ­ The modified variables are then loaded into a VP avatar model in the VCE and the case is started. ­ The clinician avatar then enters the virtual environment and examines, diagnoses, and/or treats 86Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments Table 1:Deploying and managing a VCE with DVPs. Clinical Non-clinical Deploy ­ Learning objectives Manage ­ Active VP management ­ Standardised patients ­ Scenario development ­ Authoring of DVPs ­ Standardised patients developed and trained ­ Application training to students and staff Assessments ­ Scheduling IT, staff, etc. ­ Access to online services ­ Technical support ­ Assessment systems Figure 5:Summary graphic showing active patient management within CliniSpace [21]. ­ ­ the patient using the available interactive medical objects, e.g., blood pressure cuff, medications, intubation, imaging, etc. Depending on the intervention, this may alter the underlying pathophysiological model (DynaPatient), which then, in turn, can be viewed in real time as signs and symptoms in the VP avatar or available data displays, e.g., skin colour change, urine output up, SPO2 steady, etc. This iterative process of examination, diagnosis, and treatment then continues for the duration of the scenario with the interactive VP responding in real time. Separately, an "invisible" facilitator can simultaneously view the actions of the clinician avatar during their session and can be available to assist or intervene. Finally, individual actions during the whole session are logged by individual user by CliniSpace and, if required, the whole session can be streamed to video for post-session analysis. Individual mentoring may need to be available to assist. Assessment systems have to be in place ­ debriefing, surveys, evaluation, action analysis, video analysis, etc. As is clear from Table 1, the skill set required to deploy and manage these simulations is primarily non-IT technical. Delivering complex real-time Virtual Clinical Experiences with Dynamic Virtual Patients: a new discipline? As in other sectors, the arrival of COTS platforms together with accessible user interfaces facilitates the ready deployment of particular technologies. The integration of interactive VPs in virtual environments with a simple-touse authoring tool is an example of these developments in the clinical simulation space. This paper has demonstrated in practice that integrating DVPs with VCEs results in a complex environment containing ­ A VCE, ­ Virtual actor(s), i.e., DVPs, ­ Real actors playing standardised patients and represented as avatars (optional), ­ Participating and collaborating professional actors, i.e., trainees, faculty, etc. In some senses, a simulation with VCEs and DVPs is more akin to staging a real-time piece of experiential theatre The deployment of immersive environments with interactive VPs requires a number of coordinated elements. These elements, which could also be regarded as constraints or limitations, include ­ Students and faculty have to be scheduled where realtime in-world interaction is to occur. ­ Access has to be scheduled to the online resources. ­ Technical support has to be in place. ­ Application training has to be given to faculty and students. Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments87 where the participants ­ both real and "robotic" ­ are a part of the play. The key differences are that the "play" is not designed to entertain, but is an "enactment" of professional practice that has to be accurate in both its context and components as it is a part of a strictly audited clinical education process and as such is subject to professional body oversight. The skill sets required to set up and implement these real-time professional collaborative clinical experiences could be viewed as a hybrid consisting of expertise in professional clinical education, real-time virtual simulation development and deployment, and live collaborative experiential theatre. It is perhaps somewhat surprising, therefore, that this novel mix of careful professional practice and complex of new technologies are increasingly being embraced by clinical professionals in a range of contexts. Some recent reported examples include their use to improve communication skills in nursing students [22], and a broader overview of their use in nursing, emergency departments, disaster response, and chronic disease management [23]. A recent study, reported earlier, on the introduction of Virtual Clinical Worlds and DVPs into Inter-professional Education for students in clinical settings have shown encouraging results [17], and the results of a randomised trial will shortly reinforce these results. It is not suggested, however, that VPs in virtual environments are suitable for all clinical education requirements. First, not all virtual environments are created equal [24], so the choice of virtual environment should be matched to educational requirements