Download PPTX - MedBiquitous

moancapableAI and Robotics

Nov 17, 2013 (3 years and 8 months ago)

58 views

Designing a Virtual Patient for
Communication Training

April Barnes, M.S., Ph.D. Candidate
1,2

Jennifer Cloud
-
Buckner., Ph.D. Candidate
1,2


Jennie Gallimore, Ph.D.
1,2,3


Phani Kidambi, Ph.D.
1


Rosalyn Scott, M.D., M.S.H.A.
1,2,3


Ohio Center of Excellence in Human
-
Centered Innovation
1

Department of Biomedical, Industrial, and Human Factors Engineering
2


Department of Surgery, Boonshoft School of Medicine
3

Wright State University, Dayton, OH, USA

Source: http://research.bidmc.harvard.edu/VPtutorials/default.htm


Information presented to user through video or text


Best for training clinical
reasoning/decision
-
making
skills
(Cook & Triola 2009)

Picture/video

of patient


Navigation
Menu


Develop high

fidelity,
interactive

VP


Realistic appearance (3D, animated, full body, non
-
verbal
behavior)


Speech recognition


Natural, conversational capability


Animated facial expressions, gestures


Adaptive responses


Emotion Detection



Develop training related to communication skill
performance


Evaluate learner performance


Provide constructive feedback



Speech
Recognition


Signal
Processing
(tone,
inflection)


Key word
Processing
(learning
algorithm)


Evaluation/
Coding of
Context in
Communication

Model


Selection of
Responses
(emotion, non
-
verbal, verbal


Communication

Analysis for
Learner
Feedback


VP Output


Learning
Objective,
Scenario
Development


INPUT


Multidisciplinary team of subject
-
matter
experts and experienced clinicians


Extensive literature review


Observation of real SP training and
performance for iterative VP design


Prototype VP


Most software is Freeware


Speech recognition


Script matching based on keyword in user query


Randomly selects 1 of 3 responses to each question


Develop rubric for performance evaluation




Communication models



Cognitive
-
Affective Model of Organizational
Communication Systems (CAMOCS)



Roter Interaction Analysis System (RIAS)


Communication
Complexity

Inputs

task, distance,
values, norms

Process
g
oals, media,
strategies,
message form

Impact
mutual
understanding,
relationship


Extensive
research article
cites 301
sources in
business
organizational
communication


Main factors of
communication
complexity:
inputs,
process,
impact


Commonly
used measurement framework of
healthcare
communication


Classifies
task
-
focused communication and social
-
emotional
communication


Coding scheme for video
or audio of physician
-
patient
interactions


Utterances divided into >40 classifications, plus 12
global dimensions
of socio
-
emotional affect



(
D. Roter, 2006; D. Roter & Larson, 2002).

Personal remarks, social
conversation

Laughs, tells jokes

Shows approval

Gives compliment

Shows agreement or
understanding

Empathy

Shows concern or worry

Shows concern or worry

Reassures, encourages
or shows optimism

Legitimizes

Partnership

Self
-
Disclosure

Shows disapproval

Shows criticism

Asks for reassurance


Transition words

Gives orientation,
instructions

Paraphrase/Checks for
understanding

Bid for repetition

Requests for services

Asks for understanding

Asks for opinion

Asks questions


Closed/open
-
ended


Medical condition


Therapy


Lifestyle


Psychosocial
-
Feelings

Gives information

Counsels or directs
behavior


Design
the

system
to support the
needed impact, goals, strategies,
media characteristics
, inputs and
learning outcomes.


Representative case of Mr. Y and Dr. X:


65
-
year
-
old white male with no
significant past medical history

Coughing for 3 months (no fever,
infection, chills)

Former smoker

Possible mass on chest x
-
ray

Analysis Components

Affective distance

Adjusting to feedback

Interactivity

Tasks

Shared understanding

Contextualized content

Explicit directions

Goals

Cognitive distance

Adjusting to feedback in
communicating a difficult
diagnosis



Physician must be sensitive to body
language and patient’s reactions to
moderate how much information is
delivered in the initial diagnosis
.

For example, if Mr. Y dismisses the
urgency of the news, Dr. X may give
a more explicit explanation of why
these tests are needed and why the
timing of them is important.

Analysis Components

Affective distance

Adjusting to feedback

Interactivity

Tasks

Shared understanding

Contextualized content

Explicit directions

Goals

Cognitive distance

Shared understanding


Need shared knowledge between
participants to improve dialogue


Contextualized, explicit content

The surgeon may want to explicitly present
treatment options, with various risks and
percentages associated with them.

