Engineering Strategies for Advancing Optical Signal Outputs in Smartphone‐Enabled Point‐of‐Care Diagnostics

Abstract

The use of smartphone‐based analysis systems has been increasing over the past few decades. Among the important reasons for its popularity are its ubiquity, increasing computing power, relatively low cost, and capability to acquire and process data simultaneously in a point‐of‐need fashion. Furthermore, smartphones are equipped with various sensors, especially a complementary metal–oxide–semiconductor (CMOS) sensor. The high sensitivity of the CMOS sensor allows smartphones to be used as a colorimeter, fluorimeter, and spectrometer, constituting the essential part of point‐of‐care testing contributing to E‐health and beyond. However, despite its myriads of merits, smartphone‐based diagnostic devices still face many challenges, including high susceptibility to illumination conditions, difficulty in adapter uniformization, low interphone repeatability, and et al. These problems may hinder smartphone‐enabled diagnosis from passing the FDA regulations of medical devices. This review discusses the design and application of current smartphone‐based diagnostic devices and highlights challenges associated with existent methods and perspectives on how to deal with those challenges from engineering aspects on constant color signal acquisition, including smartphone adapter design, color space transformation, machine learning classification, and color correction. This review discusses the design and application of current smartphone‐based diagnostic devices and highlights challenges associated with existent methods and perspectives on how to deal with those challenges from engineering aspects on the constant color signal acquisition, including smartphone adapter design, color space transformation, machine learning classification, and color correction.

Full text

Engineering Strategies for Advancing Optical Signal
Outputs in Smartphone-Enabled Point-of-Care Diagnostics
Kexin Fan, Weiran Liu, Yuchen Miao, Zhen Li, and Guozhen Liu*
1. Introduction
Point-of-care testing (POCT) has been flourishing in the past
decade due to its affordable, sensitive, specific, user-friendly,
rapid, equipment-free, delivered (ASSURED) properties.
[1]
It
enables untrained patients to monitor their
health conditions without the use of bulky
laboratory equipment and professional
techniques. The role of POCT is particu-
larly crucial and essential during a pan-
demic such as COVID-19 due to limited
medical resources.
[2]
The POCT devices
should be portable and easy to use.
[3]
However, conventional biosensing systems
in hospitals and laboratories are usually
cumbersome, including sophisticated
detection equipment and a computer as
an analyzer.
[4]
These instruments are
expensive and difficult to assemble, and
expert training is required to operate those
machines. Therefore, smartphone-based
diagnostics have shown great promise in
POCT based on their embedded features,
such as high portability, compacity, versa-
tility, and ability of data processing to real-
ize the digital health.
[5]
In 2021, 80.63% of
the world’s population own a smart-
phone.
[6]
It is predicted that in 2025, there
will be 7.33 billion smartphone users in the
world.
[7]
The universality makes smartphones the most suitable
platform for POCT. Moreover, numerous sensors have been inte-
grated in smartphones, such as magnetic field sensors,
gyroscopes, high-resolution complementary metal–oxide–
semiconductor (CMOS) image sensors, global positioning
system sensors, accelerometers, gyroscopes, barometers, prox-
imity sensors, and so on, making it not only a communication
tool, but also versatile biosensing platforms.
[8]
The ability of its
wireless connection, including Bluetooth, near-field sensors,
wireless fidelity, and 3 G/4 G/5 G, allows data transfer from
POCT devices to smartphones or from smartphones to the inter-
net.
[9]
The technology of cloud computing and big data facilitates
the fast and precise analysis of health data.
[10]
In addition, doctors
can access patient health data online or remotely and give sug-
gestions without meeting their patients. The smartphone-based
diagnosis has significantly eased the pressure on hospital facili-
ties and resources by having patients do tests/screening at home.
Among all of the sensors integrated in the smartphone, CMOS
image sensors are the most applicated sensors in smartphone-
based diagnostic devices. It has been hypothesized that the pixel
count of smartphone cameras follows Moore’s law, which means
that it doubles every two years.
[11]
As a result of its powerful abil-
ity to record light signals, the CMOS sensors on smartphones
may act as an alternative to conventional bulky optical biosensing
equipment and allow on-site measurement. A wide range of opti-
cal biosensing techniques, including colorimetry, fluorimetry,
K. Fan, W. Liu, Y. Miao, G. Liu
School of Medicine
The Chinese University of Hong Kong
Shenzhen 518172, China
E-mail: liuguozhen@cuhk.edu.cn
K. Fan, W. Liu, Y. Miao, G. Liu
Ciechanover Institute of Precision and Regenerative Medicine
The Chinese University of Hong Kong
Shenzhen 518172, China
Z. Li
School of Science and Engineering
The Chinese University of Hong Kong
Shenzhen 518172, China
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/aisy.202200285.
© 2022 The Authors. Advanced Intelligent Systems published by Wiley-
VCH GmbH. This is an open access article under the terms of the
Creative Commons Attribution License, which permits use, distribution
and reproduction in any medium, provided the original work is
properly cited.
DOI: 10.1002/aisy.202200285
The use of smartphone-based analysis systems has been increasing over the past few
decades. Among the important reasons for its popularity are its ubiquity, increasing
computing power, relatively low cost, and capability to acquire and process data
simultaneously in a point-of-need fashion. Furthermore, smartphones are equipped
with various sensors, especially a complementary metal–oxide–semiconductor
(CMOS) sensor. The high sensitivity of the CMOS sensor allows smartphones to be
used as a colorimeter, fluorimeter, and spectrometer, constituting the essential part
of point-of-care testing contributing to E-health and beyond. However, despite its
myriads of merits, smartphone-based diagnostic devices still face many challenges,
including high susceptibility to illumination conditions, difficulty in adapter uni-
formization, low interphone repeatability, and et al. These problems may hinder
smartphone-enabled diagnosis from passing the FDA regulations of medical devices.
This review discusses the design and application of current smartphone-based
diagnostic devices and highlights challenges associated with existent methods and
perspectives on how to deal with those challenges from engineering aspects on
constant color signal acquisition, including smartphone adapter design, color space
transformation, machine learning classification, and color correction.
REVIEW
www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (1 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

and spectrometry, has been utilized to detect enormous
biomarkers.
[12]
The colorimetric and fluorescent biosensing
can be realized in gas phase,
[13]
liquid phase,
[14]
or on lateral flow
assay (LFA) strips,
[15]
cellulose test paper,
[16]
microfluidic
chips
[17]
and so on. The color information of images captured
by smartphones can be analyzed instantly using a predesigned
smartphone application (APP) or software/program on the
computer after data transfer.
[18]
However, the constancy of color information is difficult to
achieve for smartphone-based optical detection considering its
high susceptibility to illumination conditions,
[18]
difficulty in
adapter uniformization,
[19]
and low interphone repeatability.
[20]
The constant and uniform illumination conditions can be main-
tained by attaching extra devices with independent light sources
to smartphones. However, the adapters are difficult to be applied
to all types of smartphones considering their different shapes
and sizes, especially for those of fluorescent purposes. The
requirement of having to be lightproof restricts the flexibility
of adapters.
[21]
In addition, photos taken by smartphones of dif-
ferent models could show significantly different colors due to dif-
ferent postprocessing and inherent differences in the spectral
sensitivity of the camera.
[22]
The raw signals captured by
CMOS sensors using a color filter array (CFA) are processed non-
linearly in the process of color space transformation (CST) and
lossy compression before storing in the memory, increasing the
difficulty of color calibration.
[23]
Image autocorrections such as
automatic exposure correction, white balance (WB), tune manip-
ulation, brilliance adjustment, saturation adjustment, and con-
trast enhancement can highly affect the color values.
[24]
Moreover, the color of photos taken by the same smartphone
varies as well with different shooting conditions, including shoot-
ing distance, imaging angle, camera settings, such as
International Standardization Organization (representing sensi-
tivity or brightness), exposure time, aperture, etc. One strategy to
solve these problems is to generate a calibration curve for each
type of smartphone camera with fixed settings and capture the
assay signals with exactly the same settings. However, this pro-
cedure is apparently inconvenient. Some researchers circum-
vented accurate color acquisition by developing count-based
[25]
or distance-based
[26]
biosensors. Unfortunately, not all types of
assays can be adapted to these forms. It is still necessary to
achieve accurate color acquisition. These challenges associated
with smartphone-based POCT have attracted researchers’wide
attention.
There have been many reviews on smartphone-based POCT
systems. González et al.
[27]
provided an overview of adapter
design in different types of biosensors. Liu et al.
[28]
summarized
the features of biosensors (paper-based sensor, flexible device,
microfluidic chip, et al.) currently widely used in smartphone-
based POCT. Nonno et al.
[9]
investigated different classes of
smartphone-based devices (smartphone-based colorimeters,
photo- and spectrometers, and fluorimeters). Sun et al.
[29]
sum-
marized the development and application of mobile APPs for
smartphone-based POCT. Kordasht et al.
[30]
summarized differ-
ent types of smartphone-based immunosensors based on electro-
chemical, colorimetric, and optical techniques. Kap et al.
[31]
analyzed different smartphone-based colorimetric detection sys-
tems for glucose monitoring. Yang et al.
[32]
investigated different
detection platforms based on colorimetric, luminescent, and
magnetic assays. Biswas et al.
[33]
summarized recent develop-
ments in smartphone spectrometer sample analysis. Qin
et al.
[34]
emphasized different signal analysis algorithms for lat-
eral flow immunoassay detection. Although most of the reviews
pointed out the current challenges of smartphone-based optical
diagnostics, no recent reviews provided a comprehensive analy-
sis and summary of existing strategies on solving the signal read-
out quality. This review emphasizes four engineering strategies
on the constant color signal acquisition, including smartphone
adapter design, CST, machine learning (ML), and color correc-
tion, as shown in Scheme 1.
2. Smartphone Adapter Design and Engineering
The color signal detected by the camera sensor is determined by
the product of irradiance, reflectance of imaging target, and the
spectral sensitivity of camera.
[35]
Although the adapters to smart-
phones for light shielding do not ensure the same spectral sen-
sitivity of camera sensors, they do guarantee the constancy of
Scheme 1. The schematics summarizing the contents of this review.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (2 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

irradiance and reflectance to a great extent with simple operation,
highly improving the performance of smartphone-based optical
biosensing.
2.1. Colorimetry
The conventional approach of colorimetric biosensing is to derive
a calibration curve based on a single or the combination of
multiple channels of a color model which leads to the highest
correlation between the color intensities and analyte concentra-
tions. Even though the calibration curve performs well in a con-
trolled environment, the color intensity values will be biased
under ambient light sources due to their high sensitivity to
the illumination sources.
[36]
Especially, when taking photos from
above, the smartphone or users can always block some light and
cast shadows on the sensing platform. Some researchers took
photos at a certain angle and allowed ambient light to come
through from one side.
[37]
However, the light intensities at
different places and different times will also be different, leading
to inaccurate results. To solve this problem, some researchers
use the ratio of color intensity in the test zone and control zone
or background region,
[38]
but the interference may not be
eliminated completely. Some other researchers constructed
the standard curve together with the measurement of samples
to ensure the same light conditions.
[39]
However, this will greatly
increase the complexity of the detection procedure. Therefore, to
make the measurement without constructing the standard curve,
an adapter that can block the light from surroundings with extra
light sources providing homogenous and constant light is often
used, so a standard curve can be prepared in advance and stored
in APP (Figure 1). Patients only need to take one photo and
results can be directly given. Although an extra device induces
inconvenience, using housing is the easiest way and ensures
high constancy of color intensities. To achieve a homogenous
light environment in the housing, an optical diffuser, such as
translucent glass, can be covered on the light source
(Figure 1a,b).
[40]
Or instead of irradiating samples directly,
reflected light from the wall of the housing is also used. For water
sample imaging, to avoid light reflectance from the water surface,
the light source can be placed under the water sample, and the
transmission of light is measured.
[41]
Or a polarizer can be placed
in front of the smartphone camera to reduce reflections from the
target surface.
[42]
The light sources can be light-emitting diodes
(LEDs), lasers, smartphone flashlights, electroluminescent
sheets, etc. They are usually powered by batteries or smartphones
through universal serial bus (USB)-on the go (OTG) connection
(Figure 1b).
[40a,43]
Interestingly, Iqbal et al.
[44]
used phone’s
screen as the light source and front-view camera as the detector,
so no extra light source is required. However, the emission spec-
trum of different smartphone screens is different, which may
induce variation in illumination conditions. This is the same case
as using the flashlight, which does not provide the same illumi-
nation for different smartphones and either does not ensure even
lighting.
[45]
The size of the adapter is primarily limited by the
long focus distance of smartphones, which can be reduced by
an extra biconvex lens (Figure 1a,b).
[46]
Fu et al.
[47]
developed
a very small cylinder-shaped adapter (Figure 1c). Instead of using
the camera at the back of the smartphone, they used the ambient
Figure 1. Illustration of smartphone-based colorimeters. a,i) Schematic of a smartphone adapter with components including lens and a light diffuser;
ii) The internal structure of the imaging stage with a light illumination pathway during a readout. Reproduced with permission.
[40b]
Copyright 2020, American
Chemical Society. b,i) Schematic of a smartphone adapter with components including lens and a light diffuser; ii) The attachment with the smartphone;the
sensor was connected to the smartphone through USB-OTG. Reproduced with permission.
[40a]
Copyright 2021, MDPI. c) Schematic of a smartphone
adapter using the ambient light sensor. Reproduced with permission.
[47]
Copyright 2016, Royal Society of Chemistry. d) Illustration of i. a reflection film
module and ii) a mirror surface module in the detection chamber. Reproduced with permission.
[52]
Copyright 2017, Springer Nature.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (3 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

light sensor to measure the light signals, so no extra convex lens
or diffuser was needed. Park et al.
[48]
also used the ambient light
sensor, and it was claimed that it could exclude the influence of
different illumination conditions. Overall, a generalized smart-
phone adapter for colorimetric biosensing includes a convex
lens, LED modules, an optical diffuser, a color checker (will
be mentioned in Section 5), and a holder for all of the compo-
nents. The adapter can be fixable to the smartphone, so the posi-
tion of the sensing platform relative to the smartphone camera is
defined, making it easier to target the detection signal. However,
this usually restricts the versatility of the adapter
(i.e., the adapter has to be a certain shape and size to fit certain
smartphones). Advanced adapters can be designed to be adjust-
able to fit different smartphone models and camera focal lengths.
There are other types of optical biosensors that are based on
luminescence, including chemiluminescence,
[49]
thermochemi-
luminescence,
[50]
and electrochemiluminescence.
[51]
The
housing design is similar to the aforementioned structures.
Because the light signal is emitted from the sample itself, no
extra light sources are needed. However, a completely lightproof
housing becomes crucial for detection. Moreover, to gather more
photons, a mirror or reflective materials can be coated in the
adapter (Figure 1d).
[50,52,53]
Longer exposure time of the camera
is often involved for luminescence detection.
[54]
The adapter can
be designed to be flexible, but this increases the cost and makes it
difficult to guarantee light tightness. Some studies used a simple
light box that allowed users to manually put the smartphone on
the box, although an additional automatic color-retrieving algo-
rithm is usually required
[55]
(Figure 2), or the test region needs
to be manually selected by users.
[39]
Some applications of smart-
phone-based colorimeters are listed in Table 1 and 2.
2.2. Fluorimetry
Fluorimeter measures the intensity of emitted light from a
fluorophore with excitation by a certain wavelength of light.
An extra light source and a lightproof housing are always
required. The major difference from the colorimeter is that
the excitation light source is placed perpendicular (Figure 3a)
or nearly perpendicular (Figure 3b,c) to the emitted light.
[56]
If
the sensing platform is 2D, such as a test paper, it will be difficult
to set a 90° angle between emission light and excitation light. To
solve this problem, a dichroic mirror, which allows light with cer-
tain wavelengths to pass through while reflecting light with other
wavelengths, is usually utilized to separate emission light and
excitation light (Figure 3d).
[57]
To reduce background signal
and increase detection accuracy, a light filter can be embedded
to block excitation light and only pass emission light to the
CMOS sensor.
[58]
The excitation light sources are usually
LEDs with certain wavelength or lasers, on the grounds that
Figure 2. Illustration of automatic color-retrieving results and algorithms. a) The color chart localization. Reproduced with permission.
[142]
Copyright
2020, IEEE. b) Circle shape test region localization. Reproduced with permission.
[145]
Copyright 2017, IEEE. c) Automatic color-retrieving algorithm:
i) Noise reduction and image simplification; ii) Edge detection; iii. Initial contour extraction; iv) Final contour labeling; v) Automatic position determi-
nation to obtain reference colors and target colors; vi) Retrieved reference colors and target colors. Reproduced with permission.
[55a]
Copyright 2017,
MDPI. d,i) Automatic color-retrieving algorithm; ii) Original image; iii) ROI extraction. Reproduced with permission.
[155]
Copyright 2019, MDPI.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (4 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