and, second, it is evident that some skills, e.g., technical (procedural) skills, are best taught by demonstration, through practice, and by observation in classrooms or laboratories, and that declarative (cognitive) skills in the medical domain are best taught on the one hand by lecture and on the other hand by mentoring the development of complex hypotheses by the student about real clinical cases, often in a real-world clinical context. In addition, a clear economic case for their deployment has already been cogently presented [25], which argues that the substantial further development of existing physical simulation centres will be constrained by the cost of manikins, professional staff, and physical space. This has been validated elsewhere, "Although the randomised controlled study did not show that the virtual patient simulation was superior to the mannequin-based simulation, both simulations have demonstrated to be effective refresher learning strategies for improving nursing students" clinical performance. Given the greater resource requirements of mannequin-based simulation, the VP simulation provides a more promising alternative learning strategy to mitigate the decay of clinical performance over time [26]. This paper proposes that, with the arrival of commercial software platforms that enable non-IT clinical staff to author VP simulations, the clinical simulation community can and should embrace these novel technologies. This is because, in addition to the new possibilities that these cost-effective virtual simulation environments offer for teaching and learning, they will offer new avenues to explore in clinical settings, such as action research [7], new clinical procedure development, and clinical environment planning. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. Competing interests: The funding organisation(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Bio-Algorithms and Med-Systems de Gruyter

Authoring, deploying, and managing dynamic Virtual Patients in Virtual Clinical Environments

Loading next page...
 
/lp/de-gruyter/authoring-deploying-and-managing-dynamic-virtual-patients-in-virtual-E8ERL5qq0A
Publisher
de Gruyter
Copyright
Copyright © 2015 by the
ISSN
1895-9091
eISSN
1896-530X
DOI
10.1515/bams-2015-0009
Publisher site
See Article on Publisher Site

Abstract

Following their introduction at the beginning of the 21st century, interactive or dynamic Virtual Patients are beginning to be used more widely in clinical education. They can be seen as being at the end of a continuum of simulation technical complexity, having been earlier developed on a wide range of "media": human actors, paper, video, physical mannequins, etc. This paper focuses on the current emergent more complex Virtual Patients in three-dimensional (3D) immersive clinical environments. In these environments, in silico 3D patient avatars interact directly in response to virtual clinical interventions undertaken by avatars, each of which is controlled by one or more users. The paper explores the issues of authoring, deploying, and managing these real-time, dynamic Virtual Patients using as an example the immersive clinical environment CliniSpace. As clinician-accessible Virtual Patient authoring is now becoming available in immersive clinical environments, so these wider clinical and managerial non-technical issues are coming rapidly to the fore. Keywords: clinical simulation; clinical virtual worlds; interactive Virtual Patients; Virtual Patients; virtual worlds. Introduction The earliest dynamic avatar noted in Sean P. Egan's "History of Avatars" review was Ananova, "designed to *Corresponding author: Dick Davies, Exec. Producer for Ambient, Ambient Performance, Suite 336, 43 Bedford Street London WC2E 9HA, UK, E-mail: dick.davies@ambientperformance.com LeRoy Heinrichs: Department of Obstetrics and Gynecology, Emeritus, Stanford University School of Medicine, Stanford, CA, USA Parvati Dev: Innovation in Learning, Stanford, CA, USA deliver the latest news over the Internet and on mobile devices, the 28-year-old, green-haired, British, 3D avatar had a full range of facial expressions and could speak multiple languages. She was designed to be attractive, with a global appeal, and trustworthy and believable, based on the principle that face-to-face communication is one of the oldest and most trusted forms of obtaining information, a theory still adhered today by companies creating or deploying avatars" [1]. Since Ananova's appearance in 2000, many interactive avatars have been developed on a multitude of online platforms representing a variety of sectors in business, art, marketing, gaming, and education, including medical education. Applications in clinical medicine include VitalSims, CliniSpace, The Anatomy of Care, Virtual Heroes, OLIVE, and others referenced in an excellent, illustrated, online Medscape review by L. Stokowski, R.N., in 2013 focused on nursing [2]. Each of