Analysis Components

Affective distance

Adjusting to feedback

Interactivity

Tasks

Shared understanding

Contextualized content

Explicit directions

Goals

Cognitive distance

Message Goal



In a follow
-
up appointment, oncologist
discovers that Mr. Y has not been getting
all of his chemo pills; Mrs. Y had
postponed a couple of doses because it
was making her husband too sick.


When physicians want to instruct or
influence difficult patients, they may
want to use highly formal language with
explicit instructions so that they can
better convey the importance to the
patient of a particular course of
treatment.


Analysis Components

Affective distance

Adjusting to feedback

Interactivity

Tasks

Shared understanding

Contextualized content

Explicit directions

Goals

Cognitive distance

Cognitive distance



A physician explaining a complex
diagnosis to a patient with limited
medical understanding will require
more explicit explanations, more
formal information, and probably
multiple methods of presenting
information (visual, verbal) for the
patient to get then and to reference
later.

Analysis Components

Affective distance

Adjusting to feedback

Interactivity

Tasks

Shared understanding

Contextualized content

Explicit directions

Goals

Cognitive distance


Conduct study: comparison of training with SP
alone to training with VP and SP


Measures: same used to evaluate performance
using SP


Move from prototype to build a VP in a gaming
environment with more realistic non
-
verbal
movements


Development of the virtual human is being
created in an Army project to develop learning
for cross
-
cultural competencies focusing on non
-
verbal behaviors




Prototype Proof of Concept



Haptek

SDK and body models from
Haptek


Free speech recognition


Microsoft Speech


Free synthetic speech generation


JAVA




New System Under Development



Unreal Tournament SDK game engine for virtual
environment


Stereoscopic 3D display


Maya 3D object editing software for body and object
creation


FaceFX

for visual expressions and matching speech
phonemes with mouth movements.


Custom creation of different looks and custom developed
facial action movements not available in
FaceFX
.


Natural
Speaking Professional
for speech recognition.


Ipisoft

and
Playstation

video cameras (6) for creating natural
body movements into characters.



Future adds


Learning software development for interpreting
speech and providing feedback vs discrete
scripted feedback.


Measures of learner interaction


Eye tracking (when not using 3D stereo)


Face tracking


Speech context


Emotion detection (facial and verbal)


Thank You!


Questions?


Accreditation Council for Graduate Medical Education (ACGME) (2005). Advancing Education in Interpersonal and
Communication Skills: An educational resource from the ACGME Outcome Project. Retrieved from
http://www.acgme.org/outcome/implement/interperComSkills.pdf
.


Association of American Medical Colleges. (1999).
Contemporary Issues in Medicine: Communication in Medicine
.
Report 3 of the Medical School Objectives Project. Washington, DC


Cook, D.A. &
Triola
, M. M. (2009). Virtual patients: a critical literature review and proposed next steps.
Medical
Education
,
43
(4), 303
-
311.



Stone, M. &
Silen
, W.
A 50 year
-
old woman with Lower Abdominal Pain.
Retrieved from
http://
research.bidmc.harvard.edu/VPtutorials/default.htm
.


Issenberg
,

S.B.,
McGaghie
, W.C.,
Petrusa
, E.R., Gordon, D.L. &
Scalese
, R.J. (2005). Features and uses of high
-
fidelity
medical simulations that lead to effective learning: a BEME systematic review*.
Medical Teacher
,
27
(1), 10
-
28.


Makoul

, G. (2001). Essential elements of communication in medical encounters: the Kalamazoo consensus statement.
Academic Medicine
. 76:390
-
393.


Paul, D. L. (2006). Collaborative activities in virtual settings: A knowledge management perspective of telemedicine.

Journal of Management Information Systems, 22
(4), 143
-
176.


Roter, D., & Larson, S. (2002). The Roter Interaction Analysis System (RIAS): Utility and flexibility for analysis of
medical interactions.

Patient Education and Counseling, 46
(4), 243
-
251.


Roter, D. (2006).
The Roter Method of Interaction Process Analysis.

Retrieved May 1, 2009, from
http://
rias.org/manual.pdf



Smothers, V., Azan, B.,
Ellaway
, R.(2010).
MedBiquitous

Virtual Patient Specifications and Description Document.
Retrieved from
http://www.medbiq.org/working_groups/virtual_patient/VirtualPatientDataSpecification.pdf



Stone, M. &
Silen
, W.
A 50 year
-
old woman with Lower Abdominal Pain.
Retrieved from
http://
research.bidmc.harvard.edu/VPtutorials/default.htm
.


Te'eni
, D. (2001). Review: A cognitive
-
affective model of organizational communication for designing IT.

MIS Quarterly,
25
(2), 251
-
312.


Toussaint
, P.,
Verhoef
, J.,
Vliet

Vlieland
, T., &
Zwetsloot
-
Schonk
, J. (2004).
The impact of ICT on communication in
healthcare.

Proceedings of MEDINFO’04