Table 1. Designs and applications of smartphone-based colorimeters.
Analyte[s] Sample Sample
platform
Light source Sensor Components Data
processing
Color
space
Analysis mode limit of detection (LOD)/
Detection range
Analysis
time
Merits Ref.
H
2
S N/A Gas Flashlight CMOS A plastic optical
fiber; a 3D-printed
connector socket
APP Gray Qualitative N/A 5 min Cost-effective,
scalable, and
compact
[161]
K
þ
Treated blood
plasma
Liquid A green LED CMOS A LED housing; a
Lens tube; an optical
diffuser; a bi-convex
lens; a glass-slide
assembly; a clamp
MATLAB
and C
þþ
HSV Quantitative 1.5–7.5 mM N/A Low cost [40a]
Altenariol
monomethyl ether
Cherry, apple
and orange
extract solution
Microfluidic
chip
N/A CMOS N/A APP HSL Quantitative 200 pg/mL <15 min Simple, real-time,
sensitive, and low
cost
[162]
Salmonella
typhimurium
Spiked fresh-cut
vegetables salad
Microfluidic chip An adjustable
LED
CMOS LED APP HSL Quantitative 6.0 10
1
cfu mL <45 min Simple, portable,
and low cost
[163]
Urinary C-
telopeptide
fragment of type II
collagen
Urine Paper Smartphone
flash
CMOS A prism; a channel
holder; a dispersion
film; a paper-based
signal guide; a cover
Sketch filter N/A Semiquantitative 0–10 ng mL <10 min Simple, low cost,
cost-effective,
high productivity,
and intuitive
[25a]
Glucose, cholesterol,
albumin, alkaline
phosphatase,
creatinine, aspartate
aminotransferase,
alanine
aminotransferase,
and urea nitrogen
PBS (3D)-μPADs A white light
bulb
CMOS A black cardboard
box; a white light
bulb
APP and ImageJ Gray Quantitative Glucose
(0–20 mmol L
1
), cholesterol
(0–10 mmol L
1
), albumin
(0–7gdL
1
), alkaline
phosphatase (0–800 U L
1
),
creatinine (0–500 μmol L
1
),
aspartate aminotransferase
(0–800 U L
1
), alanine
aminotransferase (0–1000
UL
1
), and urea nitrogen
(0–7.2 mmol L
1
).
3 min Multiple
detections
[164]
Carcinoembryonic
antigen
Serum Multilayered
μPADs
Smartphone
flashlight
CMOS N/A Program on
smartphone
RGB Quantitative 0.015 ng mL <1h Multi-image
processing
algorithm and
autonomous ROI
identification
[165]
Organophosphorus
pesticide
Apple-banana
juice
Hydrogel White light CMOS An iron box;
a light source;
heating pads;
tin foil
APP Gray Quantitative 0.5 ng mL 15 min High reliability
and stability
[166]
Nucleocapsid
phosphoprotein
Serum Liquid Laser (655 nm) Light sensor A black housing;
adjustable
resistance; a laser
source; a battery; an
ELISA plate; a light
sensor; USB
APP (iENV) Optical
density
Quantitative 10 pgmL
1
<1 h Cost-effective [167]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (5 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

Table 1. Continued.
Analyte[s] Sample Sample
platform
Light source Sensor Components Data
processing
Color
space
Analysis mode limit of detection (LOD)/
Detection range
Analysis
time
Merits Ref.
charging port; a
photometer
Glucose, nitrite,
protein, and
phenylpyruvate
Urine Tape LED CMOS Three-sided sealed
box; a 40 W LED
APP
(Color Grab)
RGB Quantitative 0.33 mM, 18.96 μM, 3.50 μM
and 0.32 mM for glucose,
nitrite, protein and
phenylpyruvate, respectively.
N/A Flexible, waterproof,
and adhesive
[168]
Reduced glutathione
(GSH) and
glutathione
reductase (GR)
Serum Liquid N/A CMOS N/A APP RGB Quantitative 0.1 μM and 10 μUmL
1
for GSH and GR
N/A High selectivity [169]
Nucleic acid TE buffer Liquid A white LED CMOS A box; a film heater;
a white LED; a loop-
mediated isothermal
amplification chip; a
DC converter; a 5 V
battery
APP
(qLAMP)
Hue Quantitative 10
2
copies per chamber 40 min Low-cost, high
portability, simplicity
in operation, and
user-friendliness
[79]
Human serum
albumin
Artificial urine Paper N/A CMOS N/A N/A N/A Quantitative 4.9 mg L 2 min High sensitivity [170]
Oxalate Urine Hydrogel White light CMOS An iron hood; a
white light source
ImageJ Hue Quantitative 8.0 μmol L
1
10 min Simplicity in
operation, cost-
efficiency, and good
selectivity
[171]
Glucose Blood Paper An
electroluminescent
sheet
CMOS A housing; an
electroluminescent
sheet; a battery
ImageJ RGB
(log(R
B
/
R
C
))
Quantitative 0.12 mM 12 min No need for
pretreatment, one-
step operation
[172]
Penicillinase Milk Paper An
electroluminescent
sheet
CMOS A housing; an
electroluminescent
sheet; a battery
APP RGB (R/G) Quantitative 2.67 10
2
mU μL
1
12 min Dual readout, high
sensitivity,
selectivity, and
reproducibility
[38a]
cTnI Serum Paper/soluble
polymer
LED CMOS A smartphone
mount; a LED; lens;
a translucent light
diffusing cover; a
battery
ImageJ N/A Quantitative 0.92 pg mL
1
20 min Simple, low-cost,
mass-producible,
highly sensitive, and
automated
[40b]
Uric acid Urine Liquid A UV flashlight
(365 nm)
CMOS A black housing; a
UV flashlight; two
optical filters
(365 nm, 400-
700 nm)
ImageJ Gray Quantitative 20.3 μM<10 min High sensitivity [173]
DNA Various types
of solution
(blood, urine,
milk, river
water etc.)
Microfluidic
chips
A white LED CMOS An integrated chip; a
foldable detection
box (LED, lens)
APP (POCKET
version 1.0)
Gray (ratio
to
reference)
Semiquantitative <10
3
copies mL
1
80 min Sample-to-answer ,
ultraportable
[174]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (6 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

the fluorescence occurs only at specific wavelengths of light. This
limits the generality of smartphone-based fluorimeter
housing, because the light sources and the light filters need to
be changed for different biosensors. Lee et al.
[59]
somehow solved
this problem by developing an adapter with a revolvable filter
case including four bandpass filters and a laser sheath which
allows the change of the light source (Figure 3e). Some applica-
tions of smartphone-based fluorimeters are listed in Table 3.
Generally, a smartphone adapter for fluorescent biosensors
should include the components of a convex lens, excitation light
sources, excitation light filters, dichroic mirrors, emission light
filters, and a lightproof holder. To achieve the detection of fluo-
rescent signals with different emission wavelengths and excita-
tion wavelengths, a filter set can be embedded in the adapter.
However, this method is obviously nonconvenient and limited
because it is impossible to cover the excitation and emission
spectrum of all of the fluorophores. The spatial approaches,
which rely on the dispersion of light using either a prism or
diffraction grating as discussed in Section 2.3, can be considered
for the future design of adapters for fluorescent biosensors. After
separating lights according to their wavelength, lights with
desired wavelengths can be selected according to their location
on the light spectrum using adjustable slits and gathered by
photomultipliers.
2.3. Spectrophotometry
A spectrophotometer is used to separate and measure the
spectral components of a beam of light.
[33]
Compared with color-
imeters and fluorimeters, which only give us information about
superimposed light with a wide range of wavelengths or light
with one certain wavelength, spectrometers give us a spectrum,
therefore enabling surface plasmon resonance (SPR)-based or
surface-enhanced Raman spectra (SERS)-based biosensing.
SPR spectroscopy is a powerful label-free technique that enables
noninvasive real-time monitoring of noncovalent molecular
interactions.
[60]
By observing peak shifts of the spectrum, the
concentration of analytes can be determined. Fan et al.
[61]
used
smartphone-based SPR spectroscopy for the detection of CA125
and CA15-3 (Figure 4a). The limit of detection reached 4.2
and 0.87 U mL
1
, respectively. Walter et al.
[62]
detected 25-
hydroxyvitamin D in human serum samples using a gold nano-
particle (AuNP) aptamer-based assay and achieved a sensitivity of
0.752 pixel nM
1
. The light was guided by optical fibers and inte-
grated with microfluidic chips (Figure 4b). Except for SPR-based
spectroscopy, another popular spectroscopy is based on SERS,
which refers to enhanced Raman scattering with the assistance
of nanomaterials.
[63]
The spectrum of Raman scattering of each
molecule can serve as a fingerprint and therefore can be used for
the detection of various analytes.
[64]
To obtain the light spectrum through smartphones, additional
components are required. First, the light beam can be separated
according to the wavelength through a prism or diffraction
grating (Figure 4c,d).
[65]
Light with different wavelengths will
be diffracted by different angles, resulting in the formation of
a spectrum. The spectrum can be captured by smartphones,
and the light intensity of each column can be calculated. After
calibration, the wavelength can be determined by its location,
Table 1. Continued.
Analyte[s] Sample Sample
platform
Light source Sensor Components Data
processing
Color
space
Analysis mode limit of detection (LOD)/
Detection range
Analysis
time
Merits Ref.
Glucose, cholesterol,
and lactate
Saliva Paper Lamp CMOS N/A ImageJ RGB (R) Qualitative N/A 10 min Multiplexed [175]
Anionic and cationic
species
N/A Paper A light source CMOS A black box; a light
source
ImageJ N/A Quantitative 5 10
6
mg mL 500s Sensitive, easy to
establish, simple in
operation and fast
[176]
Uric acid Saliva Paper LED CMOS A black box; a LED MATLAB/APP RGB (R) Quantitative 0.1 ppm 2 min Sensitive, low-cost,
easy to fabricate
[177]
SARS-CoV-2 Saliva Tubes Common day light CMOS N/A APP HSL (L) Quantitative 1 copy/μL 90 min Ultrasensitive, [178]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (7 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

Table 2. Designs and applications of smartphone-based detection platforms for luminescent sensing.
Analyte[s] Sample Sample platform Sensor Components Power
supply
Data
transfer
Data
processing
Color space/
measurement
index
Analysis
mode
LOD/
Detection
range
Relative
standard
deviation
Analysis
time
Merits Ref
PfHRP2 N/A Microchannel
capillary flow
assay
Silicon
photomultiplier
sensors
A black housing;
high-sensitive
optical detector
N/A USB-
OTG
APP Voltage Quantitative 8 ng mL N/A 10 min Ultrahigh sensitive,
no external flow
control, low cost,
power supplied from
cellphone, disposable,
and portable
[43]
Cholesterol Human serum Alginate
hydrogels
CMOS A black housing N/A N/A ImageJ Gray Quantitative 7.2 μM 0.6–6.6% 1 min High stability [179]
Valproic acid Blood and
saliva
Paper based CMOS A black housing with
an internal reflective
coating,
planoconvex lens;
a miniheater;
a magnet
N/A N/A ImageJ N/A Quantitative 4 μgmL
1
for blood
and
0.05 μgmL
1
for saliva
1%–4% for
blood and
3–7% for
saliva
10 min Simple, easy-to-use,
and reagentless
[50]
Hemin Human blood
serum
Test paper Photomultiplier tube
and CMOS
A black housing N/A N/A ImageJ Relative
light units
Quantitative 20 nM 1.8–2.6% 5 min High sensitivity
and selectivity;
cost-effective
[180]
Vitamin B
12
and Vitamin C
Blood diluted
by 10 times
Polyimide
sheet
CMOS A black housing;
a DC-DC Buck
Boost Converter;
cables
Smartphone N/A ImageJ Relative light
unit /Gray
Quantitative 0.109 nM
for VB
12
and
0.96 μM for VC
N/A N/A High sensitivity;
multiple detection
[181]
SARS-COV-2
antigen
N/A Paper CMOS N/A (dark
environment)
N/A N/A Image-Pro
Plus
software
N/A Quantitative 0.1 ng mL N/A 16 min Fast, highly sensitive,
and highly selective
[182]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (8 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

and a spectrum can be generated. Second, a light source is nec-
essary. Normally used light sources include LEDs,
[61]
smart-
phone flashlights,
[62]
or sunlight,
[66]
with the LEDs being the
most commonly used due to their low cost, accessibility of a wide
range of wavelengths, and application variability.
[9]
CloudMinds
Inc. invented a portable smartphone-based SERS spectrophotom-
eter (Figure 4e). By applying the slit space coupling technique
and volume-phase holographic transmission grating, the sensi-
tivity is improved while the size of device is reduced.
[67]
The
device can be integrated to the back of the smartphone and com-
municated through Smartport interface. However, instead of
using the CMOS sensor of smartphones, a charge-coupled device
is used for signal acquisition. The data will then be transferred to
a cloud network through Wi-Fi or 3 G/4 G/5 G. The server will
smooth the spectrum, and the baseline will be fit and subtracted
to obtain a pure Raman spectrum.
[68]
Several research groups
have verified this device. The details are listed in Table 4.
Because in this case smartphone only serves as a result receiver,
the difference in smartphone camera spectral sensitivity is cir-
cumvented. However, if we use the smartphone camera for sig-
nal acquisition, color correction is necessary to ensure that
different smartphones produce the same color value for the same
color signal. The details will be discussed in Section 5.
3. Color Space Transformation
3.1. Color Spaces
The color changes in the liquid, test strips, or hydrogels can often
be used for quantitative detection of various analytes through dif-
ferent colorogenic reactions. These colors could be abstracted
numerically as a mathematical formula so that they can be stored
and interpreted on devices/computers. The value of its
Figure 3. Illustration of smartphone-based fluorimeters. a) Schematic diagram of the smartphone and microfluidic chip platform for immunofluorescent
signal detection. Reproduced with permission.
[58]
Copyright 2018, Royal Society of Chemistry. b) Smartphone-based particle diffusometry platform.
i) Optics setup within the platform includes an external ball lens, filter, and laser at a 15° incident angle; ii) Schematic of the adapter; iii.
Integration of adapter to smartphone. Reproduced with permission.
[156]
Copyright 2020, Elsevier. c) Schematic diagram of a smartphone-based fluores-
cence spectrum reader. Reproduced with permission.
[157]
Copyright 2018, Elsevier. d) Working principle of a dichroic mirror. Reproduced with permis-
sion.
[158]
Copyright 2019, Elsevier. e,i) Schematic diagram of a smartphone-based fluorimeter with a selectable filter; ii) A demonstration of fluorescent
imaging with a real object. Reproduced with permission.
[59]
Copyright 2017, Elsevier.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (9 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