these platforms enables avatars to demonstrate basic physiologic responses that teach users about typical responses to different clinical situations. In these computer-generated environments, people enter simulated worlds in real time from their PC or Pad to interact and collaborate with others represented by avatars in real time and make real-time decisions as they would in the "real world." When virtual world technology is implemented in the medical context, it can replicate most clinical environments: the emergency department, the ward, or a chaotic pre-hospital experience. These virtual worlds when deployed in a clinical context have been termed "Virtual Clinical Worlds" [3]. Combining Virtual Patient (VP) technology with Virtual Clinical Worlds offers, in principle, a powerful platform for developing and deploying realistic clinical experiences. The barrier until recently, however, has been that for clinicians to deploy these kinds of integrated experiences, substantial input from IT professionals was necessary. The arrival of Commercial-off-the-Shelf (COTS) suppliers is changing the game. Why do COTS change the game? The emergence of a COTS-supplied Virtual Clinical Environment (VCE)/VP integrated platforms now means that questions that were 80Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments previously distant, due the fact that these technologies were emergent and so not ready for widespread deployment, are now coming to the fore as these technologies allow clinical professionals to configure and deploy them rather than IT professionals. Possibly, we could now argue for a "tipping point" in terms of clinical acceptance and possibly uptake. The issues therefore become increasingly clinical and non-IT technical, i.e., how to introduce and manage the use of these simulation technologies on a day-to-day basis for mainstream clinical use. Further, these virtualisation technologies offer some unique features that give rise to some wider questions around the construction of virtual experiences for professionals. One hope that does come through strongly in the literature is that, "Future progress must be made to develop rich, guided authoring tools that will allow medical educators, not just computer scientists, to develop virtual patients..." [4]. This wish is the focus of this paper. This paper will ­ Provide a briefing background on VPs integrated into VCEs including some of their possible applications in clinical settings; ­ Show that issues that arise from their use in the mainstream are in the main now clinical and non-technical rather than IT technical using the CliniSpace VCE as the case study; ­ Explore the wider non-technical issues around the design and deployment of these technologies in professional contexts. use the rules of a game; simulations attempt to model a system in a manner that is consistent with reality [6]. It is important that this conceptual distinction is held in mind for this paper. Dynamic (or interactive) VPs (DVPs) and VCEs are subgenres of simulation that arise from very distinct and different heritages. VPs very much arise from professional clinical practice and have only recently embraced computer technologies. Virtual Clinical Worlds, on the other hand, have arisen from the repurposing and exploitation of existing video gaming technology approaches in the 2000s to develop realistic simulated clinical environments [7]. These two strands of virtualisation whilst originally from different contexts are now being brought together due to the fact that computing power and fast networks are democratising access to these powerful tools. What are virtual worlds? Virtual worlds are now an accepted technology, deployed widely in consumer spaces and with which most of the millennial generation, i.e., those born post-1980, are familiar and most are comfortable [8]. Known also as immersive environments, virtual worlds can be defined as a "synchronous, persistent network of people, represented by avatars, facilitated by networked computers" [9]. There are a wide range of variations, but virtual worlds are computer simulations offering some or all of the following [10]: ­ Three-dimensional (3D) spaces; ­ People represented by avatars; ­ Objects in world are persistent and maybe interacted with, e.g., moveable chairs, drivable vehicles; ­ Communication is usually in real time via voice, text, and gesture. Dynamic Virtual Patients and Virtual Clinical Environments: into the mainstream? According to Sauvé et al. [5], games and simulations are distinctive concepts. A game is a fictitious or artificial situation governed by rules that structure their actions in view of an objective, which is to win or to overcome an obstacle. They are integrated into an educational context when the learning objectives are associated formally to the content and the game enhances learning in the cognitive, affective, and/or psychomotor domains. In that case, they are