Evaluating the Performance of Artificial Intelligence-Based Large Language Models in Orthodontics—A Systematic Review and Meta-Analysis
Farraj Albalawi,
No information about this author
Sanjeev B. Khanagar,
No information about this author
Kiran Iyer
No information about this author
et al.
Applied Sciences,
Journal Year:
2025,
Volume and Issue:
15(2), P. 893 - 893
Published: Jan. 17, 2025
Background:
In
recent
years,
there
has
been
remarkable
growth
in
AI-based
applications
healthcare,
with
a
significant
breakthrough
marked
by
the
launch
of
large
language
models
(LLMs)
such
as
ChatGPT
and
Google
Bard.
Patients
health
professional
students
commonly
utilize
these
due
to
their
accessibility.
The
increasing
use
LLMs
healthcare
necessitates
an
evaluation
ability
generate
accurate
reliable
responses.
Objective:
This
study
assessed
performance
answering
orthodontic-related
queries
through
systematic
review
meta-analysis.
Methods:
A
comprehensive
search
PubMed,
Web
Science,
Embase,
Scopus,
Scholar
was
conducted
up
31
October
2024.
quality
included
studies
evaluated
using
Prediction
model
Risk
Bias
Assessment
Tool
(PROBAST),
R
Studio
software
(Version
4.4.0)
employed
for
meta-analysis
heterogeneity
assessment.
Results:
Out
278
retrieved
articles,
10
were
included.
most
used
LLM
(10/10,
100%
papers),
followed
Google’s
Bard/Gemini
(3/10,
30%
Microsoft’s
Bing/Copilot
AI
(2/10,
20%
papers).
Accuracy
primarily
Likert
scales,
while
DISCERN
tool
frequently
applied
reliability
indicated
that
LLMs,
ChatGPT-4
other
models,
do
not
significantly
differ
generating
responses
related
specialty
orthodontics.
forest
plot
revealed
Standard
Mean
Deviation
0.01
[CI:
0.42–0.44].
No
observed
between
experimental
group
(ChatGPT-3.5,
Gemini,
Copilot)
control
(ChatGPT-4).
However,
exhibited
high
PROBAST
risk
bias
lack
standardized
tools.
Conclusions:
extensively
variety
tasks
demonstrated
advanced
encouraging
outcomes
compared
thus
can
be
regarded
valuable
enhancing
educational
learning
experiences.
While
responses,
is
compromised
absence
peer-reviewed
references,
necessitating
expert
oversight
applications.
Language: Английский
Automated Age Estimation from OPG Images and Patient Records Using Deep Feature Extraction and Modified Genetic–Random Forest
Diagnostics,
Journal Year:
2025,
Volume and Issue:
15(3), P. 314 - 314
Published: Jan. 29, 2025
Background/Objectives:
Dental
age
estimation
is
a
vital
component
of
forensic
science,
helping
to
determine
the
identity
and
actual
an
individual.
However,
its
effectiveness
challenged
by
methodological
variability
biological
differences
between
individuals.
Therefore,
overcome
drawbacks
such
as
dependence
on
manual
measurements,
requiring
lot
time
effort,
difficulty
routine
clinical
application
due
large
sample
sizes,
we
aimed
automatically
estimate
tooth
from
panoramic
radiographs
(OPGs)
using
artificial
intelligence
(AI)
algorithms.
Methods:
Two-Dimensional
Deep
Convolutional
Neural
Network
(2D-DCNN)
One-Dimensional
(1D-DCNN)
techniques
were
used
extract
features
patient
records.
To
perform
feature
information,
Genetic
algorithm
(GA)
Random
Forest
(RF)
modified,
combined,
defined
Modified
Genetic–Random
Algorithm
(MG-RF).
The
performance
system
in
our
study
was
analyzed
based
MSE,
MAE,
RMSE,
R2
values
calculated
during
implementation
code.
Results:
As
result
applied
algorithms,
MSE
value
0.00027,
MAE
0.0079,
RMSE
0.0888,
score
0.999.
Conclusions:
findings
indicate
that
AI-based
employed
herein
effective
tool
for
detection.
Consequently,
propose
this
technology
could
be
utilized
sciences
future.
Language: Английский
The Applicability of the Demirjian and Willems Standards to Age Estimation of 6–9-Year-Old Portuguese Children
Humans,
Journal Year:
2025,
Volume and Issue:
5(1), P. 6 - 6
Published: Feb. 28, 2025
This
study
evaluates
the
applicability
of
Demirjian
and
Willems’
methods
for
age
estimation
in
Portuguese
children
aged
6–9
years
based
on
orthopantomographs
(OPGs).
The
main
objective
was
to
compare
precision
both
estimating
chronological
(CA).
analyzed
160
OPGs,
equally
distributed
by
sex,
dental
(DA)
calculated
twice,
using
methodologies.
findings
reveal
that
Demirjian’s
method
consistently
overestimated
an
average
1.47
males
1.45
females.
Similarly,
Willems
also
but
a
lesser
extent,
with
mean
differences
1.18
0.91
Statistical
analysis
confirmed
significantly
overestimate
age,
most
considerable
discrepancies
observed
8-year-old
individuals.
Despite
providing
slightly
more
accurate
results,
neither
reliable,
particularly
male
subjects.
highlights
need
further
refinement
these
methods,
considering
their
tendency
especially
specific
groups.
research
improves
techniques
forensic
clinical
settings,
within
pediatric
population.
Language: Английский
Künstliche Intelligenz in der forensisch-radiologischen Altersdiagnostik
Rechtsmedizin,
Journal Year:
2025,
Volume and Issue:
unknown
Published: Feb. 27, 2025
Efficacy of artificial intelligence in radiographic dental age estimation of patients undergoing dental maturation: A systematic review and meta-analysis
International Orthodontics,
Journal Year:
2025,
Volume and Issue:
23(4), P. 101010 - 101010
Published: May 2, 2025
Language: Английский
Insights into dental age estimation: introducing multiple regression data from a Black South African population on modified gustafson’s criteria
Fabian Rudolphi,
No information about this author
L. Steffens,
No information about this author
D.E. Shay
No information about this author
et al.
International Journal of Legal Medicine,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Aug. 22, 2024
Dental
Age
Estimation
(DAE)
is
an
effective
instrument
of
the
rule
law
for
verifying
dubious
age
claims
in
living
individuals.
Once
tooth
development
complete,
only
degenerative
dental
characteristics
can
be
used
this
purpose.
The
influence
ethnicity
on
these
has
not
been
clarified.Degenerative
changes
were
examined
using
modified
Gustafson's
criteria
including
secondary
dentin
formation,
cementum
apposition,
periodontal
recession
and
attrition
Olze
et
al.
(2012)
staging
scales.
Orthopantomograms
1882
black
South
Africans,
consisting
934
females
948
males,
from
12.00
to
40.96
years
chronological
utilized.
Two
independent
examiners
performed
evaluations,
with
one
two
evaluating
all
radiographs
twice.The
relationship
between
individual
was
analyzed
multiple
regression
analysis
as
dependent
variable.
resulting
R
Language: Английский