Table 3. Designs and applications of smartphone-based fluorimeters.
Analyte[s] Sample Sample Platform Light Source Power
Supply
Sensor Components Data Processing Color Space Analysis
Mode
LOD/Detection
Range
Analysis
Time
Merits Ref
Quantum dots N/A LFA A violet LED Smartphone
through USB-OTG
CMOS A black matboard
housing; violet LED;
paper-based filters
MATLAB RGB Quantitative 2.5–20 nM N/A Cost effective [183]
Tetracycline Milk Cuvette or test
paper
A UV lamp
(302 nm)
A USB interface
via a regulator
N/A Black housing; a UV
lamp (302 nm)
Color
Picker APP
RGB (B/R) Quantitative 6.6 nM 2 min Low cost, easy to
carry, simple
operation, and high
selectivity and
repeatability
[38b]
Tetracycline Milk Test paper A UV lamp
(365 nm)
N/A CMOS N/A Color recognizer
APP
RGB (R/B) Quantitative 20 nM 5 min High selectivity [38c]
Dipicolinic acid N/A Test paper A UV lamp
(254 nm)
N/A N/A A dark case Color analyzer
APP
RGB (R/G) Quantitative 67 nM N/A N/A [38d]
cTnI Human serum Microfluidic
chip
A blue LED Battery CMOS Two blue LEDs;
two bandpass
excitation filters;
one bandpass
emission filter; one
20microscope
objective
MATLAB/APP Luminosity value Quantitative 94 pg mL
1
12 min Easy operation,
miniature
[58]
Dipicolinic acid A boric acid–
borate buffer
solution
Liquid A UV lamp
(254 nm)
N/A CMOS Dark room APP RGB (G/R) Quantitative 0–300 μM N/A High sensitivity
and selectivity
[38e]
Thiram residue Tap water,
apple
peel, and milk
Test paper A UV lamp Powered by
any device
though a
USB interface
CMOS A dark housing;
afilter
(400–1000 nm)
APP RGB (R/B) Quantitative 59 nM N/A Rapid, sensitive,
and instrument free
[38f]
Ochratoxin A,
Hg
2þ
,
salmonella,
hepatitis B
virus, and growth
stimulation
expressed
gene 2
N/A LFA A laser (980 nm) N/A CMOS A white housing; a
laser diode; a
dichroic mirror; an
infrared filter
APP
(UCNP-LFA
analyzer)
RGB Quantitative 5–10-fold lower
compared
with the clinical
cutoff value
15 min Real time, high
sensitivity, and
multiplex
[158]
Zearalenone Maize and millet
extraction
LFA UV light N/A CMOS N/A APP (ZEN
reader)
RGB Quantitative 0.0058 ngmL 20 min Colorimetric and
fluorescence dual
mode, high
sensitivity, and
repeatability
[184]
Mycotoxins 7% methanolPBS Liquid A laser (574 nm) A battery CMOS A battery; laser
diodes; filters; a
single-cylinder
microscope
APP Gray Quantitative 0.1 ng mL 1 h 45 min Multiplex [185]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (10 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

Figure 4. a) Schematic of attachments of the smartphone biosensor system with multitesting units based on localized SPR integrated with microfluidic
chips. i) The working principle of the system; ii) The schematic of attachments; iii) The detail of the case; iv) The detail (top view) of the stage of microhole
array and small-lens array; v) The picture of attachments and the smartphone. Reproduced with permission.
[61]
Copyright 2020, MDPI. b) Conceptual
design of the all-optical planar polymer-based biochip sensor platform for smartphones. Reproduced with permission.
[62]
Copyright 2020, MDPI. c) Light
diffraction through a prism. d) Light diffraction through diffraction grating. e) Schematic illustration of the SERS measurement using the smart SERS
terminal, adapter, and SERS chips. Reproduced with permission.
[159]
Copyright 2019, Royal Society of Chemistry.
Table 4. Verification of smartphone-based portable SERS spectrometers from cloud minds inc.
Analyte[s] Sample Sample Platform LOD/Detection Range Analysis Time Merits Ref
12 kinds of pesticides N/A SERS chip 10
9
M
10 s Miniature, sensitive,
and high optical
collection efficiency
[68]
Rhodamine
6 G, crystal violet,
and malachite
green
Ethanol Paper-based
SERS chip
1 pmol for rhod amine 6 G
and 10 pmol for both crystal
violet and malachite green
10 s High resistance to friction,
miniature, cost-effective,
and good repeatability
[186]
Crystal violet N/A Paper-based SERS chip 10
9
M
10 s High stability [187]
Serum amyloid
A (SAA) and C-reactive
protein (CPR)
N/A LFA 0.1 and 0.05 ng mL
1
for SAA and CRP
20 min N/A [188]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (11 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

mathematical representation varies with different color models,
including RGB, HSV, HSL, CIE XYZ, CIE L*a*b* (or LAB),
etc. The most common color model is RGB. It is an additive color
model in which the red, green, and blue monochromatic (single
wavelength) primary colors of light are added together in various
combinations to reproduce multiple colors.
[69]
Different devices
may use different primary colors and yield different RGB color
spaces, meaning that the same set of tristimulus values may refer
to different colors (Figure 5a). Thus, RGB is specific to one imag-
ing device and is termed as device dependent. RGB is coded on
256 levels, from 0 to 255. The value represents the light intensity
of its corresponding primary beam. For instance, zero value of R,
G, and B indicates no existence of light and represents black
color. The maximum value of R, G, and B means superimposi-
tion of fully-on light beams, representing white color. Different
colors can be represented by different combinations of R, G, B
values. However, it is not easy to intuitively tell the tristimulus
values of a specific color. Therefore, HSV and HSL were created
to accommodate easier color adjustment (Figure 5b). HSV is a
cylinder color model that remaps colors into three interdepen-
dent dimensions that are more understandable to humans,
which are hue, saturation, and value. Hue describes the domi-
nant color family, saturation represents purity, and value controls
the brightness. HSL is another cylinder color model, whose H
and S dimensions are the same as those of HSV. L represents
lightness, indicating the color’s luminosity. It is different from
the V dimension of HSV in that the purest color is halfway the
black and white ends of the scale (Figure 5b). HSV and HSL are
both derived from RGB space and therefore are device
dependent.
One issue with RGB color model is that not all the colors that
humans can perceive can be represented in RGB without intro-
ducing negative values. Therefore, in 1931, International
Commission on Illumination (CIE) created the CIE 1931 XYZ
color space, which uses imaginary colors as primary colors, so
as to avoid negative values for all visible colors.
[70]
CIE XYZ is
also a standard representation for colors. By mapping the color
from a device’s own color space to CIE XYZ space, color can be
measured or translated to another device-dependent color space.
In the XYZ color model, Y corresponds to relative luminance, Z
is approximately equal to B dimension of CIE RGB, and X is a
blend of all three.
[71]
Note that any color space with an explicit
relationship to XYZ is said to be device independent. For exam-
ple, once the three primary colors of an RGB space have deter-
mined values in XYZ color space, this RGB space becomes device
independent. One limit of XYZ color space is perceptual
Figure 5. a) Illustration of different RGB color spaces in the chromaticity diagram. The three vertices of each triangle indicate the three primary colors of
the corresponding color space. The area inside the triangle is the gamut of that color space. b) HSV and HSL color spaces. c) Comparison between linear
RGB and LAB color space. d) Illustration of LAB color space. Reproduced with permission.
[160]
Copyright 2009, Springer Nature.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (12 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

nonuniformity, which means that a Euclid distance on the CIE xy
chromaticity diagram is not proportional to the perceived differ-
ence between two colors (Figure 5c).
[72]
Several attempts have
been made to derive perceptually uniform color spaces from
the CIE XYZ color space. In 1976, the CIE suggested using
two quasiuniform color spaces, CIE L*a*b* (or LAB)
(Figure 5d) and CIE L*u*v* (or LUV).
[73]
Both of the systems
use L* as lightness value, which correspond to the cube-root of
the luminance. The other two axes, a* and b* for CIELAB or u*
and v* for CIELUV, define colorfulness. LAB and LUV are based
on four psychological primary colors, which are red, green, yel-
low, and blue. Red and green compose a pair of complementary
colors, and yellow and blue form another pair. The mix of colors
could happen across pairs but not within a pair.
[74]
For example,
orange is a mix of red and yellow, but the clash of red and green
produces a nearly black color. The value of a*(or u*) and b*
(or v*) represents color changes from red to green and yellow
to blue, respectively.
3.2. Applications of Different Color Spaces in Biosensing
Color quantification methods with smartphones using different
color spaces have been extensively reported. Each color space per-
forms differently in different applications, and some color spaces
have been reported to be able to reduce color variation caused by
illumination conditions or produce higher accuracy. Yang
et al.
[75]
used one of the LAB color spaces, CIEDE2000, for
colorimetric urinalysis. With fixed illumination conditions, the
accuracy was comparable to a urine analyzer. They claimed it
was necessary to convert RGB color space to a uniform color
space because the distance between two colors in RGB cannot
represent the human sense of color. However, no comparison
between RGB and CIEDE2000 was reported in this research.
Although they listed the results of other research using RGB,
CIE XYZ, and HSV, showing CIEDE2000 outperformed other
color spaces, these experiments were not conducted under the
same conditions and cannot be used for comparison. Ba ş
et al.
[76]
compared the performance of RGB and LAB on the quan-
titative measurement of glucose. Under constant illumination
conditions, both RGB and LAB gave calibration curves with high
linearity. In contrast, under ambient light which might cause
gray background, LAB can still produce high linearity
(R
2
¼0.996), while RGB gave relatively poor linearity
(R
2
¼0.925). However, why LAB is more resistant to illumina-
tion variations was not mentioned. Moreover, whether the varia-
tion of illumination conditions causes a curve shift in color
intensity was not reported as well. Kılıc et al.
[39]
compared two
color-matching algorithms for semiquantitative water quality
detection based on a single reference image. The first method
is based on color correlation, which calculates correlation coef-
ficients between test images and trained images. The second
method is based on the Euclidean distance between the test
image and the reference image in LAB color space. It was shown
that the second method outperformed the first method.
Similarly, Komatsu et al.
[77]
also used the Euclidean distance
in LAB color space for the measurement of pH values. The
pH indicator showed multiple color changes. It was claimed that
the detectable pH range of this method was wider than the typical
Figure 6. Properties of different color spaces. a,i) RGB did not show a clear trend over pH from 2–9; ii) ΔEdemonstrated an obvious increasing trend with
low relative standard deviation for each pH value. Reproduced with permission.
[77]
Copyright 2016, Royal Society of Chemistry. b,i) Monotonal color
signals were converted into bitonal color signals after tinting; ii) Hue value difference was created after tinting. Reproduced with permission.
[81]
Copyright
2017, Royal Society of Chemistry.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (13 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

grayscale-based image analysis (Figure 6a). Alvarez et al.
[78]
used
different mathematical combinations of RGB parameters,
including grayscale intensity, effective absorbance of R, and
Euclidean distance between signal color and background color
in RGB color space, for the measurement of tropospheric ozone.
Effective absorbance of Rshowed the best performance. Nguyen
et al.
[79]
demonstrated that using hue value can eliminate the
influence of light source intensities given that the assay was
based on bitonal or multitonal color change. Cantrell et al.
[80]
also
verified that hue value provided better results than RGB for bito-
nal optical sensors. However, most of assays depend on the con-
centration of one specific color indicator; therefore, they are
monotonal. To solve this problem, Krauss et al.
[81]
developed a
tinting method, for example, changing white background to blue,
and yellow color signal to green, so as to induce a larger change
in hue value (Figure 6b). Kestwal et al.
[82]
used the saturation
parameter of HSV color space for beta-glucan paper-based detec-
tion. The signal was normalized by subtracting the background
signal, which was prepared from a blank or control paper device.
High linearity with R
2
equal to 0.9953 was achieved. However, no
comparison with other parameters or color spaces was reported.
The study of Yang et al.
[83]
found that S of HSV and S of HIS
were not affected by smartphones for the prediction of soil
organic matter content. Shalaby et al.
[84]
examined the perfor-
mance of four nonuniform color spaces (RGB, CMY, XYZ,
and Yxy), six uniform color spaces (LAB, LCH, Hunter-LAB,
LUV, HSL, HSV), ΔE
LAB
, and ΔE
LUV
on the prediction of
Cr(VI) concentration, showing that the use of Yxy slightly
improved the linearity of the calibration line and all uniform
color spaces gave at least one signaling parameter exhibiting
extraordinary sensitivity and linearity. However, the conclusions
may only be applicable to Cr(VI). An investigation of multiple
dyes is desired. Nelis et al.
[85]
used liquids, pH strips, and
LFAs for the investigation of color space channels’(RGB,
HSV, LAB, and ΔRGB) efficiency. The color changes were intro-
duced by dropped NPs often used for LFAs (i.e., gold, latex, or
carbon black NPs) and oxidized TMB solutions in ELISA wells.
The performance of each channel was compared under varying
background illumination conditions. Moreover, six smartphones
were used for imaging and their calibration curves were com-
pared. Contrary to the conclusions drawn from most of other
researchers, Nelis et al.
[85]
showed that L and V performed well
in most systems but never outperformed the best RGB channel
for a specific test. The H channel performed satisfactorily for
color change, but poorly for color intensity quantization, and
never surpassed the best RGB channel in a particular test either.
R worked best for color changes, while B and G functioned best
for color intensity changes. The ΔRGB values showed some
robustness to errors in individual channels of RGB. They also
demonstrated that the calibration curve derived from one smart-
phone could not be directly applied to another smartphone.
Overall, opinions regarding which color space performed best
vary widely with assays. Which color space is favorable might
depend on the overlap between the absorbance spectrum of color
signal substances and the spectrum of color channel, meaning
that each color channel might need to be tested for each assay.
Generally, uniform color spaces, such as LAB and LUV, may out-
perform nonuniform color spaces for measuring small changes
in color considering that uniform color spaces resemble human
perception and human eyes are more sensitive to relatively low
intensity of light and smaller changes in color compared with
camera sensors. New color spaces or algorithms can be invented
to increase the linearity of calibration curve and to be more resis-
tant to illuminations.
4. Machine Learning Classification
ML is a subset of artificial intelligence that studies computer
models and algorithms (e.g., neural networks) in a data-driven
way. Then the trained model with learnt pattern is applied to
infer the rest of data.
[86]
It has emerged as a powerful tool for
classification problems due to its flexibility and adaptability to
dynamic conditions based on the features extracted from colori-
metric information.
[87]
It can deal with changes in illumination
and smartphone models in one stage. Traditional ML typically
requires three components to learn a pattern: a dataset, features
engineering, and an classification algorithm (Figure 7). The
larger the size of data and features provided, the higher the
accuracy.
4.1. ML Classification-Based Colorimetric Detections
Solmaz et al.
[88]
used support vector machine (SVM) and least
squares-support vector machine (LS-SVM) classifier algorithms
to classify distinct pH values. SVM is a supervised learning
model which clusters training data to distinctive groups by map-
ping training examples to high-dimensional feature space so as
to maximize the gap between the two categories. LS-SVM is the
least-squares version of SVM which solves a set of linear equa-
tions instead of a convex quadratic programming problem for
classical SVMs.
[89]
In this research, the influence of image format
and illumination conditions is investigated. It was hypothesized
that JPEG images were highly processed and compressed; there-
fore, the relationship with incoming light intensity is nonlinear,
so the final color value cannot be fully trusted. RAW images
derived from raw sensor data from a digital camera and
RAW-corrected (RAWc) images after WB and color transforma-
tion to CIE 1931 XYZ color space of RAW images were compared
with JPEG images. To mimic the diverse illumination conditions
in real life, three different homogeneous light sources were used:
sunlight (S), fluorescent (F), and halogen (H). To include more
versatile conditions, the strips with the same pH level were
imaged as a group of four with six different orientations and
alignments including the pictures with variable rotations and
Figure 7. Flowchart for smartphone-based colorimetric detection by ML
classification.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (14 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