often termed "serious games." On the contrary, a simulation does not necessarily involve competition, and the users are not trying to win. Many educational simulations, unlike games, can function without human intervention. Games suspend the rules of reality in order to Immersive Clinical Environments or Virtual Clinical Worlds Where virtual worlds are used specifically for clinical purposes, they could be termed Immersive Clinical Environments or Virtual Clinical Worlds. Heinrichs et al. [3] went on to propose that in clinical practice, virtual worlds were deployed because they offer the following attributes: Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments81 ­ ­ ­ ­ ­ ­ Presence, Immersion, i.e., engagement, Team-based activities, Real workplace settings, Safe "play spaces," Relatively low cost (compared to custom health-care game development), within a clinical or, in certain cases where the focus is still on the clinical, pre-hospital settings. The paper now looks at VPs before moving on to examine how VPs can work within virtual worlds. Virtual Patients The genesis of VPs is the "human standardised patient" developed by Dr. Howard Burrows in 1963. Whilst computer simulations of patients started in the 1960s, it is only recently with the widespread uptake of computers in medical education that VPs, as they are now understood, have expanded. The term "Virtual Patients," however, is a portmanteau term for a very wide range of distinct approaches. "Such approaches include case presentations, interactive patient scenarios, virtual patient games, human standardised patients, high fidelity software simulations, high fidelity manikins and virtual conversational agents" [4]. The conceptualisation and building of VPs had its beginning in 1911 with a life-size "Mrs. Chase doll," an early mannequin specifically developed for nurses to practice their clinical skills [11]. Since then, an entire industry has developed responding to the market for physical mannequins. As computer technology and online learning has emerged, building avatars as VPs is offering to replace physical mannequins as the learning objects. A major difference only mentioned here are the actions of caregiver avatars that are able to move about in a 3D space, but appropriate patient avatar actions would be focused on movements onto and off a chair, couch or bed, or into or out of a wheelchair, or rotation side-to-side or sitting up in bed. These are actions that mannequins are unable to perform, except as users position them manually. In contrast, virtual avatars can be directed to run or walk and change direction as the keyboard's pointer keys are activated. Torso twisting or flexion is programmed to be displayed upon compressing a () key, or another chosen by the coder. Arm, head, and facial movements are guided in coordination with spoken words. For example, auscultation of the chest or heart integrates the appropriate sounds with a stethoscope positioned and held by the user. Abnormal sounds are encoded as appropriate for the scenario. The mention of auscultation sounds leads to focusing on the intrinsic physiologic system that resides in all living organisms, and in realistic VPs. The current generation of VPs will usually be designed to encompass either a basic or perhaps a more sophisticated system, and include physiologic sounds of the cardiovascular and gastrointestinal systems. Such sounds can be programmed to represent either normal or pathological conditions. The physiology model selected may be extremely sophisticated based on established physiological principles such as those incorporated in HuMod [12] or upon reports of big data derived from clinical practice/ experience. The latter "pragmatic model" continues to be selected for CliniSpace's VPs [13]. The fundamental parameter or "driver" for the pragmatic model(s) is blood volume upon which an individual's vital signs are based. Representations of population-based data reflect the height, weight, and blood volume differences among men vs. women and the body mass index (BMI) of healthy people and of those overweight and obese individuals [14]. Similarly, the increasing blood volume of pregnant women reflects the changing BMI of that physiological condition [15]. In the CliniSpace Virtualsimcenter, the patient monitor displays the vital signs of avatars based on this fundamental principle. One thus recognizes the core value of using blood volume in the architecture of VPs that can be designed to accurately reflect and teach this feature of human physiology. The term "Virtual Patients," however, is a portmanteau term for a range of distinct approaches. "Such approaches include case presentations, interactive patient scenarios, Virtual Patient games, human standardised patients, high fidelity software simulations, high fidelity manikins and virtual conversational agents" [4]. Given that the VPs developed in CliniSpace cover a number of these