Simulation of a Virtual Patient with Cranial Nerve Injury Augments Physician-Learner Concern for Patient Safety - PowerPoint PPT Presentation

Loading...

PPT – Simulation of a Virtual Patient with Cranial Nerve Injury Augments Physician-Learner Concern for Patient Safety PowerPoint presentation | free to download - id: 3d96d3-NGJmY



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Simulation of a Virtual Patient with Cranial Nerve Injury Augments Physician-Learner Concern for Patient Safety

Description:

Aaron Kotranza, Ben Lok, Juan Cendan University of Florida Colleges of Medicine and Engineering Gainesville, FL MedBiquitous Annual Conference, London, 2010 – PowerPoint PPT presentation

Number of Views:31
Avg rating:3.0/5.0
Slides: 28
Provided by: medbiqOrg
Learn more at: http://www.medbiq.org
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Simulation of a Virtual Patient with Cranial Nerve Injury Augments Physician-Learner Concern for Patient Safety


1
Simulation of a Virtual Patient with Cranial
Nerve Injury Augments Physician-Learner Concern
for Patient Safety
  • Aaron Kotranza, Ben Lok, Juan Cendan
  • University of Florida
  • Colleges of Medicine and Engineering
  • Gainesville, FL
  • MedBiquitous Annual Conference, London, 2010
  • Presentation 39558

2
NERVE
  • Neurological Exam Rehearsal Virtual Environment
    (NERVE)
  • Virtual Patient targets abnormal physical exam
    findings of the cranial nerves
  • Interaction includes
  • Verbal communication with natural speech
  • Gestures
  • Virtual tool interactions
  • Patient Vision Feedback (PVF) allows the user to
    share the patient visual experience

3
Motivation
  • Cranial nerve palsy/injury is uncommon but
    dramatic and critical physical finding
  • Cannot be feigned by SP
  • Opportunities for clinical teaching are rare
  • Identified as a particularly relevant niche for
    virtual human representation

4
Current State of the Art
  • Traditional lecture, textbook-video didactics,
    some simulation systems
  • Fundoscopic exams
  • UC Davis disembodied eyes system
  • http//cim.ucdavis.edu/eyes/version15/eyesim.html
  • Rick Lasslo,M.D., M.S., Gary Henderson, PhD, and
    John Keltner, M.D. , UC Davis School of Medicine

5
System Description
  • NERVE uses a life-size virtual patient with
    cranial nerve injury
  • Learner performs examination using
  • Speech
  • Virtual tools (ophthalmoscope, eye chart, hand
    and fingers) manipulated by a Nintendo Wii Remote
    (WiiMote)
  • Representation from the VP
  • Restricted eye motion
  • Double vision
  • Relevant history

6
NERVE System
  • VP modeled in Autodesk Maya
  • Rendered with open source Ogre 3D
  • VP portrayed on a 52 LCD
  • User wears IR fiducials for location/angle of
    vision tracking
  • Naturalpoint Optitrack

7
Patient Vision Feedback
  • Real-time experience
  • Allows learner to view a virtual room through the
    eyes of the patient
  • Goals
  • Provide learner with more information for
    understanding how a particular CN injury affects
    the patients vision
  • Provide insight into the patients overall
    wellbeing and safety

8
Typical Interaction
  • Learner greets VP and queries with natural speech
  • Virtual People Factory
  • Learner performs examination
  • Ophthalmoscope with and without light
  • Fundoscopic exam
  • Hand fingers tool for ROM
  • Convergence
  • Peripheral Vision
  • Eye chart with cover one eye

9
Interaction with VP using text box
  • Typical exchange between user and virtual patient
    - can occur in natural spoken language or in
    text-box

10
Additional Responses
  • Also able to reproduce on command
  • Head tilt
  • Chin to chest
  • Smile, frown
  • Stick out tongue
  • Raise eyebrows and wink
  • Puff out cheeks
  • Turn head side-to-side
  • Collision detection allows examination of Cranial
    Nerve 5 (virtual finger and facial sensation)