illumination intensities. The luminance on each strip is slightly
different compared to the others due to the positioning with
respect to the light source. For each light source and each image
format, 90 images captured by LG G4 smartphone were used for
training. The mean RGB values extracted from the testing region
of strips were fed to both SVM and LS-SVM. Almost 100% accu-
racy was achieved for LS-SVM, while SVM performed signifi-
cantly worse. Moreover, being different from the hypothesis
that RAW images outperform JPEG images, the effect of format
varies with pH values, and generally JPEG performs better than
RAW images and RAWc images using the SVM classifier. To
further test the robustness of LS-SVM, images captured under
dual-illumination conditions (combined lighting conditions con-
sisting of the fluorescent–halogen [FH], fluorescent–sunlight
[FS], and halogen–sunlight sources [HS]) were used for testing,
showing 80% of accuracy, which could be improved if images in
dual-illumination conditions were used for training.
Do˘
gan et al.
[87]
developed an ML-based smartphone APP
called “Hi-perox Sens”capable of image capture and analysis
of color signals on microfluidic paper-based analytical devices
(μPADs) for nonenzymatic colorimetric determination of
H
2
O
2
through iodide-mediated TMB-H
2
O
2
reaction system.
The Hi-perox Sens uses a Firebase cloud system for both trans-
ferring the image to the remote server and receiving the classifi-
cation result back to the APP. The μPADs were prepared by
adding only two indicators, TMB and KI, or KI only, to the test
zones. The images were captured with four different smartphone
models (Oppo A5 2020, Reeder P10, iPhone 5SE, and iPhone 6S)
under seven different illumination conditions (H, F, S, HF, HS,
FS, and halogen–fluorescent–sunlight). A total of 33 features,
including 9 color channels (R, G, B, H, S, V, L*, a*, b*), mean,
skewness, and kurtosis values for each color channel, texture fea-
tures (contrast, correlation, homogeneity, and energy), entropy,
and intensity values were extracted from the test zone on μPADs
and fed into different ML classifiers. 23 ML classifiers were
trained, and their performances were compared in terms of clas-
sification accuracy. Among these classifiers, linear discriminant
analysis and ensemble bagging classifier outperformed others
for KI and TMB þKI, respectively. Classification accuracy of
97.8% was reached for TMB þKI in the 0–5 mM concentration
range, and classification accuracy of 92.3% was reached for KI in
the 0.2–50 mM concentration range with good interphone
repeatability at t¼30 s.
Instead of putting the information of all of the color channels
into a classifier, Khanal et al.
[24]
evaluated the performance of
RGB, HSV, and LAB separately with four ML classifiers, includ-
ing logistic regression (LR), SVM, random forest, and artificial
neural network (ANN). The images of food color and PADs
for pesticide assays captured under different illumination condi-
tions with four different smartphone models (Huawei SCC-U21,
iPhone 6, Honor 8C, and Samsung Galaxy J7 Max) and users
were used for training and testing. Additionally, a reference zone
was created aside from the test zone. The reference zone has the
same components as the test zone, while a blank solution with-
out pesticide was added to the reference zone as running the
assay. Images of both reference zone and test zone were fed into
classifiers in contrast to most of the other studies that imaged
only the test zone. According to the results, this one-point cali-
bration method provided better accuracy, indicating that
reference color may partially eliminate the influence of different
imaging conditions. ANN model and LAB color space produced
the best concentration prediction accuracy (0.966) for food color
assay, and SVM model and LAB color space produced the highest
accuracy (0.908) for enzyme inhibition assay. However, this does
not necessarily mean LAB color space works better than RGB and
HSV. For example, RGB outperformed other color spaces in
terms of the crossvalidation accuracy for food color assay using
LR, SVM, and ANN models. Kim et al.
[90]
performed a similar
study, where three different classifiers (linear discriminant anal-
ysis [LDA], SVM, and ANN) and four color spaces (RGB, HSV,
YUV, and Lab) were evaluated. No explicit trends among color
spaces and classifiers were found. Therefore, the selection of
ML models and color spaces should be based on the applications
and experimental data. More details and examples of ML-based
colorimetric detection are listed in Table 5.
ML solves problems in a human-like manner. Utilizing the
ML models to study from the data is a promising method to fac-
tor out the influence of ambient light and interphone variation.
However, to achieve high accuracy and high resolution, a very
large training dataset, usually over thousands of images, is
required. Moreover, ML based on classification only gives
semiquantitative results. With coarse-level classifications, the
accuracy could reach 100%. However, when performing fine-
grained concentration classifications, the performance may not
be satisfying. To improve the resolution of assays, the training
data set needs to be enlarged as well. For example, to obtain
results with a fractional value, one needs to have a training
set with labels specified to decimals.
[88]
Nevertheless, this would
require extensive labor work to take all possible values into con-
sideration on a continuous scale. How to produce enough data in
a small amount of time is a problem that needs to be solved
urgently. Luo et al.
[91]
developed a high-throughput colorimetric
sensor for total organic carbon analysis of environmental water
samples. Five different types of ink were overprinted on a carrier
to form thousands of interaction points, with each point repre-
senting a microreaction and producing independent data.
Therefore, thousands of tests could occur simultaneously, and
the results can be captured in one image. However, this method
may only apply to assays based on simple chemical reactions.
Another challenge for ML classification is the training dataset,
which has to cover all of the possible conditions. If a domain shift
happens in test data, the prediction results will be much unreli-
able. For instance, it has been tested that although three different
light sources have been used for training, as the illumination
condition was changed to the combination of these three light
sources, the accuracy was decreased from 100% to 80%.
[89]
If
users captured the assay signal under different light sources
and different light intensities, the accuracy might decrease fur-
ther. Yang et al.
[83]
also demonstrated that different smartphones
did not fit in the model constructed from other smartphones very
well. Training images captured with more diverse conditions
may ensure the accuracy, but it also means including a very large
amount of training set to the classifier, which needs massive
resources to function. In the past few years, convolutional neural
networks (CNNs) have been very successful in computer vision.
However, CNNs have limited advantages over other ML models
because colorimetry on paper-based devices does not provide tex-
ture and shape variation. Developing a PAD that expresses
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (15 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

changes in the shape and texture according to the analyte con-
centrations can allow us to take full capacity of CNNs and build
more accurate methods for estimating analyte levels.
4.2. Accuracy Improvement Strategies
Increasing the accuracy of models is always the driving force for
the development of ML algorithms. Many approaches are proven
effective in improving accuracy, all of which can be divided into
two main categories, improving the dataset and the model.
The size and the quality of the dataset are the key factors that
affect classification accuracy. Experiments on multiple models
including CNN and ResNet prove that the accuracy of the model
increases with the size of the dataset.
[92]
The most straightfor-
ward method for enlarging datasets is to generate more labeled
data, but this method may not be applicable to all tasks. In the
case of biosensing and biomedical imaging, data acquisition
requires consent from patients and labeling by healthcare profes-
sionals, which limits the size of the dataset. In this case, data
augmentation methods are used. For optical sensors, the input
of the ML algorithm is usually an image, and there are many
augmentation strategies for images. The simple augmentation
techniques include geometric transformations, color space aug-
mentations, kernel filters, mixing images, random erasing,
etc.
[93]
Recently, generative adversarial networks (GANs) have
been used in data augmentation, and they can augment sources
of variance which are not produced in traditional augmentation
methods.
[94]
Using the augmentation strategies, the size of the
dataset is enlarged, and the accuracy of the ML algorithm is
improved.
The quality of the dataset is determined by many aspects. The
three most important factors influencing the quality of a dataset
are comprehensiveness, correctness, and variety.
[95]
The compre-
hensiveness of data describes how well the dataset includes all
representative samples. A lack of comprehensiveness may lead
to a biased model. Missing value treatment is the process of gen-
erating samples that fill up the gap in the sample population, and
it can increase the comprehensiveness of the dataset.
[96]
The cor-
rectness of data corresponds to the validity of data and the accu-
racy of labels. The ILSVRC12 ImageNet dataset is the most
frequently used one in classification tasks. Russakovisky et al.
pointed out that there are mislabeled images in their dataset.
[97]
Automatic error detection and cleanup algorithms can correct the
mislabeled data and improve the accuracy in ImageNet dataset
classification by 2–2.4%.
[98]
The variety of the dataset is the mea-
surement of the convergence of the distribution of data samples
to the overall population. Improving the variety of the dataset is
mostly relied on the sampling procedure, in which the
Table 5. Smartphone-based colorimetric detections via ML.
Analyte[s] Platform Classes Classifier
algorithm[s]
Illumination
conditions
Image format Color spaces Optical detection
devices
Analysis mode Ref
Nucleic Acids Tubes Fluorescent CNN Blue (470 nM) or
UV lights
N/A N/A Smartphone camera
or Peiqing automatic
gel imaging analysis
system
Qualitative [189]
H
2
O
2
Test strips Colorimetric LS-SVM; RF Incandescent,
fluorescent
light bulbs, sunlight,
and
their combination
JPEG RGB; HSV; LAB REEDER P10,
SAMSUNG Galaxy
Note 3, LG G4, HTC
One M7, and SONY
Xperia Z1
Semiquantitative [36]
Alcohol Test strips Colorimetric LDA, SVM, ANN Constant
illumination
(smartphone
flashlight
with housing)
N/A RGB, HSV, YUV,
and LAB
Samsung Galaxy S4,
Samsung Galaxy
Note 3, LG G2,
Oneplus One, and
Motorola Moto G
Semiquantitative [90]
Glucose μPADs Colorimetric LDA,
gradient
boosting, RF
Sunlight,
fluorescent, halogen,
and their
combination
N/A RGB; HSV; LAB iPhone 6S, iPhone 7,
Reeder P10, and
Samsung J7
Semiquantitative [190]
Salmonella spp. LFA Colorimetric SVM, discriminant
analysis,
naïve bayes, nearest
neighbors, decision
trees, and
neural networks
Constant
illumination
(LED with housing)
N/A RGB, LAB OnePlus One Quantitative [191]
C-reactive protein μPADs Colorimetric MLP, CNN, and
ResNet
Natural light, indoor
light, and flashlight
N/A RGB, HSV Xiaomi 8 Semi-quantitative [192]
Bilirubin Paper Colorimetric Random Forest Ambient light N/A RGB, CYMK Digital camera Semiquantitative [193]
Glucose Cuvette Colorimetric LDA, bagging
classifier, RF
White LEDs JPEG RGB, HSV, LAB LG G6 Semiquantitative [194]
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (16 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

percentage of a class of samples should be approximately the
actual percentage of that class. By increasing the quality of the
dataset, the accuracy of the algorithm can be improved.
[99]
There are many approaches to improving the accuracy of a
neural network by working on the models. The two major
approaches in optimizing the model are the selection of back-
bone networks and the tuning of hyperparameters. The rapid
development of neural networks provides researchers with many
choices of models, from the traditional convolution neural net-
works (CNNs) to the popular transformer models. Selecting the
correct model for the task will increase the accuracy by a great
margin.
[100]
There is no gold standard for choosing between
models, only by experiments can researchers pick out the
optimal model. Tuning of the hyperparameters is the most
important procedure to ensure a highly accurate model.
Hyperparameters refer to the parameters that cannot be trained
during the training process, most of which are related to the
model’s architecture and the training settings.
[101]
The most typ-
ical hyperparameter in model design is the number of hidden
layers and the number of neurons in each layer. The model’s
ability to analyze data increases with the size of the model. If
the size of the model is too small, the model may not be able
to utilize all the features of the sample, causing underfitting
problems.
[102]
On the other hand, if the size of the model is
too big, it may learn features from the noise of the training data,
causing a decrease in the model’s generalization ability and
resulting in overfitting.
[103]
The hyperparameters regarding the
training procedures also influence the accuracy of the model.
Three of the most researched training settings are learning rates,
optimizers, and batch sizes. Learning rates decide the length of
each step in the backpropagation process. Large learning rates
can converge to a global minimum faster, while small learning
rates may locate a more optimal minimum.
[104]
The goal is to find
a learning rate that can quickly locate an optimal minimum.
Several methods can be used to find an adequate learning rate.
Learning rate decay methods like stepwise LR decay
[105]
and
exponential LR decay
[106]
decrease the learning rate in the train-
ing process; this could make the training both efficient and accu-
rate. Optimizers are the optimization algorithms that find the
minimum in the loss function; different optimizers will give dif-
ferent training speeds and accuracy. The most adopted
optimizers include stochastic gradient descent (SGD),
[107]
root
mean square prop (RMSprop),
[108]
and Adaptive momentum esti-
mation (Adam).
[109]
The choice of the optimizer is a complicated
matter, which depends on the quality of the datasets and the
structure of the model. In most cases, Adam should be used
as the default optimizer as it performs consistently in a variety
of tasks.
[110]
Batch size is the number of training samples that is
used to calculate the gradient in each iteration. Theoretically, a
larger batch size can give a better estimation of the gradient. But
larger batch size will have a larger memory cost, and it can even
lead to loss of generalization ability.
[111]
Studies also show that
batch size and learning rate should be picked in pairs; a large
batch size performs best with a large learning rate.
[112]
In sum-
mary, model selection and hyperparameter tuning have a great
impact on the accuracy of the neural network.
In conclusion, approaches to improving the accuracy of ML
algorithms can be divided into two main categories, improve-
ment to the dataset and to the model. The dataset should have
a high level of comprehensiveness, correctness, and variety, and
the model should be designed with a suitable structure and be
trained with the appropriate procedure.
[95]
5. Color Correction
5.1. Camera Image Signal Processing
Color correction aims to eliminate the influence of illumination
and achieve color constancy for different smartphones. To under-
stand the principle of color correction, it is important to have a
general knowledge of how smartphone cameras capture images
and how they are processed. An image signal processing (ISP)
pipeline can be divided into two stages (Figure 8). The first stage
involves the conversion of raw RGB data from camera-specific
color space to device-independent color space, including
preprocessing, WB, CST, etc. The second stage includes gamma
encoding, postprocessing, etc. The purpose of the second stage is
mainly for image compression and color manipulation.
The color signal detected by camera sensor is determined by
the product of irradiance, reflectance, and the spectral sensitivity
of camera. Following the matrix notation of Karaimer et al.,
[113]
let lrepresent the illumination as a 1 Nvector, where Nis
Figure 8. A diagram of a typical in-camera imaging pipeline.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (17 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