approaches, then the term "dynamic Virtual Patient" is proposed. The characteristics of the DVP are as follows: ­ 3D clinical environment, ­ A human physiology engine, ­ 3D avatars, ­ Data displays, ­ Medical procedure capability, ­ Reporting or assessment capabilities. In CliniSpace, each DVP (DynaPatient) has an underlying pathophysiological model consisting of 82Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments 22 y/o pregnant female, auto injury/traumatic hemorrhage, EOIS=5; GASHD 1.0 250 TR-005-traumatic hemorrhagic shock-untreated 200 Response of model Dysrrhythmias, bradycardia, or PEA, (pulseless electrical activity) PR SaO2 BP-S RR 50 11 min to death 0 8 9 7 Time, min CliniSpaceTM Figure 1:Graphical representation of a pathophysiological model for a patient with traumatic haemorrhage without intervention. (Copyright IIL.) ­ ­ ­ ­ ­ ­ ­ A set of states (healthy, deteriorating, fibrillation, sepsis, severe sepsis, shock, asystole, death). Mathematical equations that define the smooth changes within that state. Conditions that define transitions between states. A physiology model that interacts within the environment. Each model has a "driver" or independent variable. Some of the variables are "observable," for example, as the vital signs appear on a monitor. The AI of the VP is linked to the AI of the environment. Each object in the world has its own rules: e.g., administering fluids at the IV stand temporarily decreases the heart rate and increases the blood pressure. An important lesson to be learnt is that acute blood loss must be stopped before administration of fluids will become helpful (see Figure 1). Authoring, deploying, and managing Virtual Clinical Experiences In the introduction to this paper, it is claimed that these new VCE/DVP integrated platforms now allow clinical professionals to configure and deploy them without the requirement for IT professionals. The consequences of the arrival of these modifiable platforms is that the issues are now becoming clinical and non-IT technical, i.e., how to introduce and manage the use of these simulation technologies on a day-to-day basis for mainstream clinical use with clinician educators. This section investigates, using CliniSpace as a case study example, how to author and deploy/manage these virtual simulation technologies. The focus of this paper is on DVPs embedded in Virtual Clinical Worlds, and what is clear is that at the centre of these new VCEs is the underlying human physiology engine. Poulton [16] has argued that, "It is now possible to consider the extension of the current relatively lightweight Virtual Patient into a truly interactive patient simulation, an "e-human" or "digital avatar." At this stage, the simulation takes on new capabilities, offering authentic patient management, clinical and communication skills training; the potential capability to mimic the health or disease of any citizen." This paper explores these emergent "new capabilities" for simulation. Authoring Once the overarching structure and plan for a clinical virtual world is established, one must consider "What type of Virtual Patient(s) would one like to design and develop? The answer to this question relates directly to the teaching/learning objective(s). Choosing an avatars' appearance, whether youthful or aged, male or female, thin or obese, etc., is guided by that objective. For example, one would be unlikely to select a young thinly built individual to exhibit hypertension, but this clinical condition would not be a surprising finding in a young obese avatar. In contrast, both young and older, thin and Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments83 Figure 2:Students in CliniSpace cooperating in treating a VP. (Copyright IIL and Charles Drew University.) obese individuals are subject to systemic infections that may present with flushed appearances or skin rashes [17]. In a study among 60 students at Charles Drew University School of Medicine, an affiliate of the University of California, Los Angeles, a large majority, 90%, reported ease of use, and 70% reported effective learning among team members about antibiotic overdose because the avatars in an infection scenario developed a skin rash unique to the excessive drugs use. The virtual rash subsided and the skin colour of the avatar returned to normal after the excessive rate of administration was corrected. This example illustrates the dynamic and innovative design of CliniSpace's DynaPatientsTM as learning objects (Figure 2). Authoring of VPs is an "art" in which experience, reflection, and learning through the opinions of users supports engagement, adoption, and better learning scores. A Carnegie Mellon University webpage [18] offers a cogent and relevant discussion of learning objectives. Objectives, assessments, and instructional strategies are key elements in authoring. As one uses the intrinsic CliniSpace, or any other authoring tool, the most