PreprintPDF Available
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Diabetes is a chronicle disease where the body of a human is irregular to dissolve the blood glucose properly. The diabetes is due to lack of insulin in human body. The continuous monitoring of blood glucose is main important aspect for health care. Most of the successful glucose monitoring devices is based on methodology of pricking of blood. However, such kind of approach may not be advisable for frequent measurement. The paper presents the extensive review of glucose measurement techniques. The paper covers various non-invasive glucose methods and its control with smart healthcare technology. To fulfill the imperatives for non-invasive blood glucose monitoring system, there is a need to configure an accurate measurement device. Noninvasive glucose-level monitoring device for clinical test overcomes the problem of frequent pricking for blood samples. There is requirement to develop the Internet-Medical-Things (IoMT) integrated Healthcare Cyber-Physical System (H-CPS) based Smart Healthcare framework for glucose measurement with purpose of continuous health monitoring. The paper also covers selective consumer products along with selected state of art glucose measurement approaches. The paper has also listed several challenges and open problems for glucose measurement.
Content may be subject to copyright.
Everything You Wanted to Know About
Noninvasive Bloodless Glucose Monitoring
Prateek Jain Amit M. Joshi Saraju P. Mohanty
Dept. of ECE Dept. of ECE Computer Science and Engineering
MNIT, Jaipur, India MNIT, Jaipur, India University of North Texas, USA
Diabetes is a chronicle disease where the body of a human is irregular to dissolve the blood glucose
properly. The diabetes is due to lack of insulin in human body. The continuous monitoring of blood
glucose is main important aspect for health care. Most of the successful glucose monitoring devices is
based on methodology of pricking of blood from ear lobe or finger. There is requirement to develop
the Internet-Medical-Things (IoMT) integrated Healthcare Cyber-Physical System (H-CPS) based Smart
Healthcare framework for glucose measurement with purpose of continuous health monitoring. To fulfill
the imperatives for non-invasive blood glucose monitoring system, there is a need to configure an accurate
measurement device. Noninvasive glucose-level monitoring device for clinical test overcomes the problem
of frequent pricking for blood samples.
Index Terms
Smart Healthcare, Internet-of-Medical-Things (IoMT), Healthcare Cyber-Physical System (H-CPS),
Diabetes, Glucose measurement, Non invasive measurement, Spectroscopy and calibration
The glucose is considered as important source of energy for the human body. The body requires blood
glucose of normal range (80 to 150 mg/dl) in order to perform the daily activities [1]. However, the higher
or lower value of glucose would lead to various complication inside the body. At the same time, insulin
is also crucial hormone generated inside the body from the food intake. The glucose is produced from
the food digestion which enters the blood cell to supply the energy and also helps in the growth. In case,
the insulin is not properly generated then blood would accumulate the high glucose concentration. Fig. 1
illustrates the closed-loop of glucose generation and consumption in human body [2]. A consistently high
blood glucose concentration is possible if the generation of αcells is larger as compared to that of the
βcells. Because of this condition, enough insulin is not secreted in the body for glucose consumption.
This condition refers to as the Diabetes Mellitus. Diabetes is termed as chronic disease which defines
high blood glucose levels inside the human body. The unbalanced glycemic profile is main reason for
the cause of diabetic condition. The rate of prevalence for Non Communicable Diseases (NCD)/Chronic
Disease has increased with many fold from last several years. There are around 20 million death reported
yearly through cardiovascular disease, for which high blood glucose is significant predisposing factors.
Moreover, people with diabetes are more affected during the viral pandemic outbreaks [3], [4], [5].
There has been exponential growth of diabetes patients over past few years because of obesity, unhealthy
diet plan, old-age population, and inactive lifestyle. Diabetes is considered as one of the fastest growing
health challenges, with the number of adults living with diabetes having more than tripled over the past
2 decades (Refer Fig. 2) [6]. The prevalence of diabetes around the world was 9.3% during 2019 with
approximate 463 million people. It is expected to rise to 578 million by 2030 with 10.2% prevalence
arXiv:2101.08996v1 [] 22 Jan 2021
Insulin Secretion
by βCells
by Insulin
Food Intake Exercise
by αCells
Glucose Deliverance
by Liver
CGM by
High Glucose Concentration
Food Intake
CGM by
iGLU Normal Blood
Glucose Concentration
Fig. 1: Illustration of the closed loop form of glucose generation and consumption [2].