the number of spectral samples in the camera’s sensing range
(e.g., from 400 to 700 nm). Each vector coordinate is the intensity
of light with the corresponding wavelength. We use Rto repre-
sent the reflectance of imaging targets as an NMmatrix, where
Mis the number of imaging materials. For easier understanding,
we specify the imaging target as a color rendition chart. Then M
is the number of color patches on the color chart. We use C
cam
to
represent the spectral sensitivity of the camera as a 3 N matrix,
with each row corresponding to one color channel, (i.e., R, G, and
B). Using this notation, the camera’s response in illumination l
can be represented as
Φl¼CcamdiagðlÞR¼Ccam LR (1)
where Φ
l
is a 3 Mmatrix. Each column is the RGB value of a
color patch. For imaging the same material, the camera’s
response is therefore dependent on the illumination and camera
spectral sensitivity. Efforts have been made to achieve color con-
stancy in the factory. A colorimetric mapping is performed to
convert sensor RGB values from their camera-specific color space
to a perceptual color space. The target perceptual color space can
be expressed as
Φxyz ¼CxyzR(2)
where C
xyz
is the CIE 1931 XYZ matching function. The illumi-
nation matrix Lis ignored in this equation because it is assumed
that the image is captured under ideal white light (i.e., all entries
of lare equal to 1). To derive the mapping function from Φ
l
to
Φ
xyz
, we need to know the exact value of Φ
xyz
. Therefore, a color
checker with established reflectance is usually imaged for calibra-
tion. The colorimetric mapping can be divided into two steps:
WB and CST. The whole process of colorimetric mapping is
summarized in Figure 9. Note that the CST in this section is
different from the concept in Section 3.
5.1.1. White Balance
Human visual system is able to maintain the color appearance of
objects under a wide variation of light conditions through a pro-
cess called chromatic adaptation.
[114]
The small changes in color
can be discounted after being accommodated to the illumination.
For example, a white paper would still be perceived as white at
dusk, even though the actual color may be yellowish. In contrast,
camera sensors do not have the ability of chromatic adaptation
and only record the actual light, which nevertheless may not
appear natural to humans. Therefore, a WB, which is a linear
transform, is performed to remove the illumination’s color
cast.
[115]
One simple method for WB is using RGB equalization,
or wrong von Kries model, with reference to a white standard and
a black standard.
[35]
The gray patches on the color chart equally
reflect light with different wavelengths. The R, G, and B values
Figure 9. This figure is modified after Karaimer et al.,
[113]
which illustrates the colorimetric mapping from the camera-specific color space to the standard
CIE XYZ color space.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (18 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

recorded by the camera are expected to be equal for such surfa-
ces. However, this is never the case due to the properties of illu-
mination and spectral sensitivity of the camera sensor. RGB
equalization counteracts this error. However, although it is easy
to implement, it does not ensure the natural look of images.
More complex models, such as chromatic adaptation transform
(CAT) (or white point conversion), provide better human percep-
tion. Commonly used CAT models include Von Kries,
[116]
Bradford,
[117]
Sharp,
[118]
and CMCCAT2000.
[119]
Another method for WB is to calculate a diagonal 3 3 matrix,
W
d
, that minimizes the ideal camera response in white light for a
white object and the actual camera response. The subscript d
means W
d
is a diagonal matrix. Because the neutral object
reflects the light spectrum equally, the reflectance is equal to
the illumination, and therefore all entries of Rare equal to 1.
Once the illumination has been estimated, W
d
can be computed
directly from the estimated illumination parameters as
Wd¼diagðCcamlÞ1(3)
There are many ways to estimate illumination, including gray
world,
[120]
white patch-based method,
[121]
bootstrapping
method,
[122]
statistical methods,
[123]
gamut-based methods,
[124]
and ML methods.
[125]
The fundamental idea is to put a white
calibration object in the imaged scene. The chromatic cast on
the white object is the color of the illumination in the camera’s
color space. Interestingly, Abdelhamed et al.
[126]
leveraged the
difference in the response of the two rear-facing cameras in
the smartphone to estimate the illumination by training a small
neural network.
The obvious drawback of WB is that it cannot guarantee that
the nonneutral scene materials are appropriately corrected.
[113]
Thus, a CST is applied to convert the raw tristimulus values
captured by the camera from the camera-specific color space
to a perceptual device-independent color space (e.g., CIE XYZ)
for subsequent processing.
5.1.2. Color Space Transform
To generate chromatic images, the camera splits the light into
three channels by putting a CFA (usually a Bayer mosaic filter)
in front of the photodiode to record the tristimulus values. The
filter properties vary with camera models, so different cameras
record different tristimulus values for the same scene even in the
same illumination. CST corrects and ensures the reproducibility
of image color by converting the tristimulus values from a device-
dependent color space to a standard, device-independent color
space. The 3 3 CST matrix, T
l
, can be computed by minimizing
the following.
Tl¼argTlminkCxyzRTlWl
dΦl
camk2(4)
where Φl
cam is the camera response under illumination l, and Wl
d
is the WB matrix for illumination l. The calculated parameters
are stored in the firmware of the camera, so customers do not
need to re-calculate the CST using a color chart every time before
taking a photo. However, as we can see, the mapping function is
derived from one specific illumination, which means the map-
ping function can only accurately correct the images that are
captured under exactly the same illumination. In the ideal situa-
tion, one mapping function needs to be calculated per illumina-
tion. However, this is impossible to fulfill in the real situation. To
address this problem, two CSTs for two fixed illuminations that
are selected far apart in terms of correlated color temperature
(CCT) are generally precomputed in the factory. The actual
CST applied to the image is interpolated from the two precom-
puted CSTs according to the estimated illumination.
[127]
However, because the interpolation process is based on estima-
tion, the color reproduction accuracy is greatly affected.
5.1.3. Gamma Correction
Humans perceive light differently from the camera. Most of the
camera sensors record light linearly, which means when twice
the number of photons hit the sensor, it outputs twice the signal.
However, this is not the case for human eyes. Humans do not
perceive light twice as the brightness as the light which is one-
half of the luminance. Instead, they are perceived to be much
closer (Figure 10a). Therefore, if we directly store the signal that
the camera received, many storage bits would be used to describe
the brighter tones that humans perceive with less sensitivity, and
fewer bits would be devoted to the darker tones that are more
valuable for human perception. To address this problem, a non-
linear gamma encoding is applied to expand the bits used for
describing darker tones and compress the bits used for describ-
ing lighter tones (Figure 10b). The encoding gamma is defined as
Vout ¼AVγ
in (5)
where γ<1, V
in
is the input signal after CST, Ais a constant and
generally equal to 1.
Because gamma encoding is performed before the captured
image is converted to a JPEG file, a set of linear colorimetric sig-
nals would not appear linear anymore in the image. This is one of
the reasons why colorimetric signals recorded by smartphones
Figure 10. a) Difference between camera response and human eye
perception. b) Comparison between linearly encoded color signal and
gamma-encoded color signal. More bits are used for storing weaker light
signals for which humans are more sensitive after gamma encoding.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (19 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

usually deviate from Beer’s law. One confusing point is that the
screen still needs to display the actual luminance same as the
luminance recorded by the camera sensor so as to reproduce
the scene for natural human perception. Therefore, after gamma
encoding, the monitor needs to perform gamma decoding before
displaying the image. Decoding gamma has the same equation as
encoding gamma, but with γ>1. The decoding gamma of the
display needs to match the encoding gamma of the camera
(i.e., the product of encoding gamma and decoding gamma is
equal to 1), so as to generate the correct color signal. One thing
to notice is that the function of gamma is to optimize storage
efficiency, instead of color correction. Linear RAW image data
would still appear natural for human eyes but only on a linear
gamma display. The following diagram illustrates the overall
process of gamma correction (Figure 11).
Because color measurement occurs before gamma decoding,
we need to perform gamma decoding manually to recover the
linearity of signal. However, mismatched encoding gamma
and decoding gamma would fail in achieving this goal.
Unfortunately, the applied gamma correction value varies
depending on ambient light and other automated settings,
and its value is not easy to get access as the applied gamma
curves are usually proprietary.
5.1.4. Postprocessing
For aesthetic purposes, the images usually go through a series of
modifications, such as brightness modulation, brilliance
enhancement, saturation manipulation, contrast adjustment,
noise reduction, etc. Those modifications are mostly nonlinear,
and modification methods and specific parameters differ signifi-
cantly with manufacturers. Therefore, it is difficult to reverse this
process without a re-calibration step (e.g., imaging a color chart).
5.2. Color Correction
Many methods have been proposed to enhance color reproduc-
tion accuracy by improving the first stage or reversing the second
stage of ISP.
5.2.1. Color Correction Targeting on the First Stage
Due to the complexity of postprocessing, it would be beneficial to
stop the ISP before it goes through postprocessing and read the
intermediary pixel values. However, this is complicated by the
lack of access to the camera hardware, because the first stage
is applied onboard the smartphone. Some studies used the
screenshot captured in the preview or video mode to circumvent
device-specific postprocessing of images.
[128]
Karaimer and
Brown
[129]
addressed this restriction further by introducing a
software-based camera emulator that is compatible with a wide
variety of cameras. This software platform is able to stop the ISP
pipeline at intermediate steps and allows the acquisition and
modification of intermediary pixel values. For different imaging
devices, the color value of images right after WB and CST should
show higher constancy than those after postprocessing.
Figure 11. The overall process of gamma correction. Linear raw color signal is stored nonlinearly in a JPEG file. The display performs gamma decoding.
Because the net effect of gamma encoding and gamma decoding is linear, the monitor emits the same light as the original scene. Therefore, the picture
displayed on the monitor appears naturally to humans. However, because the measurement step occurs before gamma decoding by computer monitor,
gamma correction needs to be performed manually to restore the linearity of color signal.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (20 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

However, as mentioned above, the potency of existent WB and
CST is not enough to eliminate the interphone image color var-
iations. Therefore, several methods have been raised to improve
WB and CST. Karaimer et al.
[113]
proposed a method of incorpo-
rating an additional calibrated illumination with a CCT at
5000°K into the interpolation process and a second method
of using a full-color balance matrix and a fixed CST. A
Bayesian classifier is trained to estimate the full-color balance
matrix for camera images captured under arbitrary lighting con-
ditions. The performance of mobile phone cameras was
improved up to 33% and 59% for the first method and the second
method, respectively. The first method runs fast and can be easily
incorporated into the existing camera pipeline, while the second
method involves ML, which means it is not suitable for use
onboard a camera but can be helpful to subsequent image analy-
sis on the computer. Finlayson et al.
[118]
applied a spectral sharp-
ening transform in the form of a 3 3 full matrix to enhance the
performance of WB. To estimate the sharpening matrix, it was
necessary to capture materials with known reflectance spectra
under different lighting conditions. Chong et al.
[130]
refined this
method by directly solving the sharpening matrix using the spec-
tral sensitivities of the camera. Unfortunately, camera manufac-
turers generally do not provide detailed information on spectral
sensitivities, so the burden of obtaining them is imposed on the
users. Notable methods include direct approaches and indirect
approaches. Direct approaches include recording the camera
response and spectroradiometer response to light at each wave-
length. Thus, a monochromator is generally used to generate
monochromatic light (Figure 12).
[131]
Although this method
ensures high accuracy, it is time-consuming and costly. Soda
et al.
[132]
extended this method by replacing the monochromator
with a movie that swept the wavelength of the light emitted from
the screen in a predefined range and rate. Indirect approaches
refer to deriving the spectral sensitivity by imaging a color chart
under known illumination. This method faces the problem of
solving high-dimensional matrices, and the matrices are seri-
ously rank deficient even with a large number of color patches.
Tominaga et al.
[133]
improved the estimation efficiency and accu-
racy using a small number of color patches by referring to the
smartphone camera spectral sensitivity database and extracting
the features of spectral function shapes. Jiang et al.
[134]
applied
principal component analysis to the database and proposed two
methods to estimate the spectral sensitivities for both known and
unknown illumination. Those approaches that combine direct
methods and indirect methods are described as being more
accurate than conventional indirect methods.
[135]
If we can image a color chart under the same scene of the
imaging target, and have access to the raw data, the color correc-
tion would be easy by deriving a mapping from sensor RGB
directly to CIE XYZ color space without the need for illumination
correction. Notably, Funt and Bastani
[136]
solved the problem of
color calibration under nonuniform illumination across the color
chart using a numerical optimizer. They later proposed a faster
and easier technique based on least-squares regression on the
unit sphere.
[137]
Hong et al.
[138]
tested different polynomial trans-
fer matrices for converting device-dependent color space to CIE
XYZ color space using least-squares fitting technique. It was
described that the polynomial model with higher order yielded
better results than 3 3 linear transforms. However, polynomial
color correction is more susceptible to camera exposure
compared with linear color correction, as the exposure alters
the vector of polynomial matrix in a nonlinear way, which leads
to hue and saturation shifts.
[139]
Finlayson et al.
[139]
later pro-
posed a fractional polynomial to address this problem by taking
the k-th root of each k-degree polynomial term. Bianco et al.
[140]
pointed out that color correction transform would amplify the
illuminant estimation errors. They demonstrated that it was
possible to mitigate the error amplification by considering the
probability distribution of the illumination estimation algorithm.
5.2.2. Color Correction Targeting on the Second Stage
Methods regarding the reversal of the second stage to undo non-
linear processing have been extensively reported. Like the deri-
vation of CST, we can use a color chart to restandardize the image
color. Before the full color of the image is corrected using all col-
ors in the color chart, a gamma correction is usually applied
using the grayscale patches of the imaged color checker.
Takahashi et al.
[141]
created a color chart to increase the color
accuracy of telemedicine, where the patient’s skin or tongue
color was examined. They applied gamma correction and then
color correction for the images captured by patients with a color
chart placed aside using the algorithm of multiple regression.
They also considered that the display’s performance varied
depending on the model, so the same color could not be output
even if the same RGB was input. Therefore, a color correction for
doctors’displays was also performed using the color chart and a
colorimeter to ensure the display color reproducibility. You
et al.
[142]
proposed a low-cost, contactless chicken meat quality
evaluation method by examining the color image of chicken
meat. The image color was corrected using a multivariate linear
regression model with reference to a color chart placed beside the
meat. Polynomial models with different orders and parameters
were tested. It was shown that a small number of elements in the
model were unable to correct the color well, while a large number
of elements increased the calculation burden. After experiments
on a set of sample images, a second-order polynomial model of
nine parameters was adopted. Moreover, a new color card locali-
zation method without complicated computation was proposed
with steps of global three-channel thresholding, contour extrac-
tion, and color block localization (Figure 2a). Kim et al.
[55a]
pro-
posed a smartphone-based colorimetric pH detection method
Figure 12. Experimental setups for measuring the spectral sensitivity of
smartphone cameras using a monochromator and a spectroradiometer.
Reproduced with permission.
[133]
Copyright 2021, MDPI.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (21 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