appropriate directive is, "At the end of this to-be-authored scenario, what do I want/expect the learner to be able to learn and/ or do?" The answer to this seemingly simple question must consider the learner's prior knowledge and experience and comfort level with computers, online e-learning, video games play, and adoption of new methods and technologies. A common strategy is starting simply, and increasing complexity as one proceeds, thus allowing for multiple entry points for users. This design permits a progressive scale for entry into the set of scenarios, or entry at selected points. This feature enables instructors to use the scenarios for testing. Most [19, 20] of the few VP models available for deployment in virtual environments are runtime, i.e., they run from start to finish and cannot be started or stopped, or even most cases parameterised simply. Two types of VP designs can be distinguished: a "narrative" or passive structure and a "problem-solving" or active structure. In the narrative/passive cases, the simulation represents a single medical state, often in considerable detail, and with relevant graphics, audio, and visual media displaying the patient's medical condition. Fewer simulations support the evolution of the "problem solving"/active patient's state, both with and without medical intervention. In the problem-solving/active model, one specifies both gradual changes in physiologic variables as well as a number of discrete important "states," with the patient moving from state to state based on the VP's condition and on the actions taken by the learner. From this the question arises as to why "active" should be preferred to "passive" VPs? Two reasons are advanced. First, "passive" VPs are experienced as pale imitations of real-world patients so medical professionals find it difficult to take them seriously and, second, they can be "learnt" and then "gamed" by the end user. To expand 84Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments on this, the problem with passive models, in addition to the highlighted poor end-user experience already mentioned above, is that most professional educators are in organisational contexts that have distinctive clinical and other associated procedures that are followed for specific clinical situations. They, therefore, need the clinical environment and VPs to be customisable in order to be able to train for local requirements. Offering VP models that "do not quite fit with our procedures" undermines credibility. A further issue is "gaming the game" where simple runtime models can be learnt and then "gamed" by the user over a period of time. Their passivity has a repetitive and thus predictable nature and performance. They are then viewed as not sufficiently realistic and dismissed as "simply games." Once played and their properties experienced, going back again and again is not compelling. In response to these issues, the approach taken in CliniSpace has been to build simple-to-use standalone authoring tools and enable the running VP model to be adjusted by the scenario facilitator in real time. In practice, this first enables the scenario controller to rapidly "tweak" the scenario prior to runtime, so changing various elements of the scenario, and second, by adjusting the running VP model in real time, the scenario controller is able to alter the original course of the scenario in response to real-time activities and happenings in the immersive environment. The sequence, severity, and resolution of the activities and happenings can be built dynamically into the scenarios [14]. CliniSpace offers authoring of the clinical environment and the in silico VP via a stand-alone set-up and authoring tool. This tool ­ ­ Parameterises and sets triggers in the default VP pathophysiological model in the CliniSpace VCE, Sets up various physical aspects of the CliniSpace environment, e.g., rooms, equipment, dialogue, imaging graphics, etc., and then loads them via a simple file upload into the live single or multiuser CliniSpace environment. Each CliniSpace Virtual Clinical Experience can therefore be individualised for the end user(s) by the author/facilitator to link to a specific educational requirement. Headers for each of 12 tabs of the "CliniSpace Authoring Tool" are Learning Objectives, Patient Introduction, General Observations, Patient History, Physical Exam, Differential Diagnosis, Recommended Case Protocol, Vital Signs, Imaging, Labs, Medications, and Intravenous Options. Each of these is represented in a mobile, interactive, electronic medical record computer sited nearby each patient's bed. The page for each topic has key words listed for which relevant and unique "answers" can be introduced. These pages are further organised as templates to enable authors to create six consecutive "states" for each VP over whatever time period is selected. These features are designed to facilitate efficient authoring. As an example (Figure 3), the Patient State interface sets up the parameters to the VP's initial clinical state. Further interfaces set up the patient hospital room with the objects to be visible at the beginning of a case, medications available to be administered, imaging to be ordered and viewed, intravenous options, etc. To add Figure 3:Patient State tab in CliniSpace. (Source: CliniSpace, IIL.) Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments85 Figure 4:Event Authoring Tab in CliniSpace. (Source: CliniSpace, IIL.) realism to the "patient," an Event Authoring function (Figure 4) offers a range of functionality that allows the author to set up and manage events in CliniSpace. Multiple conditions (Flags) can be added to a procedure to deliver a VP clinical response, e.g., the administration of a combination of drugs to lower a patient's heart rate. Event Authoring is a key and new component of the CliniSpace stand-alone set-up application and consists of the following elements: ­ Procedures ­ general, medications and intravenous options; ­ Attributes ­ vital signs, animations, sounds, texts, states; ­ Flags ­ conditions that are added to Procedures; ­ Transitions ­ Start and Stop conditions. The other key aspect of VP management is real-time facilitator intervention. In this, the running patient physiological model can be moved forward in time or stopped. This function is particularly useful, for example, after administering a drug such as an antibiotic, which, in the real world, would naturally take some time to have an effect. In this case, the scenario can be moved on a number of hours by the facilitator. Taken together, this dual active management functionality within CliniSpace ­ offline stand-alone authoring and real-time facilitator intervention ­ provides active VP management. Deployment and managing Once the clinical virtual environment has been developed and the DVP authored, then the simulation is ready for deployment. Once deployed, however, as it is a real-time simulation experience, it has to be managed in real time. This subsection explores the issues in deploying and managing these environments again using CliniSpace as the example. Figure 5 shows the interactions between the various elements of the CliniSpace Virtual Sim Centre. It places the physiology of real patients at the centre. In detail: ­ The pathophysiology of real patients is analysed and the key elements enumerated as a set of algorithmic models. ­ These are then modelled as a "DynaPatient" with a complex, dynamic pathophysiology, e.g., trauma. ­ The physiological variables of this DynaPatient are then modified by the medical educator using a simple-to-use authoring tool and at the same time the VCE is modified, e.g., medications made available, equipment made available in the ward, bed placed in a certain position, etc. ­ The modified variables are then loaded into a VP avatar model in the VCE and the case is started. ­ The clinician avatar then enters the virtual environment and examines, diagnoses, and/or treats 86Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments Table 1:Deploying and managing a VCE with DVPs. Clinical Non-clinical Deploy ­ Learning objectives Manage ­ Active VP management ­ Standardised patients ­ Scenario development ­ Authoring of DVPs ­ Standardised patients developed and trained ­ Application training to students and staff Assessments ­ Scheduling IT, staff, etc. ­ Access to online services ­ Technical support ­ Assessment systems Figure 5:Summary graphic showing active patient management within CliniSpace [21]. ­ ­ the patient using the available interactive medical objects, e.g., blood pressure cuff, medications, intubation, imaging, etc. Depending on the intervention, this may alter the underlying pathophysiological model (DynaPatient), which then, in turn, can be viewed in real time as signs and symptoms in the VP avatar or available data displays, e.g., skin colour change, urine output up, SPO2 steady, etc. This iterative process of examination, diagnosis, and treatment then continues for the duration of the scenario with the interactive VP responding in real time. Separately, an "invisible" facilitator can simultaneously view the actions of the clinician avatar during their session and can be available to assist or intervene. Finally, individual actions during the whole session are logged by individual user by CliniSpace and, if required, the whole session can be streamed to video for post-session analysis. Individual mentoring may need to be available to assist. Assessment systems have to be in place ­ debriefing, surveys, evaluation, action analysis, video analysis, etc. As is clear from Table 1, the skill set required to deploy and manage these simulations is primarily non-IT technical. Delivering complex real-time Virtual Clinical Experiences with Dynamic Virtual Patients: a new discipline? As in other sectors, the arrival of COTS platforms together