Journal of Food Protection,
Journal Year:
2024,
Volume and Issue:
87(12), P. 100396 - 100396
Published: Nov. 8, 2024
Conventional
detection
methods
require
the
isolation
and
enrichment
of
bacteria,
followed
by
molecular,
biochemical,
or
culture-based
analysis.
To
address
some
limitations
conventional
methods,
this
study
develops
a
machine
learning
(ML)
approach
to
analyze
excitation-emission
matrix
(EEM)
fluorescence
data
generated
based
on
bacteriophage
T7
Escherichia
coli
interactions
for
in-situ
live
bacteria
in
presence
fresh
produce
homogenate.
We
trained
classification
models
using
various
ML
algorithms
3-D
EEM
with
their
phage.
These
algorithms,
including
linear
Support
Vector
Classifier
(SVC)
Random
Forest
(RF),
demonstrate
high
accuracy
(>0.85)
detecting
E.
at
10
Analytical Chemistry Letters,
Journal Year:
2024,
Volume and Issue:
14(1), P. 1 - 28
Published: Jan. 2, 2024
The
present
expeditious
advancement
of
green
analytical
chemistry
(GAC)
necessitates
the
establishment
explicit
and
succinct
GAC
principles,
which
may
serve
as
valuable
guidance
in
adoption
environment
friendly
laboratory
practises.
current
ideas
engineering
need
modification
to
effectively
address
requirements
within
context
GAC.
use
multivariate
curve
resolution
(MCR),
parallel
factor
analysis
(PARAFAC),
self-weighted
alternating
trilinear
decomposition
(SWATLD),
unfolded
partial
least
squares/residual
bi-linearization
(UPLS/RBL)
are
prevalent
approaches
for
examination
process
data
across
several
application
domains
like
detect
contaminants
water
samples.
A
special
emphasis
was
placed
on
circumstances
that
necessitate
sophisticated
customised
implementations
resolution.
This
will
involve
addressing
enhancements
pre-processing
techniques,
arrangements
from
multiple
sets,
constraints,
challenges
associated
with
non-ideal
noise
structure,
deviations
linearity.
study
furthermore
covers
a
thorough
case
studies
new
developments
discipline,
highlighting
efficacy
identification
pharmaceutical
substances
wastewater.
paper
examines
methodologies,
instrumental
analysis,
algorithms,
MCR
methods,
set
configurations,
separation
techniques
practical
applications
resource
minimization
sustainability.
Foods,
Journal Year:
2024,
Volume and Issue:
13(9), P. 1376 - 1376
Published: April 29, 2024
With
a
view
to
maintaining
the
reputation
of
wine-producing
regions
among
consumers,
minimising
economic
losses
caused
by
wine
fraud,
and
achieving
purpose
data-driven
terroir
classification,
use
an
absorbance–transmission
fluorescence
excitation–emission
matrix
(A-TEEM)
technique
has
shown
great
potential
based
on
molecular
fingerprinting
sample.
The
effects
changes
in
composition
due
ageing
stability
A-TEEM
models
over
time
had
not
been
addressed,
however,
classification
blends
required
investigation.
Thus,
data
were
combined
with
extreme
gradient
boosting
discriminant
analysis
(XGBDA)
algorithm
build
range
Shiraz
research
wines
(n
=
217)
from
five
Barossa
Valley
sub-regions
four
vintages
that
aged
bottle
for
several
years.
This
spectral
machine
learning
approach
revealed
100%
class
prediction
accuracy
cross-validation
(CV)
model
results
vintage
year
98.8%
unknown
sample
when
splitting
samples
into
training
test
sets
obtain
models.
modelling
sub-regional
production
area
showed
CV
99.5%
93.8%
split
dataset.
Inputting
sub-set
current
generated
previously
these
sub-region
yielded
accurate
2018–2020
wines,
92%
2018
91%
using
2021
included
original
modelling.
Satisfactory
also
obtained
blended
sub-regions,
which
is
significance
considering
practice
blending.
Foods,
Journal Year:
2024,
Volume and Issue:
13(20), P. 3303 - 3303
Published: Oct. 18, 2024
At
a
time
when
the
botanical
origin
of
honey
is
being
increasingly
falsified,
there
need
to
find
quick,
cheap
and
simple
method
identifying
its
origin.
Therefore,
aim
our
work
was
show
that
fluorescence
spectrometry,
together
with
statistical
analysis,
can
be
such
method.
In
total,
108
representative
samples
10
different
botanic
origins
(9
unifloral
1
multifloral),
obtained
in
2020–2022
from
local
apiaries,
were
analyzed.
The
spectra
those
determined
using
F-7000
Hitachi
spectrophotometer,
Tokyo,
Japan.
It
shown
each
variety
produces
unique
emission
spectrum,
which
allows
for
determination
Taking
into
account
difficulties
analyzing
these
spectra,
it
found
most
information
regarding
differences
their
identification
provided
by
synchronous
cross-sections
at
Δλ
=
100
nm.
addition,
this
analysis
supported
discriminant
canonical
allowed
creation
mathematical
models,
allowing
correct
classification
type
(except
dandelion)
an
accuracy
over
80%.
application
universal
(in
accordance
methodology
described
paper),
but
use
requires
spectral
matrices
(EEG)
characteristic
given
geographical
Journal of Food Protection,
Journal Year:
2024,
Volume and Issue:
87(12), P. 100396 - 100396
Published: Nov. 8, 2024
Conventional
detection
methods
require
the
isolation
and
enrichment
of
bacteria,
followed
by
molecular,
biochemical,
or
culture-based
analysis.
To
address
some
limitations
conventional
methods,
this
study
develops
a
machine
learning
(ML)
approach
to
analyze
excitation-emission
matrix
(EEM)
fluorescence
data
generated
based
on
bacteriophage
T7
Escherichia
coli
interactions
for
in-situ
live
bacteria
in
presence
fresh
produce
homogenate.
We
trained
classification
models
using
various
ML
algorithms
3-D
EEM
with
their
phage.
These
algorithms,
including
linear
Support
Vector
Classifier
(SVC)
Random
Forest
(RF),
demonstrate
high
accuracy
(>0.85)
detecting
E.
at
10