using a color adaptation algorithm performed in CIELUV color
space. A 3D-printed mini light box and a paper-printed color
chart were prepared. A third-order polynomial model of 20
parameters was used for color correction. They used the mini-
mum absolute deviation solution obtained by iterative
reweighted least square to solve the parameters. Similarly, an
automatic color-retrieving algorithm based on contour extraction
was also proposed to obtain all the color values of the captured
image automatically (Figure 2c). Zhang et al.
[143]
combined
images captured with a smartphone-based microscope with
those captured by a lens-free holographic microscope to generate
color images of specimens with high resolution. A polynomial
regression model was used for color correction (Figure 13).
Specifically, the images were first normalized and white balanced
using an empty calibration image taken without a sample. The
color information was then transformed from RGB color space to
LAB color space. A fitting function used for lightness correction
or gamma correction was calculated according to the L compo-
nent of the output and the L component of the ground truth. The
saturation of the image was enhanced to match the ground truth
by appropriately scaling the chroma component (C¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
A2þB2
p).
The scaling factor was calculated in the least-square sense.
Finally, a second-order polynomial with 12 parameters in LAB
color space was used for color correction. Cugmas et al.
[42]
sup-
plemented a typical veterinary teledermoscopy system with a con-
ventional color calibration procedure and studied two midpriced
smartphones in evaluating native and erythematous canine skin
color. It was claimed that comparably accurate colorimetrical
measurements could be achieved with a simple image normali-
zation on the white surface without the need for time-consuming
full-color calibration. They also used the linear regression model
for color correction but with first-order and only four parameters.
They did not recommend using a high-order polynomial with a
small number of measurements because it could easily lead to
overfitting. Similarly, Cebrián et al.
[144]
tested three different
colorimetric analyses using different types and brands of smart-
phones with the assistance of a 3D-printed light shield box. They
concluded that a linear least-squares fit was enough to achieve
color constancy. Overall, imaging a color chart and performing
color correction using a polynomial regression model is mostly
used to reverse the second-stage image manipulation. The spe-
cific model and parameters may depend on different assays. A
light shield box may help to simplify the correction algorithms
and improve calculation efficiency. Some researchers also found
that brighter images with higher contrast produced with longer
exposure time but without inducing clipping can sometimes
contribute to better color constancy.
[42]
Apart from using polynomial regression models, ML can also
contribute to the color correction. One possible strategy for com-
bining color calibration with ML is to design a small neural net-
work that can determine the calibration function for each image.
The general procedure for color correction is basically to find a
3D function fðR,G,BÞthat can output the pixel value of each
pixel after calibration. Suppose the ground truth pixel value of
a pixel is R
g
,G
g
,B
g
, and the actual pixel value is R,G,B, and
then the value of the calibration function at this point should be
fðR,G,BÞ¼½Rg,Gg,Bg(6)
In linear regression-based calibration, the function fðR,G,BÞ
is assumed to be a linear function, while in polynomial regres-
sion-based models, the function fðR,G,BÞis assumed to be a
polynomial function. If we consider the actual pixel values as
the input and the ground truth value as label, we can train a neu-
ral network for each image that can have a calibration function
with no fixed type. Considering the inconsistency in the camera’s
photosensitive elements and ISP, a flexible calibration method is
theoretically better than a fixed regression method. For the ML
calibration method to work, the number of reference colors with
ground truth needs to be large, meaning the image needs to be
taken with many standard colors from all possible colors the cam-
era can detect. Also, to prevent overfitting, part of the dataset
needs to be separated to form a validation set and stop the train-
ing process when the validation error is too large. By doing so, we
could train a neural network that will calibrate one single image.
The advantage of this approach is that the accuracy of color
calibration should be better than regression-based methods.
The drawback is that the model needs to be trained for every sin-
gle image from a specific camera, which is not efficient. Thus, a
unified feature mapping that is insensitive to color types may be a
promising solution in the future, especially using some
pretrained models.
Regarding the design of the color rendition chart, it could have
different colors and a different number of patches. Patterns of
the color chart can be optimized to fit different applications.
Contrary to common sense that more color patches result in bet-
ter color correction results, Akkaynak et al.
[35]
demonstrated that
the transformation error stopped decreasing after the inclusion
of the 18th patch. They also demonstrated that using patches
Figure 13. An example of color correction using polynomial regression.
Reproduced with permission.
[143]
Copyright 2016, Springer Nature.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (22 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

whose radiance spectra span the subspace of those in the scene
yielded the most accurate transforms. If the radiance spectra of
the scene only span a small range of wavelength (e.g., in a forest
that is sufficient in shades of greens and browns but rare in other
colors), the normal color chart that is abundant in chromaticity
but not specific in one color may act poorly in this situation. For
most cases of colorimetric biosensing, the changes in color
signals are small and only occur in one or two hues. There is
no need to correct the colors out of the test region. Therefore,
to increase the accuracy of testing results and also reduce the
calculation burden, a color calibration target specific to the detec-
tion signal is usually developed. For example, Takahashi et al.
[141]
only used the color patches that have similar colors to the human
skin and tongue for examining skin or tongue (Figure 14a). And
Cugmas et al.
[42]
only used the white and skin patches on the
Digital SG ColorChecker for skin color analysis (Figure 14b).
Note that the inclusion of grayscale patches is necessary because
they are important for WB and gamma correction while using
grayscale patches solely is not enough to correct chromatic
colors.
Manufacturing a color chart is not as easy as printing a color-
ful picture on paper. The surface of the color chart is better as a
Lambertian surface, which means it reflects light equally from
different angles. Especially, the glossiness should be avoided oth-
erwise it would damage the correct color acquisition (Figure 14c).
Most importantly, the reflectance of each color chart needs to be
the same as a second color chart produced in different batches
(i.e., the color chart needs to be reproducible).
There is research trying to correct the image color without
using color charts. Escobedo et al.
[145]
performed color correction
with only reference to a full black spot and a full white spot using
the wrong von Kries model. The fitting function responded well
to the Gamma correction curve. Hu et al.
[146]
used SVM to predict
the illumination condition and the corresponding color correc-
tion matrix with reference to the color difference of images taken
with/without a flashlight. It was assumed that taking an image
with a flashlight would just be like imposing an intensity value on
each channel of the image taken without a flashlight under the
same environment. Nevertheless, the imposed intensity vector
would be different if the images were captured under different
illumination conditions. Therefore, the color difference between
images captured with/without flash could be used to estimate the
current lighting environment. Without the illumination estima-
tion process, Lee et al.
[147]
leveraged the difference between two
Figure 14. Color checkers used for different applications. a) Designed color checker for tongue color analysis. Reproduced with permission.
[141]
Copyright
2020, International Society of Artificial Life and Robotics. b) Skin patches selected for skin color analysis (in red square). Reproduced with permission.
[42]
Copyright 2020, MDPI. c) A bad case of color checker imaging. Reproduced with permission.
[142]
Copyright 2020, IEEE.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (23 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

images captured under auto WB mode and a preset mode,
respectively, to derive the color conversion function. Although
those methods may work well for illumination correction, their
performance on different smartphone models was nottested.
6. Conclusion and Future Perspectives
Smartphones are ubiquitously available nowadays. Combined
with its merits of low cost, high portability, and high versatility,
it shows great advantages over cumbersome laboratory equip-
ment in the fields of telemedicine, e-health, environmental mon-
itoring, or any scenarios of point-of-need detection. Especially,
smartphone-based optical detection systems have been of
increasing interest in recent years due to their simplicity in col-
orimetric signal measurement. However, most of the studies
only focus on biosensing technology development and slightly
on the difficulties in accurate signal acquisition in different envi-
ronments. The color of images captured by smartphones is
highly susceptible to illumination conditions, camera spectral
sensitivity, and ISP pipeline. A small error in color information
could lead to an unacceptable concentration error for the analyte,
depending on the sensitivity of assays. Achieving color constancy
is therefore of great importance. In this review, we summarize
and discuss four types of recently developed methods targeting
color constancy: 1) adapter attachment, 2) CST, 3) ML, and 4v)
color correction. The adapter effectively excludes ambient light
and provides a uniform and homogenous light source. It ensures
the same illumination condition and reduces color variation to a
great extent. The development of a miniature adapter and an APP
with a color-retrieving algorithm is highly applicable to real cases.
Selecting an appropriate color space can somehow improve the
biosensor performance. For example, it has been verified that
hue is more resistant to illumination variations and sensitive
to chromatic changes, and LAB color space is more sensitive
to small color changes than RGB. ML is a potential candidate
for improving detection capability, which has been adopted
recently to overcome illumination variation and interphone vari-
ation. Hitherto, ML is mostly performed on computers or cloud
servers, on account of the relatively low-computing power of
smartphones. Because ML-based colorimetric biosensing is
mostly based on classifications, an extensive amount of training
data is required to obtain results with high precision. ML could
significantly increase the cost of POC diagnosis and may not be
suitable for resource-limited regions and the developing world.
Nevertheless, with the development of high-performance smart-
phone processors and ML algorithms, it is anticipated that smart-
phones will be able to perform ML with high speed and low cost
in the near future. For example, there have been many light-
weight networks specially designed for mobile phones, including
MobileNets,
[148]
EfficientNet,
[149]
and MixNet,
[150]
all of which
perform well on mobile devices. To further improve the detection
limit using ML, efforts should be made in an attempt to build
better datasets and to design more suitable model structures.
In the case of smartphone-based optical sensors, an ideal dataset
should have images from different models of smartphones in
different light environments. The dataset should also be aug-
mented with CSTs and shape transformations. This will not only
increase the size of the dataset but also make the readout
insusceptible to the types of smartphones. As for the model
design, current studies mainly modified the existing models,
which may not be optimal for the task of readout of POCT bio-
sensors. The semantic context of the image should be taken into
consideration when developing a neural network model. In the
case of POCT biosensor images, the features are mainly in low
levels,
[151]
and the region of interest (ROI) is usually in fixed posi-
tions. The design of U-net
[152]
considers the characteristics of bio-
medical images, and it performs better than other structures in
biomedical image segmentation tasks. If a new model is
designed specifically for the readout of POCT biosensors, it
may improve the detection limit of current methods.
Last but not least, color correction based on a color chart
showed the greatest promise in the elimination of illumination
variation and interphone variation with a relatively simple algo-
rithm and high effectiveness. To further improve the detection
limit, color correction method could be combined with ML algo-
rithms. Current color correction algorithms are based on a fixed
calibration model, which assumes the relationship of a pixel
value before and after calibration fits linear or polynomial func-
tions. Each pixel value of a photo is determined by light intensity,
the characteristics of photosensitive elements, and the ISP, so a
fixed model may not be suitable for all models of smartphones. If
ML is applied in color calibration, it could generate a nonlinear
mapping function for color correction, which maps all types of
color into a unified feature space. Thus, it may eliminate the
influence of camera inconsistency and improve detection accu-
racy. Moreover, the raw data after stage one of ISP should show
higher color constancy than what goes through the second stage
of ISP. However, most studies were focused on the reversal of the
second stage due to the lack of access to raw image data.
Although there was some research targeting the improvement
of the first stage of ISP, few studies used the raw data of the cam-
era for biosensing purposes. The summarized four strategies can
be blended with each other (e.g., using an adapter and color cor-
rection together ensures much higher accuracy of image color
than using color correction alone
[144]
) and better performance
of smartphone-based optical detection system is expected in
the near future. Although smartphone-based POCT technology
has the advantages of being rapid, low cost, and easy populariza-
tion, its performance in terms of detection sensitivity, repeatabil-
ity, and throughput awaits for further improvement compared to
the automatic detection platforms in laboratory. Besides these
emerging technologies for optical signal outputs, advances
[153]
in ultrasensitive assays (such as CRISPR/Cas biosensing,
[154]
nanomaterials-facilitated biosensors, et al.), printable biosensors
(such as standardization of biosensor fabrication and prepara-
tion), and microfluidics (such as sensor arrays) will be beneficial
to achieve desirable performance for smartphone-enabled
point-of-care diagnostics.
Acknowledgements
The work was financially supported by the National Natural Science
Foundation of China (grant 22174121 and 22211530067), Shenzhen
Bay Open Laboratory Fund 2021 by Shenzhen Bay Laboratory, and the
University Development Fund (UDF01002012) of Chinese University of
Hong Kong, Shenzhen.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (24 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