with accessible user interfaces facilitates the ready deployment of particular technologies. The integration of interactive VPs in virtual environments with a simple-touse authoring tool is an example of these developments in the clinical simulation space. This paper has demonstrated in practice that integrating DVPs with VCEs results in a complex environment containing ­ A VCE, ­ Virtual actor(s), i.e., DVPs, ­ Real actors playing standardised patients and represented as avatars (optional), ­ Participating and collaborating professional actors, i.e., trainees, faculty, etc. In some senses, a simulation with VCEs and DVPs is more akin to staging a real-time piece of experiential theatre The deployment of immersive environments with interactive VPs requires a number of coordinated elements. These elements, which could also be regarded as constraints or limitations, include ­ Students and faculty have to be scheduled where realtime in-world interaction is to occur. ­ Access has to be scheduled to the online resources. ­ Technical support has to be in place. ­ Application training has to be given to faculty and students. Heinrichs et al.: Dynamic Virtual Patients in Virtual Clinical Environments87 where the participants ­ both real and "robotic" ­ are a part of the play. The key differences are that the "play" is not designed to entertain, but is an "enactment" of professional practice that has to be accurate in both its context and components as it is a part of a strictly audited clinical education process and as such is subject to professional body oversight. The skill sets required to set up and implement these real-time professional collaborative clinical experiences could be viewed as a hybrid consisting of expertise in professional clinical education, real-time virtual simulation development and deployment, and live collaborative experiential theatre. It is perhaps somewhat surprising, therefore, that this novel mix of careful professional practice and complex of new technologies are increasingly being embraced by clinical professionals in a range of contexts. Some recent reported examples include their use to improve communication skills in nursing students [22], and a broader overview of their use in nursing, emergency departments, disaster response, and chronic disease management [23]. A recent study, reported earlier, on the introduction of Virtual Clinical Worlds and DVPs into Inter-professional Education for students in clinical settings have shown encouraging results [17], and the results of a randomised trial will shortly reinforce these results. It is not suggested, however, that VPs in virtual environments are suitable for all clinical education requirements. First, not all virtual environments are created equal [24], so the choice of virtual environment should be matched to educational requirements and, second, it is evident that some skills, e.g., technical (procedural) skills, are best taught by demonstration, through practice, and by observation in classrooms or laboratories, and that declarative (cognitive) skills in the medical domain are best taught on the one hand by lecture and on the other hand by mentoring the development of complex hypotheses by the student about real clinical cases, often in a real-world clinical context. In addition, a clear economic case for their deployment has already been cogently presented [25], which argues that the substantial further development of existing physical simulation centres will be constrained by the cost of manikins, professional staff, and physical space. This has been validated elsewhere, "Although the randomised controlled study did not show that the virtual patient simulation was superior to the mannequin-based simulation, both simulations have demonstrated to be effective refresher learning strategies for improving nursing students" clinical performance. Given the greater resource requirements of mannequin-based simulation, the VP simulation provides a more promising alternative learning strategy to mitigate the decay of clinical performance over time [26]. This paper proposes that, with the arrival of commercial software platforms that enable non-IT clinical staff to author VP simulations, the clinical simulation community can and should embrace these novel technologies. This is because, in addition to the new possibilities that these cost-effective virtual simulation environments offer for teaching and learning, they will offer new avenues to explore in clinical settings, such as action research [7], new clinical procedure development, and clinical environment planning. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Research funding: None declared. Employment or leadership: None declared. Honorarium: None declared. Competing interests: The funding organisation(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

Journal

Bio-Algorithms and Med-Systemsde Gruyter

Published: Jun 15, 2015

References