rate and the same would be 10.9% with 700 million population by 2045. It has been observed that
prevalence is quite higher in urban to 10.8% whereas 7.2% in rural region. Almost half of the diabetes
patients unaware about their situation due to lack of knowledge. The diabetes has indeed global outbreak
which has affected presently almost 1 in 10 people around the world. It is projected that more than 0.5
billion adults would suffer from the diabetes in the next decade [7]. As per the report from International
Diabetes Federation (IDF), the death from diabetes has large number than combined death from Malaria
(0.6mio), HIV/AIDS (1.5mio) and tuberculosis (1.5mio) [8]. There are around 8 million new patients are
being added to diabetic community every year. This has grown the demands immensely for the effective
diabetic management. It is important to monitor the blood glucose over time to time for avoiding late-
stage complication from diabetes. This has necessitate the design of various reliable and robust solutions
for efficient diabetes management. The market of diabetes devices has also grown rapidly with significant
requisite for frequent glucose measurement for better glycemic profile control.
Diabetes is one of the major chronic disease which has long-term impact of the well-being life of
a person. Diabetes Mellitus (DM) is considered as physiological dysfunctions with high blood glucose
because of insufficient insulin, insulin resistance, or excess generation of glucagon [9]. It is the critical
health issue of 21st century. Type 2 Diabetes (T2DM) has shown rapid growth around the world from
past few years. Any form of diabetes may lead to complications in various body parts which increase the
possibility of premature death. The higher value of blood glucose known as hyperglycemia, would lead
to thickening of blood vessels which could resulted in kidneys damage and loss of sight and some times
even to these organs failure. Diabetes is also associated with limb amputation, peripheral vascular diseases
and myocardial. Contrary, the low blood glucose defined as hypoglycemia may occur in Type 1 Diabetes
Patients (T1DM) for excessive insulin dosage [10]. The most common symptoms for hypoglycemia
pateints are dizziness, sweating and fatigue and in the worst case it can lead to coma and death. The
diabetic patients would have several common symptoms such as thirsty, tiredness, changes in vision,
consistently hungriness, unexpected weight loss and the excretion of urine within short durations [11].
If the diabetes remain untreated over the period of time, it may cause blindness, heart stroke, kidney
disease, lower limb amputation and blindness. It would lead to increase the probability of death almost
50% higher in comparison of the patients without diabetes. The diabetes also brings the additional financial
burden for the treatment and point of care. The diabetic patients could also result in loss of productivity
at workplace and may lead to disability. There are several health issues which may also arise from
Chronological Year
Population in Millions
Fig. 2: Global trend of Diabetes, Adopted from [6].
diabetes like depression, digestive problem, anxiety disorders, mood disorder and eating habits change.
The diabetes could be controlled with proper diet plan, through some physical exercise, insulin dosage
and medicines. The early stage of diabetes is possible to control with oral medicines. The diabetes control
also helps to reduce the associated risk of high blood pressure, cardiovascular and amputation.
The rest of the article is organized in the following manner: Section II briefly presents different
types of diabetes while making case for the need of glucose level monitoring. Section III presents
overview of various types of glucose-level measurement mechanisms. Section IV provides details of
available approaches for noninvasive glucose-level monitoring. Section V has discussions on various
post-processing and calibration techniques for noninvasive glucose-level monitoring. Section VI briefly
discusses various consumer products for noninvasive glucose level measurement. Section VII presents
the approaches for glucose-level control and corresponding consumer products. Section VIII provides the
Internet-of-Medical-Things (IoMT) perspectives of glucose level measurements and control in healthcare
Cyber-Physical Systems (H-CPS) that makes smart healthcare possible. Section IX outlines the shortcom-
ings and open problems of glucose-level measurements and control. Section X summarizes the learning
of this comprehensive review work.
This Section presents details of different types of diabetes, the health issues arise due to diabetes,
while making case for the need of glucose level monitoring.
A. Types of Diabetes
The diabetes occurs because of insufficient insulin with respect to glucose generated inside the body.
The insulin from body is either insufficient or not any which is generated from beta cells of the pancreas.
In case of diabetes, the cells of liver, muscles and fat unable to balance glucose insulin effectively. The
diabetes are classified mainly in three categories: Type 1 diabetes, Type 2 diabetes and gestational diabetes
(Refer Fig. 3) [12].