Conflict of Interest
The authors declare no conflict of interest.
Author Contributions
G.L. proposed the contents of the article while providing administration
and supervision. K.F. and G.L. designed the structure of the article. K.F.
drafted the initial manuscript, drafting, and revisions. W.L., Y.M., and
Z.L. provided the drafting and revisions of the part of ML. All authors
further edited and adapted the manuscript and approved the final version
of the article.
Keywords
e-health, illumination variations, machine learning, optical signal
corrections, point-of-care diagnostics, smartphone based digital detection
Received: September 5, 2022
Revised: November 27, 2022
Published online: February 21, 2023
[1] J. Ma, S. Q. Yan, C. Y. Miao, L. M. Li, W. W. Shi, X. M. Liu, Y. Luo,
T. J. Liu, B. C. Lin, W. M. Wu, Y. Lu, Adv. Healthcare Mater. 2019,8,
1801084.
[2] Y. Shen, T. B. Anwar, A. Mulchandani, Sens. Actuators Rep. 2021,3,
100025.
[3] D. C. Christodouleas, B. Kaur, P. Chorti, ACS Central Sci. 2018,4,
1600.
[4] S. M. Khan, A. Gumus, J. M. Nassar, M. M. Hussain, Adv. Mater.
2018,30, 1705759.
[5] J. Hu, X. Cui, Y. Gong, X. Xu, B. Gao, T. Wen, T. J. Lu, F. Xu, Biotechnol.
Adv. 2016,34, 305.
[6] S. C. Kishore, K. Samikannu, R. Atchudan, S. Perumal,
T. N. J. I. Edison, M. Alagan, A. K. Sundramoorthy, Y. R. Lee,
Chemosensors 2022,10, 55.
[7] V. Mody, V. Mody, S. Parekh, in Int. Conf. on Smart Electronics and
Communication (ICOSEC), IEEE, Piscataway, NJ 2020, pp. 830–836.
[8] P. Daponte, L. De Vito, F. Picariello, M. Riccio, Measurement 2013,46,
3291.
[9] S. Di Nonno, R. Ulber, Analyst 2021,146, 2749.
[10] J. Luo, M. Wu, D. Gopukumar, Y. Zhao, Biomed. Inform. Insights 2016,
8, BII. S31559.
[11] S. Zhang, Z. Li, Q. Wei, Nanotechnol. Precis. Eng. 2020,3, 32.
[12] L. Huang, S. Tian, W. Zhao, K. Liu, X. Ma, J. Guo, Analyst 2020,145,
2828.
[13] K. Iitani, S. S. Ramamurthy, X. Ge, G. Rao, Curr. Opin. Biotechnol.
2021,71, 198.
[14] M. Nehra, N. Dilbaghi, R. Kumar, S. Kumar, Mater. Lett. 2022,308,
131235.
[15] T. Mahmoudi, M. de la Guardia, B. Baradaran, Trends Analyt. Chem.
2020,125, 115842.
[16] S. Kamel, T. A. Khattab, Biosensors 2020,10, 67.
[17] F. Mi, C. Hu, Y. Wang, L. Wang, F. Peng, P. Geng, M. Guan, Analyt.
Bioanalyt. Chem. 2022,1, 2883.
[18] Z. Geng, X. Zhang, Z. Fan, X. Lv, Y. Su, H. Chen, Sensors 2017,17,
2449.
[19] T. Alawsi, Z. Al-Bawi, Eng. Rep. 2019,1, 12039.
[20] J. Nelis, A. Tsagkaris, M. Dillon, J. Hajslova, C. Elliott, Trends Analyt.
Chem. 2020,129, 115934.
[21] A. Roda, M. Guardigli, D. Calabria, M. M. Calabretta, L. Cevenini,
E. Michelini, Analyst 2014,139, 6494.
[22] a) M. Nixon, F. Outlaw, T. S. Leung, PLoS One 2020,15, 0230561;
b) A. Skandarajah, C. D. Reber, N. A. Switz, D. A. Fletcher, PloS
One 2014,9, 96906.
[23] Z. Liang, J. Cai, Z. Cao, L. Zhang, IEEE Trans. Image Process. 2021,30,
2248.
[24] B. Khanal, P. Pokhrel, B. Khanal, B. Giri, ACS Omega 2021,6, 33837.
[25] a) S. Kim, H. J. Chun, K. W. Lee, H. C. Yoon, Biosens. Bioelectron. 2020,
150, 111932; b) K. R. Kim, K. W. Lee, H. J. Chun, D. Lee, J.-H. Kim,
H. C. Yoon, Biosensors Bioelectron. 2022,196, 113722.
[26] a) X. Wei, T. Tian, S. Jia, Z. Zhu, Y. Ma, J. Sun, Z. Lin, C. J. Yang,
Analyt. Chem. 2016,88, 2345; b) C. Jiang, F. Wang, K. Zhang,
T. Min, D. Chen, Y. Wen, ACS Sensors 2021,6, 2395; c) T. Tian,
J. Li, Y. Song, L. Zhou, Z. Zhu, C. J. Yang, Lab Chip 2016,16, 1139.
[27] D. Quesada-González, A. Merkoçi, Biosens. Bioelectron. 2017,92, 549.
[28] J. Liu, Z. Geng, Z. Fan, J. Liu, H. Chen, Biosens. Bioelectron. 2019,132,
17.
[29] B. R. Sun, A. G. Zhou, X. Li, H.-Z. Yu, ACS Sensors 2021,6, 1731.
[30] H. Kholafazad-Kordasht, M. Hasanzadeh, F. Seidi, Trends Analyt.
Chem. 2021,145, 116455.
[31] Ö. Kap, V. Kılıç, J. G. Hardy, N. Horzum, Analyst 2021,146, 2784.
[32] J. Yang, K. Wang, H. Xu, W. Yan, Q. Jin, D. Cui, Talanta 2019,202, 96.
[33] P. C. Biswas, S. Rani, M. A. Hossain, M. R. Islam, J. Canning, IEEE J.
Select. Topics Quantum Electron. 2021,27,1.
[34] Q. Qin, K. Wang, J. Yang, H. Xu, B. Cao, Y. Wo, Q. Jin, D. Cui, Analyst
2019,144, 5659.
[35] D. Akkaynak, T. Treibitz, B. Xiao, U. A. Gürkan, J. J. Allen, U. Demirci,
R. T. Hanlon, J. Opt. Soc. Am. A Opt. Image Sci. Vis. 2014,31, 312.
[36] M. E. Solmaz, A. Y. Mutlu, G. Alankus, V. Kılıç, A. Bayram,
N. Horzum, Sens. Actuators B: Chem. 2018,255, 1967.
[37] T. San Park, J.-Y. Yoon, IEEE Sens. J. 2014,15, 1902.
[38] a) J. Xu, X. Chen, H. Khan, L. Yang, Sens. Actuators B: Chem. 2021,
335, 129707; b) T. Wang, Q. Mei, Z. Tao, H. Wu, M. Zhao, S. Wang,
Y. Liu, Biosens. Bioelectron. 2020,148, 111791; c) Y. Bai, Y. He,
Y. Wang, G. Song, Microchim. Acta 2021,188, 1; d) J. Wang,
D. Li, Y. Qiu, X. Liu, L. Huang, H. Wen, J. Hu, Talanta 2020,220,
121377; e) M. Xu, W. Huang, D. Lu, C. Huang, J. Deng, T. Zhou,
Analyt. Methods 2019,11, 4267; f ) S. Chu, H. Wang, X. Ling,
S. Yu, L. Yang, C. Jiang, ACS Appl. Mater. Interfaces 2020,12, 12962.
[39] V. Kılıç, G. Alankus, N. Horzum, A. Y. Mutlu, A. Bayram,
M. E. Solmaz, ACS Omega 2018,3, 5531.
[40] a) A. S. Hidayat, H. Horino, I. I. Rzeznicka, Sensors 2021,21, 4751;
b) G.-R. Han, H. J. Koo, H. Ki, M.-G. Kim, ACS Appl. Mater. Interfaces
2020,12, 34564.
[41] B. Berg, B. Cortazar, D. Tseng, H. Ozkan, S. Feng, Q. Wei,
R. Y.-L. Chan, J. Burbano, Q. Farooqui, M. Lewinski, D. Di Carlo,
O. B. Garner, A. Ozcan, ACS Nano 2015,9, 7857.
[42] B. Cugmas, E. Štruc, Sensors 2020,20, 6234.
[43] S. Ghosh, K. Aggarwal, T. Nguyen, J. Han, C. H. Ahn, Microsyst.
Nanoeng. 2020,6,1.
[44] Z. Iqbal, R. B. Bjorklund, Talanta 2011,84, 1118.
[45] D. Kim, S. Kim, H.-T. Ha, S. Kim, Curr. Appl. Phys. 2020,20, 1013.
[46] a) Y. Xing, Q. Zhu, X. Zhou, P. Qi, Sens. Actuators B: Chem. 2020,319,
128254; b) C. Khamkhajorn, S. Pencharee, J. Jakmunee,
N. Youngvises, Microchem. J. 2022,174, 107055.
[47] Q. Fu, Z. Wu, F. Xu, X. Li, C. Yao, M. Xu, L. Sheng, S. Yu, Y. Tang, Lab
Chip 2016,16, 1927.
[48] Y. M. Park, Y. D. Han, H. J. Chun, H. C. Yoon, Biosens. Bioelectron.
2017,93, 205.
[49] a) D. Calabria, M. Zangheri, I. Trozzi, E. Lazzarini, A. Pace,
M. Mirasoli, M. Guardigli, Biosensors 2021,11, 381; b) L. Montali,
M. M. Calabretta, A. Lopreside, M. D’Elia, M. Guardigli,
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (25 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

E. Michelini, Biosens. Bioelectron. 2020,162, 112232;
c) M. Zangheri, F. Di Nardo, D. Calabria, E. Marchegiani,
L. Anfossi, M. Guardigli, M. Mirasoli, C. Baggiani, A. Roda, Analyt.
Chim. Acta 2021,1163, 338515.
[50] A. Roda, M. Zangheri, D. Calabria, M. Mirasoli, C. Caliceti,
A. Quintavalla, M. Lombardo, C. Trombini, P. Simoni, Sens.
Actuators B: Chem. 2019,279, 327.
[51] a) M. Bhaiyya, P. K. Pattnaik, S. Goel, Microchim. Acta 2022,189,1;
b) S. M. Khoshfetrat, P. Seyed Dorraji, M. Shayan, F. Khatami,
K. Omidfar, Analyt. Chem. 2022,94, 8005; c) M. L. Bhaiyya,
S. Gangrade, P. K. Pattnaik, S. Goel, IEEE Trans. Instrum. Measur.
2022,71,1.
[52] H. Kim, Y. Jung, I.-J. Doh, R. A. Lozano-Mahecha, B. Applegate,
E. Bae, Sci. Rep. 2017,7,1.
[53] E. Michelini, M. M. Calabretta, L. Cevenini, A. Lopreside,
T. Southworth, D. M. Fontaine, P. Simoni, B. R. Branchini,
A. Roda, Biosens. Bioelectron. 2019,123, 269.
[54] M.-Y. Lu, W.-C. Kao, S. Belkin, J.-Y. Cheng, Sensors 2019,19, 3882.
[55] a) S. D. Kim, Y. Koo, Y. Yun, Sensors 2017,17, 1604; b) N. Lopez-Ruiz,
V. F. Curto, M. M. Erenas, F. Benito-Lopez, D. Diamond, A. J. Palma,
L. F. Capitan-Vallvey, Analyt. Chem. 2014,86, 9554.
[56] Y. Ma, J. Bao, Y. Zhang, Z. Li, X. Zhou, C. Wan, L. Huang, Y. Zhao,
G. Han, T. Xue, Cell 2019,177, 243.
[57] Z. Zhu, Y. Sun, T. Ma, D. Tian, J. Zhu, Analyt. Bioanalyt. Chem. 2021,
413, 3291.
[58] C. Liang, Y. Liu, A. Niu, C. Liu, J. Li, D. Ning, Lab Chip 2019,19, 1797.
[59] W. I. Lee, S. Shrivastava, L. T. Duy, B. Y. Kim, Y. M. Son, N. E. Lee,
Biosens. Bioelectron. 2017,94, 643.
[60] R. Bakhtiar, J. Chem. Educ. 2013,90, 203.
[61] Z. Fan, Z. Geng, W. Fang, X. Lv, Y. Su, S. Wang, H. Chen, Sensors
2020,20, 446.
[62] J.-G. Walter, L. S. M. Alwis, B. Roth, K. Bremer, Sensors 2020,20, 2889.
[63] A. X. Wang, X. Kong, Materials 2015,8, 3024.
[64] D. P. Dos Santos, M. L. Temperini, A. G. Brolo, Acc. Chem. Res. 2019,
52, 456.
[65] I. Abdulhalim, M. Zourob, A. Lakhtakia, Electromagnetics 2008,28,
214.
[66] B. A. Prabowo, A. Purwidyantri, K.-C. Liu, Biosensors 2018,8, 80.
[67] C. Rathmell, D. Bingemann, M. Zieg, D. Creasey, Port. Spectrosc.
Spectrom. 2021, 115.
[68] T. Mu, S. Wang, T. Li, B. Wang, X. Ma, B. Huang, L. Zhu, J. Guo, IEEE
J. Select. Topics Quantum Electron. 2018,25,1.
[69] K. Plataniotis, A. N. Venetsanopoulos, in Color Image Processing and
Applications, Springer Science & Business Media, New York 2000.
[70] D. A. Kerr, Colorimetry 2010,1,1.
[71] W. Gao, S. Ma, Advanced Video Coding Systems, Springer, New York
2014.
[72] M. Tkalcic, J. F. Tasic, Colour Spaces: Perceptual, Historical and
Applicational Background, Vol. 1, IEEE, Piscataway NJ 2003, 304.
[73] D. H. Brainard, Sci. Color 2003,2,5.
[74] L. M. Hurvich, D. Jameson, Psychol. Rev. 1957,64, 384.
[75] R. Yang, W. Cheng, X. Chen, Q. Qian, Q. Zhang, Y. Pan, P. Duan,
P. Miao, ACS Omega 2018,3, 12141.
[76] D. Baş,Analyt. Methods 2017,9, 6698.
[77] T. Komatsu, S. Mohammadi, L. S. A. Busa, M. Maeki, A. Ishida,
H. Tani, M. Tokeshi, Analyst 2016,141, 6507.
[78] M. Cerrato-Alvarez, S. Frutos-Puerto, C. Miró-Rodríguez, E. Pinilla-
Gil, Microchem. J. 2020,154, 104535.
[79] H. Q. Nguyen, V. D. Nguyen, H. Van Nguyen, T. S. Seo, Sci. Rep.
2020,10,1.
[80] K. Cantrell, M. M. Erenas, I. de Orbe-Payá, L. F. Capitán-Vallvey,
Analyt. Chem. 2010,82, 531.
[81] S. T. Krauss, A. Q. Nauman, G. T. Garner, J. P. Landers, Lab Chip
2017,17, 4089.
[82] D. R. Bagal-Kestwal, B. H. Chiang, Sens. Actuators B: Chem. 2021,
339, 129836.
[83] J. Yang, F. Shen, T. Wang, M. Luo, N. Li, S. Que, Geoderma 2021,
402, 115365.
[84] A. A. Shalaby, A. A. Mohamed, Water Air Soil Pollut. 2020,231, 516.
[85] J. L. D. Nelis, L. Bura, Y. Zhao, K. M. Burkin, K. Rafferty, C. T. Elliott,
K. Campbell, Sensors 2019,19, 5104.
[86] B. Mahesh, Int. J. Sci. Res. 2020,9, 381.
[87] V. Do˘
gan, E. Yüzer, V. Kılıç, M. Şen, Analyst 2021,146, 7336.
[88] A. Y. Mutlu, V. Kılıç, G. K. Özdemir, A. Bayram, N. Horzum,
M. E. Solmaz, Analyst 2017,142, 2434.
[89] H. Wang, D. Hu, in Int. Conference on Neural Networks and Brain,
Vol. 1, IEEE, Piscataway, NJ 2005, pp. 279–283.
[90] H. Kim, O. Awofeso, S. Choi, Y. Jung, E. Bae, Appl. Opt. 2017,56, 84.
[91] R. Luo, G. Ma, S. Bi, Q. Duan, J. Chen, Y. Feng, F. Liu, J. Lee, Analyst
2020,145, 2197.
[92] S. Lei, H. Zhang, K. Wang, Z. Su, 2018.
[93] C. Shorten, T. M. Khoshgoftaar, J. Big Data 2019,6,1.
[94] C. Bowles, L. Chen, R. Guerrero, P. Bentley, R. Gunn, A. Hammers,
D. A. Dickie, M. V. Hernández, J. Wardlaw, D. Rueckert, arXiv
preprint arXiv:1810.10863 2018.
[95] H. Chen, J. Chen, J. Ding, IEEE Trans. Reliab. 2021,70, 831.
[96] T. Emmanuel, T. Maupong, D. Mpoeleng, T. Semong, B. Mphago,
O. Tabona, J. Big Data 2021,8,1.
[97] O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma,
Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, Int. J. Comput.
Vis. 2015,115, 211.
[98] C. Kertész, arXiv preprint arXiv:2103.16324 2021.
[99] J. Ding, X. Li, X. Kang, V. N. Gudivada, J. Data Inform. Quality 2019,
11,1.
[100] A. Canziani, A. Paszke, E. Culurciello, arXiv preprint
arXiv:1605.07678 2016.
[101] T. Yu, H. Zhu, arXiv preprint arXiv:2003.05689 2020.
[102] W. Koehrsen, Towards Data Science 2018,1.
[103] X. Ying, J. Phys. Conf. Series,2019,1168, 022022.
[104] S. Ray, in Int. Conf. on Machine Learning, Big Data, Cloud and Parallel
Computing (COMITCon), IEEE, Piscataway, NJ 2019, pp. 35–39.
[105] T. He, Z. Zhang, H. Zhang, Z. Zhang, J. Xie, M. Li, in Proc. of the
IEEE/CVF Conf. on Computer Vision and Pattern Recognition, IEEE,
Piscataway, NJ 2019, pp. 558–567.
[106] Z. Li, S. Arora, arXiv preprint arXiv:1910.07454 2019.
[107] H. Robbins, S. Monro, Ann. Math. Stat. 1951,400.
[108] T. Tieleman, G. Hinton, University of Toronto, Technical Report
2012,p.6.
[109] D. P. Kingma, J. Ba, arXiv preprint arXiv:1412.6980 2014.
[110] S. Ruder, arXiv preprint arXiv:1609.04747 2016.
[111] P. M. Radiuk, 2017.
[112] I. Kandel, M. Castelli, ICT Express 2020,6, 312.
[113] H. C. Karaimer, M. S. Brown, in 2018 IEEE/CVF Conf. on Computer
Vision and Pattern Recognition, IEEE, Piscataway, NJ 2018,
pp. 6440–6449.
[114] S. Ma, P. Hanselaer, K. Teunissen, K. A. Smet, Opt. Express 2020,28,
17266.
[115] M. Afifi, M. S. Brown, in Proc. of the IEEE/CVF Conf. on Computer
Vision and Pattern Recognition, IEEE, Piscataway, NJ 2020,
pp. 1397–1406.
[116] S. Bianco, R. Schettini, Color Res. Appl. 2010,35, 184.
[117] K. M. Lam, Ph. D. Thesis, University of Bradford 1985.
[118] G. D. Finlayson, M. S. Drew, B. V. Funt, JOSA A 1994,11, 1553.
[119] C. Li, M. R. Luo, B. Rigg, R. W. Hunt, Color Res. Appl. 2002,27, 49.
[120] G. Buchsbaum, J. Franklin Inst. 1980,310,1.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (26 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

