Procedia Computer Science,
Journal Year:
2022,
Volume and Issue:
210, P. 248 - 253
Published: Jan. 1, 2022
In
recent
years,
hospitals
around
the
world
are
faced
with
large
patient
flows,
which
negatively
affect
quality
of
care
and
become
a
crucial
factor
to
consider
in
inpatient
management.
The
main
objective
this
management
is
maximize
number
available
beds,
using
efficient
planning.
Intensive
Care
Units
(ICU)
hospital
units
higher
monetary
consumption,
importance
indicators
that
allow
achievement
useful
information
for
correct
critical.
This
study
allowed
prediction
Length
Stay
(LOS)
based
on
their
demographic
data,
collected
at
time
admission
clinical
conditions,
can
help
health
professionals
conducting
more
assertive
planning
better
service.
results
obtained
show
Machine
Learning
(ML)
models,
simultaneously
patient's
pathway,
as
well
drugs,
tests
analysis,
introduce
greater
predictive
ability
LOS.
Big Data and Cognitive Computing,
Journal Year:
2024,
Volume and Issue:
8(12), P. 178 - 178
Published: Dec. 3, 2024
Hospital
overcrowding,
driven
by
both
structural
management
challenges
and
widespread
medical
emergencies,
has
prompted
extensive
research
into
machine
learning
(ML)
solutions
for
predicting
patient
length
of
stay
(LOS)
to
optimize
bed
allocation.
While
many
existing
models
simplify
the
LOS
prediction
problem
a
classification
task,
broad
ranges
hospital
days,
an
exact
day-based
regression
model
is
often
crucial
precise
planning.
Additionally,
available
data
are
typically
limited
heterogeneous,
collected
from
small
cohort.
To
address
these
challenges,
we
present
novel
multimodal
ML
framework
that
combines
imaging
clinical
enhance
accuracy.
Specifically,
our
approach
uses
following:
(i)
feature
extraction
chest
CT
scans
via
convolutional
neural
network
(CNN),
(ii)
their
integration
with
clinically
relevant
tabular
exams,
refined
through
selection
system
retain
only
significant
predictors.
As
case
study,
applied
this
pneumonia
during
COVID-19
pandemic
at
two
hospitals
in
Naples,
Italy—one
specializing
infectious
diseases
other
general-purpose.
Under
experimental
setup,
proposed
achieved
average
error
three
demonstrating
its
potential
improve
flow
critical
care
environments.
Hospital
readmission
and
length
of
stay
prediction
provide
info
to
manage
hospitals’
bed
capacity
the
number
required
staff,
especially
during
pandemics.
We
present
a
hybrid
deep
model
called
Genetic
Algorithm-Optimized
Convolutional
Neural
Network
(GAOCNN)
with
unique
preprocessing
method
predict
hospital
in
patients
having
various
conditions.
GAOCNN
uses
one-dimensional
convolutional
layers
stay.
The
parameters
are
optimized
using
genetic
algorithm.
To
show
performance
proposed
conditions,
we
evaluate
under
three
healthcare
datasets;
Diabetes
130-US
hospitals
dataset,
COVID-19
MIMIC-III
dataset.
diabetes
dataset
has
information
on
both
stay,
while
datasets
just
include
Experimental
results
that
model’s
accuracy
for
is
97.2%
diabetic
patients.
Also,
89%,
99.4%,
94.1%
diabetic,
COVID-19,
ICU
patients,
respectively.
These
confirm
superiority
compared
existing
methods.
Our
findings
offer
platform
managing
funds
resources
diseases.
Medical Records,
Journal Year:
2023,
Volume and Issue:
5(3), P. 500 - 6
Published: July 13, 2023
Aim:
The
aim
of
this
study
is
to
utilize
machine
learning
techniques
accurately
predict
the
length
stay
for
Covid-19
patients,
based
on
basic
clinical
parameters.
Material
and
Methods:
examined
seven
key
variables,
namely
age,
gender,
hospitalization,
c-reactive
protein,
ferritin,
lymphocyte
count,
COVID-19
Reporting
Data
System
(CORADS),
in
a
cohort
118
adult
patients
who
were
admitted
hospital
with
diagnosis
during
period
November
2020
January
2021.
data
set
partitioned
into
training
validation
comprising
80%
test
20%
random
manner.
present
employed
caret
package
R
programming
language
develop
models
aimed
at
predicting
(short
or
long)
given
context.
performance
metrics
these
were
subsequently
documented.
Results:
k-nearest
neighbor
model
produced
best
results
among
various
models.
As
per
model,
evaluation
outcomes
estimation
hospitalizations
lasting
5
days
less
those
exceeding
are
as
follows:
accuracy
rate
was
0.92
(95%
CI,
0.73-0.99),
no-information
0.67,
Kappa
0.82,
F1
score
0.89
(p=0.0048).
Conclusion:
By
applying
Covid-19,
estimates
can
be
made
more
accuracy,
allowing
effective
patient
management.
Procedia Computer Science,
Journal Year:
2022,
Volume and Issue:
210, P. 248 - 253
Published: Jan. 1, 2022
In
recent
years,
hospitals
around
the
world
are
faced
with
large
patient
flows,
which
negatively
affect
quality
of
care
and
become
a
crucial
factor
to
consider
in
inpatient
management.
The
main
objective
this
management
is
maximize
number
available
beds,
using
efficient
planning.
Intensive
Care
Units
(ICU)
hospital
units
higher
monetary
consumption,
importance
indicators
that
allow
achievement
useful
information
for
correct
critical.
This
study
allowed
prediction
Length
Stay
(LOS)
based
on
their
demographic
data,
collected
at
time
admission
clinical
conditions,
can
help
health
professionals
conducting
more
assertive
planning
better
service.
results
obtained
show
Machine
Learning
(ML)
models,
simultaneously
patient's
pathway,
as
well
drugs,
tests
analysis,
introduce
greater
predictive
ability
LOS.