11
Special Challenge
  • Eye movement requires coordination of 6 muscles
  • Characteristic presentation when abnormal would
    require tremendous computational expense
  • Developed our own model that is not physically
    based, but consistent with physiologic
    presentation

12
Special Challenge
  • Visual model uses interpolation of 8 cardinal eye
    movements (left, up-left,, down)
  • Can be visualized as a 2-D pitch-yaw plane
  • For each eye the model defines a set of 8 vector
    criteria

13
Range of Motion
  • Typical Range Examination
  • Red/Green boxes as teaching tools

14
Wii Remote
  • Shaped like a hand-held tool and high degree of
    freedom control
  • Bluetooth, 11 buttons, 3 orthogonal
    accelerometers, 45 degree field-of-view IR camera
    that tracks up to 4 points
  • Displays information with integrated LEDs,
    speaker and vibration
  • 100Hz update and sub-centimeter accuracy
  • Inexpensive

15
WiiMote Applications
  • Ophthalmoscope
  • Example Contact with eye leads to vibration
    and complaint from patient
  • Trigger button used to create light for pupil
    examination

16
Pupil Reflex Example
  • Exam of the pupil using laptop
  • Can be done with WiiMote

17
Patient Vision Feedback System
  • Head mounted display
  • Virtual room is rendered twice reflecting the
    perspective of each eye

18
Methods
  • 18 students in second year of medical school
  • Had prior experience with normal neurological
    exam
  • Had not examined a patient with active pathology
  • 9 students experienced PVF prior to examining the
    patient
  • The other 9 did so after data collection

19
Methods 2 Examination Experience
PV Passive Sees what the patient sees
NPV Active Interviews and examines patient
  • Two participants at a time
  • Completed experience survey
  • NERVE and PVF were explained by instructor
    (Author AK)
  • Participant PV donned a stereoscopic HMD
    displaying the virtual room while participant NPV
    examined the NERVE
  • Both participants are synchronized to the same
    patient exam and environment

20
Methods 3 -Counterbalanced Design
Both users complete survey, switch roles
Both users complete final survey
21
Results
  • 2 participants never experienced double vision
  • One in each group
  • ? Technical, individual unclear and removed from
    analysis
  • Correct diagnosis was established equally by both
    groups
  • 9 of 9 in PV 8 of 9 in NPV (pNS)

22
Results 2
  • PVF provided enough data to accurately diagnose
    the CN injury
  • 81 were able to make the Dx using PVF only
  • 7/8 in PV group identified CN3 palsy
  • 6/8 in NPV group identified CN6 palsy
  • ?2 5.1 plt0.05

23
Results 3
  • Concern for Patient Safety
  • Affective performance (perspective taking leading
    to concern for patient wellbeing) evaluated as
    expressed (verbal or written) concern for safety
  • E.g. Informing the VP that he should not drive
    home, etc.
  • 7/8 learners in PV expressed concern vs. 0/8 in
    the NPV group (plt0.005 Fishers exact)
  • 5 on post patient vision survey
  • 1 of five told the patient directly during the
    exchange
  • 3 on the overall survey
  • 1 expressed concern in both surveys

24
Results 4
  • Note
  • All participants were primed to think about
    driving as the opening statement from the VP was
  • I was driving home from work and all of the
    sudden the lines on the road started to cross.

25
Conclusions
  • NERVE presents a life-sized virtual human
    presenting with a cranial nerve palsy
  • Novel vector methodology provides efficient
    modeling
  • Content validity has been completed on prior
    testing with 32 students and residents
  • Students can arrive at the correct diagnosis
    using traditional examination skills as well as
    through the eyes of the patient technique
  • Patient safety will be augmented by giving the
    examiner a patient-perspective of the presenting
    condition

26
Future and Support
  • Currently developing and testing a web-deployable
    version of NERVE
  • Will not initially support voice recognition but
    can type-in-box
  • Will not support patient vision
  • Widely deployable
  • Dr. Cendan and Dr. Lok receive support from NIH
    NLM R03-LM009646.

27
Early Web-Based User Data, n14
About PowerShow.com