[121] N. Bani´c, S. Lonˇcari´c, in IEEE Int. Conf. on Image Processing (ICIP),
IEEE, Piscataway, NJ 2014, pp. 605–609.
[122] B. V. Funt, V. C. Cardei, in Human Vision and Electronic Imaging IV,
Vol. 3644, SPIE, Bellingham, WA 1999, pp. 421–428.
[123] Y. Qian, S. Pertuz, J. Nikkanen, J.-K. Kämäräinen, J. Matas, arXiv
preprint arXiv:1803.08326 2018.
[124] D. A. Forsyth, Int. J. Comput. Vis. 1990,5,5.
[125] a) H.-H. Choi, H.-S. Kang, B.-J. Yun, Appl. Sci. 2020,10, 4806;
b) K. Košˇcevi´c, M. Subaši´c, S. Lonˇcari´c, IEEE Access 2020,8,
84239; c) M. Afifi, J. T. Barron, C. LeGendre, Y.-T. Tsai,
F. Bleibel, in Proc. of the IEEE/CVF Int. Conf. on Computer Vision,
IEEE, Piscataway, NJ 2021, pp. 1981–1990.
[126] A. Abdelhamed, A. Punnappurath, M. S. Brown, in Proc. of the IEEE/
CVF Conf. on Computer Vision and Pattern Recognition, IEEE,
Piscataway, NJ 2021, pp. 6637–6646.
[127] Y. Li, S. Chen, M. Wei, X. Zhang, Color Res. Appl. 2022,47, 605.
[128] a) W.-C. Kao, S.-H. Wang, L.-Y. Chen, S.-Y. Lin,IEEE Trans. on
Consumer Electronics, IEEE, Piscataway, NJ 2006,52, 1144–1152;
b) M. Bigas, E. Cabruja, J. Forest, J. Salvi, Microelectron. J. 2006,
37, 433.
[129] H. C. Karaimer, M. S. Brown, in European Conference on Computer
Vision, Springer, New York 2016, pp. 429–444.
[130] H. Y. Chong, S. J. Gortler, T. Zickler, in IEEE 11th Int. Conf. on
Computer Vision, IEEE, IEEE, Piscataway, NJ 2007, pp. 1–8.
[131] J. Nakamura, Image Sensors and Signal Processing for Digital Still
Cameras, CRC Press, Boca Raton, FL 2017.
[132] Y. Soda, E. Bakker, ACS Sens. 2019,4, 3093.
[133] S. Tominaga, S. Nishi, R. Ohtera, Sensors 2021,21, 4985.
[134] J. Jiang, D. Liu, J. Gu, S. Süsstrunk, in IEEE Workshop on Applications
of Computer Vision (WACV), IEEE, Piscataway, NJ 2013, pp. 168–179.
[135] G. Finlayson, M. M. Darrodi, M. Mackiewicz, J. Opt. Soc. Am. A
2016,33, 589.
[136] a) B. Funt, P. Bastani, in AIC (International Colour Association) Conf.,
Vol 5,2012; b) B. Funt, P. Bastani, Color Res. Appl. 2014,39, 540.
[137] P. Bastani, B. Funt, in Simplifying Irradiance Independent Color
Calibration, Vol 9015,2014, pp. 540–548.
[138] G. Hong, M. R. Luo, P. A. Rhodes, Color Res. Appl. 2001,26, 76.
[139] G. D. Finlayson, M. Mackiewicz, A. Hurlbert, IEEE Trans. Image
Process. 2015,24, 1460.
[140] S. Bianco, A. Bruna, F. Naccari, R. Schettini, J. Opt. Soc. Am. A 2012,
29, 374.
[141] M. Takahashi, R. Takahashi, Y. Morihara, I. Kin, K. Ogawa-Ochiai,
N. Tsumura, Artif. Life Robot. 2020,25, 370.
[142] M. You, J. Liu, J. Zhang, M. Xv, D. He, IEEE Access 2020,8, 170093.
[143] Y. Zhang, Y. Wu, Y. Zhang, A. Ozcan, Sci. Rep. 2016,6, 27811.
[144] P. Cebrián, L. Pérez-Sienes, I. Sanz-Vicente, Á. López-Molinero,
S. de Marcos, J. Galbán, Biosensors 2022,12, 341.
[145] P. Escobedo, M. M. Erenas, A. M. Olmos, M. A. Carvajal,
M. T. Chávez, M. A. L. González, J. J. Díaz-Mochón,
S. Pernagallo, L. F. Capitán-Vallvey, A. J. Palma, IEEE Access 2019,
7, 185677.
[146] M. C. Hu, M. H. Cheng, K. C. Lan, J. Med. Syst. 2016,40, 18.
[147] S.-H. Lee, J. Sensors 2016,2016, 6731572.
[148] A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang,
T. Weyand, M. Andreetto, H. Adam, arXiv preprint
arXiv:1704.04861 2017.
[149] M. Tan, Q. Le, in Int. Conf. on Machine Learning, PMLR, 2019, pp.
6105–6114.
[150] M. Tan, Q. V. Le, arXiv preprint arXiv:1907.09595 2019.
[151] W. Förstner, in Computer Vision —ECCV ’94 (Ed.: J.-O. Eklundh),
Springer Berlin Heidelberg, Berlin, Heidelberg, 1994, pp. 383–394.
[152] O. Ronneberger, P. Fischer, T. Brox, in Medical Image Computing and
Computer-Assisted Intervention (Eds.: N. Navab, J. Hornegger,
W. M. Wells, A. F. Frangi), Springer International Publishing,
Cham, 2015, pp. 234-241.
[153] G. Liu, Front. Bioeng. Biotechnol. 2021,9, 707615.
[154] Y. Li, S. Li, J. Wang, G. Liu, Trends Biotechnol. 2019,37, 730.
[155] B. Askarian, S.-C. Yoo, J. Chong, Sensors 2019,19, 3306.
[156] T. J. Moehling, D. H. Lee, M. E. Henderson, M. K. McDonald,
P.H.Tsang,S.Kaakeh,E.S.Kim,S.T.Wereley,T.L.Kinzer-Ursem,
K. N. Clayton, J. C. Linnes, Biosens. Bioelectron. 2020,167, 112497.
[157] N. Cheng, Y. Song, Q. Q. Fu, D. Du, Y. B. Luo, Y. Wang, W. T. Xu,
Y. H. Lin, Biosens. Bioelectron. 2018,117, 75.
[158] Y. Gong, Y. Zheng, B. Jin, M. You, J. Wang, X. Li, M. Lin, F. Xu, F. Li,
Talanta 2019,201, 126.
[159] F. Zeng, T. Mou, C. Zhang, X. Huang, B. Wang, X. Ma, J. Guo,
Analyst 2019,144, 137.
[160] B. Singh, D. Parwate, S. Shukla, AAPS PharmSciTech 2009,
10, 34.
[161] A. Sultangazin, J. Kusmangaliyev, A. Aitkulov, D. Akilbekova,
M. Olivero, D. Tosi, IEEE Sens. J. 2017,17, 6935.
[162] Y. Man, A. Li, B. Li, J. Liu, L. Pan, Analyt. Chim. Acta 2019,
1092, 75.
[163] Y. Man, M. Ban, A. Li, X. Jin, Y. Du, L. Pan, Food Chem. 2021,354,
129578.
[164] T. Pomili, P. Donati, P. P. Pompa, Biosensors 2021,11, 443.
[165] K. Wang, J. Yang, H. Xu, B. Cao, Q. Qin, X. Liao, Y. Wo, Q. Jin, D. Cui,
Analyt. Bioanalyt. Chem. 2020,412, 2517.
[166] R. Jin, F. Wang, Q. Li, X. Yan, M. Liu, Y. Chen, W. Zhou, H. Gao,
P. Sun, G. Lu, Sens. Actuators B: Chem. 2021,327, 128922.
[167] B. Liu, Z. Wu, C. Liang, J. Lu, J. Li, L. Zhang, T. Li, W. Zhao, Y. Fu,
S. Hou, Front. Microbiol. 2021,12, 692831.
[168] X. He, Q. Pei, T. Xu, X. Zhang, Sens. Actuators B: Chem. 2020,304,
127415.
[169] Y. Lai, M. Li, X. Liao, L. Zou, Molecules 2021,26, 5016.
[170] Y. Cai, J.-C. Niu, Y.-Q. Liu, X.-L. Du, Z.-Y. Wu, Analyt. Chim. Acta
2020,1100, 149.
[171] R. Jin, L. Zhao, X. Yan, X. Han, M. Liu, Y. Chen, Q. Li, D. Su, F. Liu,
P. Sun, Biosens. Bioelectron. 2020,167, 112457.
[172] J. Xu, H. Khan, L. Yang, Analyt. Chem. 2021,93, 14755.
[173] F. Lu, Y. Yang, Y. Liu, F. Wang, X. Ji, Z. He, Analyst 2021,146, 949.
[174] H. Xu, A. Xia, D. Wang, Y. Zhang, S. Deng, W. Lu, J. Luo, Q. Zhong,
F. Zhang, L. Zhou, Science advances 2020,6, eaaz7445.
[175] S. H. Baek, C. Park, J. Jeon, S. Park, Biosensors 2020,10, 187.
[176] L. Liu, M.-R. Xie, Y.-Z. Chen, Z.-Y. Wu, Lab Chip 2019,19, 845.
[177] K. Fan, J. Zeng, C. Yang, G. Wang, K. Lian, X. Zhou, Y. Deng, G. Liu,
ACS Sens. 2022,7, 2049.
[178] L. Ma, L. Yin, X. Li, S. Chen, L. Peng, G. Liu, S. Ye, W. Zhang, S. Man,
Biosens. Bioelectron. 2022,195, 113646.
[179] Y. Ning, F. Lu, Y. Liu, S. Yang, F. Wang, X. Ji, Z. He, Analyst 2021,
146, 4775.
[180] T. H. Fereja, S. A. Kitte, W. Gao, F. Yuan, D. Snizhko, L. Qi,
A. Nsabimana, Z. Liu, G. Xu, Talanta 2019,204, 379.
[181] M. Bhaiyya, P. K. Pattnaik, S. Goel, Sens. Actuators A: Phys. 2021,331,
112831.
[182] D. Liu, C. Ju, C. Han, R. Shi, X. Chen, D. Duan, J. Yan, X. Yan, Biosens.
Bioelectron. 2021,173, 112817.
[183] L. A. Ireta-Mu˜noz, E. Morales-Narváez, Biosensors 2020,10, 60.
[184] T. Bu, F. Bai, S. Zhao, Y. Cao, K. He, X. Sun, Q. Wang, P. Jia, M. Li,
X. Wang, Biosens. Bioelectron. 2021,192, 113538.
[185] W. Ji, Z. Zhang, Y. Tian, Z. Yang, Z. Cao, L. Zhang, Y. Qi, J. Chang,
S. Zhang, H. Wang, ACS Appl. Mater. Interfaces 2019,11, 8584.
[186] F. Zeng, W. Duan, B. Zhu, T. Mu, L. Zhu, J. Guo, X. Ma, Analyt.
Chem. 2019,91, 1064.
[187] M. Sun, B. Li, X. Liu, J. Chen, T. Mu, L. Zhu, J. Guo, X. Ma, J. Mater.
Sci. Technol. 2019,35, 2207.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (27 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH

[188] Y. Li, X. Liu, J. Guo, Y. Zhang, J. Guo, X. Wu, B. Wang, X. Ma,
Nanomaterials 2021,11, 1496.
[189] S. Xie, D. Tao, Y. Fu, B. Xu, Y. Tang, L. Steinaa, J. D. Hemmink,
W. Pan, X. Huang, X. Nie, C. Zhao, J. Ruan, Y. Zhang,
J. Han, L. Fu, Y. Ma, X. Li, X. Liu, S. Zhao, ACS Synth. Biol.
2022,11, 383.
[190] Ö. B. Mercan, V. Kılıç, M. Şen, Sens. Actuators B: Chem. 2021,329,
129037.
[191] H. J. Min, H. A. Mina, A. J. Deering, E. Bae, J. Microbiol. Methods
2021,188, 106288.
[192] Q. Ning, W. Zheng, H. Xu, A. Zhu, T. Li, Y. Cheng, S. Feng, L. Wang,
D. Cui, K. Wang, Analyt. Bioanalyt. Chem. 2022,414, 3959.
[193] R. P. Edachana, A. Kumaresan, V. Balasubramanian, R. Thiagarajan,
B. G. Nair, S. B. T. Gopalakrishnan, Microchim. Acta 2020,414,1.
[194] V. Kılıç, Ö. B. Mercan, M. Tetik, Ö. Kap, N. Horzum, Analyt. Sci.
2022,38, 347.
Kexin Fan is currently an undergraduate student in Biomedical Engineering Bachelor of Engineering
Programme at School of Medicine, The Chinese University of Hong Kong, Shenzhen, China. In 2021, she
started her research in biosensors and intelligent diagnosis under the supervision of Professor Guozhen
Liu. Her research interests include advanced biosensors, intelligent nanoparticles, and smartphone-based
optical detection systems.
Guozhen Liu is a professor of biomedical engineering working on integrated devices and intelligent
diagnostics, at the Chinese University of Hong Kong, Shenzhen. She was an associate professor and
ARC future fellow at UNSW. Her career is alternating between academia and industry. After finishing her
Ph.D. at UNSW, she conducted her postdoctoral research at CSIRO and UNSW, respectively, before she
was appointed as an associate professor at Central China Normal University and R&D Manager, China,
in AgaMatrix Inc. She is well recognized for her interdisciplinary and translational research with close
end-user engagement in the areas of biosensors, point-of-care diagnostics, and medical devices.
www.advancedsciencenews.com www.advintellsyst.com
Adv. Intell. Syst. 2023,5, 2200285 2200285 (28 of 28) © 2022 The Authors. Advanced Intelligent Systems published by Wiley-VCH GmbH