Why build a browser-based immersive virtual medical environment

Making Virtual Worlds More Accessible for Educators Parvati Dev, PhD W.L. Heinrichs, MD, PhD Innovation in Learning Inc. (with IndusGeeks Solutions) Adapted from a poster presentation at INACSL 2010, Las Vegas, Nevada #46

ABSTRACT A new simulation tool, virtual worlds, is being introduced into nursing curricula (1,2) but is not yet widespread. Bibliographic search on PubMed provides 151 results for the term “virtual world” but only 8 when the term “nursing” is added. To better understand the opportunities and barriers for use of virtual worlds in healthcare, we undertook a retrospective analysis of the free text answers in questionnaires and focus group sessions from our earlier studies on usability and learning performance of a number of virtual medical environments. These studies included: a) residents and medical students reviewing the ATLS protocol in a virtual emergency department, b) nurses and surgical residents conducting a mass casualty exercise in a virtual emergency department, c) nurses rehearsing communication patterns in shift change, and d) high school students practicing decision making in a situation that might require CPR. The opportunities fell into two major categories: engagement and remote learning. Engagement was expressed by phrases such as “I felt involved”, “it was like being there”, “my blood pressure went up along with the (virtual) patient’s blood pressure”, and “it was easier to participate from behind my avatar than in face-to-face role playing”. The opportunity for remote learning was expressed as “could I do this from my computer at home”, “most of us miss training sessions because they are not scheduled when we can get away”. Barriers also fell into two major categories: IT (information technology) problems, and usability. The primary IT problem arose in hospitals because firewalls prevented download and installation of the necessary client software, as well as external server access when running the software. Many sites also found that their computers had inadequate graphics processing power. Regarding usability, many felt there was a steep learning curve before they could be comfortable navigating the world and using its capabilities. Since most users expected to use this simulation tool less than once a month, the investment in learning and relearning exceeded the value received. Based on this analysis, we recommend that the next generation of virtual worlds remain engaging, require no download, be web browser based, run on commodity computers, and have a simplified user interface. We present details of our analysis and examples of a new medical virtual world designed to address the above issues. A survey of health-related activities on Second Life. Beard L, Wilson K, Morra D, Keelan J. J Med Internet Res. 2009 May 22;11(2):e17. Review. Schmidt B, Stewart S. Implementing the virtual world of Second Life into community nursing theory and clinical courses. Nurse Educ. 2010 Mar-Apr;35(2):74-8.

Virtual Worlds feel three-dimensional. In Virtual Worlds, you can move around, interact, and talk, as in the real world. Example: Virtual London (above)

Clinical Virtual Worlds Example: Duke University School of Nursing Online location for lectures and classroom activities Students from any location can join Normal classroom interaction - raise hand, ask questions by speaking Simulated manikin available

Our Studies at Stanford University Studied usability, acceptance of technology, and learning efficacy with a number of virtual worlds: 2005: Virtual ED (4th year medical students and 1st year residents) 2006: Virtual ED 2 pilot (fire fighters and first responders) 2007, 2008: Virtual ED 2 (practicing physicians and nurses) 2008: CPR SimErgency (high school students in PE course)

Virtual Emergency Department (VED): A Virtual World that accurately represents an existing real world space The Emergency Department at Stanford Hospital (Study done with Stanford Hospital and Forterra Systems) Vending machines in the hallway Ambulance at entrance to Emergency Two bays in VED

Study of learning efficacy (VED 1, 2005) Subjects: 4th year medical students and 1st year residents Task: Manage trauma cases according to ATLS protocol Study design: Pre-test scenario, 4 learning and debrief scenarios, post-test scenario Evaluation instrument: EMCRM performance rating sheet

Study results: Comparing learning with HPS manikin and Virtual ED Subjects: 4th year medical students and 1st year residents Task: Manage trauma cases according to ATLS protocol Study design: Pre-test scenario, 4 learning and debrief scenarios, post-test scenario Evaluation instrument: EMCRM performance rating sheet Outcome: No significant difference in learning with manikin compared to virtual ED Youngblood P, Harter PM, Srivastava S, Moffett S, Heinrichs WL, Dev P (2008), Design, development, and evaluation of an online virtual emergency department for training trauma teams, Simulation in Healthcare: The Journal of the Society for Simulation in Healthcare; 3(3): 146-53.

Problem: If these systems are effective for learning, why do we not see more use of virtual world simulations for healthcare education? Transcripts of focus group discussions, free text answers in questionnaires, and notes of discussions with IT personnel, were used. Answers from all studies were grouped in 4 categories System robustness Usability Engagement Usefulness To study this problem, we conducted a retrospective analysis of the questionnaires, notes and transcripts.

Questionnaire responses: Physicians and nurses at two hospitals, before and after training on a CBRNE scenario in a virtual medical environment To be published in WL Heinrichs, P Youngblood, P Harter, L Kusumoto, P Dev. “Training Healthcare Personnel for Mass-Casualty Incidents in a Virtual Emergency Department”, Prehospital and Disaster Medicine , Volume 25, Number 5, 2010.

