Professional Documents
Culture Documents
Proposal
Justice
Amoh
Prof.
Andrew
Campbell
COSC
169
Mobile
Health
January
21st,
2015
Amoh 1
Each
year,
1.8
million
emergency
department
visits
in
the
United
States
alone
result
from
episodes
of
asthma.1
Even
though
self-management
plans
have
been
shown
to
reduce
these
episodes
by
as
much
as
40%,2
patients
rarely
adhere
to
these
plans
and
fail
to
manage
their
asthma.
For
a
patient,
self-management
regimes
would
involve
manually
recording
one's
symptoms
such
as
coughs
and
wheezes
in
special
journals
called
asthma
diaries.
In
addition,
management
plans
require
patients
to
follow
a
predetermined
action
plan
to
address
exacerbations.
In
practice,
however,
laborious
manual
logging
and
human
factors
such
forgetfulness
deter
patients
from
using
asthma
diaries
and
action
plans
effectively.
So
for
my
COSC
169
project,
I
propose
to
develop
a
smartphone
app
that
employs
embedded
smartphone
sensors
to
track
asthma
symptoms
and
inform
patients
on
their
asthma
conditions.
In
devising
such
an
app,
we
can
consider
two
categories
of
asthma
symptoms
to
monitor.
First,
there
are
those
symptoms
we
can
sense
using
the
standard
inbuilt
sensors
of
the
smartphone.
Cough
symptoms
can
be
detected
from
the
embedded
microphone's
audio
stream
using
Hidden
Markov
Model
classiUiers.3
Larson
et
al
have
demonstrated
that
the
same
microphone
can
be
also
be
used
to
measure
lung
function
at
accuracies
comparable
to
conventional
spirometers.4
In
addition,
wheeze
and
dyspnea(shortness
of
breadth)
symptoms
can
be
detected
by
similar
classiUication
and
acoustic
processing
techniques,
although
they
are
much
quieter
than
coughs
and
might
require
the
use
of
microphones
closer
to
patient's
mouth
like
those
in
headphones
and
headsets.
Besides
these
acoustically
detectable
symptoms,
there
are
also
the
more
subjective
symptoms
that
cannot
be
measured
through
sensing.
Fever,
chest
tightness
and
frequency
of
medication
use
are
examples
of
such
symptoms
and
these
can
be
monitored
through
Ecological
Momentary
Assessment
(EMA)
strategies.
Apart
from
symptom
detection
and
logging,
the
proposed
mHealth
app
should
also
inform
patients
on
their
current
asthma
conditions
and
advise
on
necessary
actions
just
as
asthma
diaries
and
action
plans
do.
A
key
speciUication
of
the
app
user
interface
is
that
it
reports
on
how
well
patient
symptoms
are
being
managed
using
standard
color
codes
as
seen
in
asthma
diaries.
It
must
also
be
able
to
provide,
in
a
simple
readable
format,
a
history
of
symptoms
and
any
interesting
patterns
in
symptom
occurrence.
Additional
features
like
a
panic
or
call-my-doctor
button
and
location-based
information
on
asthma
triggers
(pollen
count,
weather,
humidity,
etc.)
can
also
be
very
helpful
to
the
asthma
patient.
In
all,
the
app
ought
to
have
a
very
minimal
yet
functional
and
informative
UI,
such
that
a
user
can
easily
keep
themselves
updated
on
their
asthma
status.
1
Ginde AA, Espinola JA, Camargo Jr CA. Improved overall trends but persistent racial disparities in emergency department
visits for acute asthma, 1993-2005. J Allergy Clin Immunol 2008;122(2):313-318.
2
P.G. Gibson, H. Powell, J. Coughlan, A.J. Wilson, M. Abramson, P. Haywood, A Bauman, MJ Hensley, and EH Walters. Selfmanagement education and regular practitioner review for adults with asthma. Cochrane Database Syst Rev, 3, 2002
3 Amoh,
Justice, and Kofi Odame. "Technologies for Developing Ambulatory Cough Monitoring Devices." Critical Reviews in
Biomedical Engineering 41.6 (2014).
4
Larson, Eric C., et al. "Spirosmart: Using a microphone to measure lung function on a mobile phone." Proceedings of the 2012
ACM Conference on Ubiquitous Computing. ACM, 2012.