System robustness These issues are primarily from notes taken during system installation and operation. Server installation is complex, requiring skilled technical support. A SaaS (Software as a Service) model is desirable, where the server is maintained at a third party or company site. Accessing an outside server, for installation or operation, is impossible when institutions have a firewall. The solution is to have the user’s client software be web browser-based, since the port for HTTP, or web browser, communication is kept open by the firewall. Client software installation is usually not difficult But it must be permitted by hospital IT policy. Once installed, the client software is robust (does not crash). This was less true of the Second Life client. Hospital-based client computers must provide security against inadvertent access of patient data (HIPAA). A long term solution is a VPN (Virtual Private Network).

This data is from observation of use by the subjects, and from transcripts of focus group discussions. Navigation and interaction were more difficult for non-gamers to learn. Subjects learnt to use the arrow keys for navigation but would prefer to point at a location and automatically move there. Multiple cameras are useful but complex. The different camera views caused the most problems: First-person view, third-person view, free movement of camera. Occasional failure of “collision” allowed the camera to move through a wall, causing confusion. Interaction with the patient was difficult (comments from transcript). “ Too hard to assess patients; takes too many steps.” “ Need a way to know what’s going on with that patient without being attached to the bed.” “ I need a different button to listen to the heart, and a different button to listen to the lungs. That’s silly, just need to assess…heart, lungs, secretions … then do the high level cognition of .. OK this patient is just going to require a little bit of mask… no big deal, ...and this person is really sick” “ I don’t need to listen to the heart; I don’t need to listen to the lungs, I don’t need to touch their skin…I just look at them and go, “sh**, we’re in trouble. OK, let’s go, now!!” Usability

Engagement From questionnaires, and transcripts of focus groups. 17 out of 22 subjects rated immersion as high or very high. “ It’s a lot more real here than in the text. We can read the text—but we forget” “ The fact that it was interactive; it stays with you.” “ What’s good about this is you actually feel more involved with it, than with the dummy (instrumented manikin-based simulation). I felt like I can immerse myself more in this than working with the dummy…” “ More realistic way to practice a mass casualty drill than the ones done once a year.” (About triage team leader versus team member). “I thought it was pretty real. In reality you are not going to know how many patients are out there; you do have to rely on somebody to tell you how sick they are and how many there are… it’s tough to figure out how many people you need where. I thought that was good.”

Usefulness From transcript of focus group, mass casualty scenario training What was your overall impression? I think it’s a more realistic way to practice a mass casualty drill than having a once a year hospital thing when you close down the ER. In the live drill it’s chaos, it’s not clear what your role is and what you’re supposed to do. If it’s something you do on a regular basis like an auto accident coming in. You’re like, not a big deal, I take care of auto accident patients all the time. Yeah, there might be 10 of them this time, but still not a big deal, I can handle that. But if you’re having to decon somebody and that’s not something you do regularly, in your regular job, then that’s a very different thing. And if you’re doing it once a year, you’re not going to remember. So if we can do it on a computer 3 or 4 times a year and then actually do a hands on it makes it so you’re like I know exactly how to do that. Yeah, because it was a lot more real here than in the text. We can read the text, but we forget. But this is almost like having a real patient. I think it’s definitely an opportunity to learn. You can kind of see how you improve. Yeah, you can do it over and over again. I think the importance of these types of simulators is more to train mid and upper level people how to do more complex systems management, more complex decision making and treatment algorithms. Then in that sense we should make the actual assessment of the patient less difficult and less cumbersome, so then we can move into those higher level skills that are more difficult, because that’s actually where you make the difference

Our Solution: CliniSpace™ We implemented a new Virtual Medical Environment, CliniSpace™ , to address the needs that we identified 3D and immersive Browser-based application Transparent download, with self-install Server maintained externally Simplified navigation and cameras Rich environment for patient care and management Realistic environment, avatars and patients Virtual patients that satisfy learning objectives for junior and advanced learners Scenarios for team and crisis management learning Some needs may remain difficult to satisfy Balancing richness of interactive environment with minimizing clicking Creating a completely realistic visual environment that remains responsive and has no delays.

Study of user feedback identified these needs: A 3D realistic immersive environment A rich patient management environment but without too many clicks Easy to install and easy to use Runs on commodity computers ________________ We also implemented the following: Avatars of normal everyday people Accurate portrayal of medical environment, procedures, scenarios Accurate patient pathophysiology Tracking and recording for performance analysis Web access to amplify functionality of environment, including social networking Summary

Thank you! [email_address] www.innovationinlearning.com See a video of CliniSpace at: http://www.indusgeeks.com/ClinispaceWalkthrough_09.06.10.mp4

Why build a browser-based immersive virtual medical environment

More Related Content

Viewers also liked (15)

Similar to Why build a browser-based immersive virtual medical environment (20)

More from Parvati Dev (10)

Recently uploaded (20)

Why build a browser-based immersive virtual medical environment