Body doesn’t
generate sufficient
insulin for glucose
Type 1 Diabetes Type 2 Diabetes Gestational Diabetes
Insufficient insulin
secretion for blood
glucose regulation
Body is not able to
produce insulin for
glucose regulation
during pregnancy
Always hungry
Unexpected weight loss
Numb or tingling hands/feet
Frequent urination
Sexual disorder
Extreme fatigue
Always thirsty
Wounds heal slowly
Peripheral neuropath
Cerebrovascular disease
Diabetic Nephropathy
Coronary heart disease
Eye damage
Fig. 3: Different types of diabetes and their symptoms.
For diabetes of type-1, the pancreas does not produce insulin inside the body which is resulted in a
weak immune system. This results in a person who is unable to generate insulin naturally [2], [13]. In
case of type 2 diabetes, the amount of insulin from pancreas is not sufficient to maintain glycemic profile
of the body. Gestational diabetes usually occurs in a pregnant woman at later stage of the delivery. in the
year 2020, total 2 billion adults around the globe suffers from overweight, and 300 million of them are
obese. In addition, a minimum of 155 million children in the world is overweight or obese. It is projected
that the prevalence of hyperglycemia is 8.0% and expected to increase to 10% by 2025 [7]. There has
been concern for diabetic people specially in developing countries due to increase in Type 2 Diabetes
cases rapidly at earlier age which have overweight children even before puberty. Whereas for developed
countries, most of people have high blood glucose at age around 60 years. Most frequently affected
are at middle aged between 35 and 64 in developed countries [6]. In 2019, 69.2 millions population in
India had Type-2 diabetes. Approximately 2.35 million adults have Type-1 diabetes. In general, there
are around 5% adults have been considered for Type-1 diabetic patients while the others 90-95% are of
Type-2 diabetic patients. Type-1 diabetic patient must have insulin to control the blood glucose level.
Type-2 diabetic patients can control their glucose level by following an optimized diet with medication
and a regular physical exercise schedule.
B. The Health Crisis due to Diabetes
The diabetes mainly occurs due to unbalanced glucose insulin level of the body where insulin is
demolished and muscles and cells are not able to generate insulin properly [14], [13]. The probability of
death would also increase upto 50% in comparison to non-diabetes case. The control action of the diabetes
would be possible using proper precautionary measure after frequent glucose measurements. Therefore,
there is a real need for smart healthcare solution which would provide instant self measurement of blood
glucose with high accuracy.
Kidney damage (nephropathy)
Cardiovascular disease
Alzheimer's disease
Eye damage (retinopathy)
Skin conditions
Nerve damage (neuropathy)
Foot damage
Hearing impairment
Sexual, &
Bladder Problems
Gum Disease &
Dental Problems
Fig. 4: Diseases in human body due to diabetes
Hyperglycemia is the major issue which has been considered by several health organizations at world-
wide level [15], [16]. There are several attempts which have been used for glucose measurement [17].
There have been substitutional work using various techniques to make the device more familiar with
clinicians and patients [18]. Diabetes is possible in the age group 18 to 80 years usually [19], [20]. The
normal range of glucose is in the range of 70-150 mg/dL and pathophysiological would be from 40
mg/dL to 550 mg/dL [21]. One of the emerging issues is to design the glucose measurement device for
continuous health care monitoring [22]. The devices for monitoring the glucose level are available for
last two decades [23].
C. Glucose Measurement: A Brief History
The glucose meter (aka glucometer) is a portable medical device for predicting the glucose level
concentration in the blood [24], [25]. It may also be a strip based dipped into any substance and
determined the glucose profile. It is a prime device for blood glucose measurement by people with
diabetes mellitus or hypoglycemia. With the objective of glucose monitoring device advancement, the
concept of the biosensor has been proposed earlier in 1962 by Lyons and Clark from Cincinnati. Clark is
known as the “father of biosensors”, and modern-day glucose sensor which is used daily by millions of
diabetics. This glucose biosensor had been composed with an inner oxygen semipermeable membrane, a
thin layer of GOx, an outer dialysis membrane and an oxygen electrode. Enzymes could be gravitated at
an electrochemical detector to form an enzyme electrode [26]. However, the main disadvantage of first-
generation glucose biosensors was that there was the requirement of high operation potential of hydrogen
peroxide amperometric measurement for high selectivity. The first-generation glucose biosensors were
replaced by mediated glucose biosensors (second-generation glucose sensors). The proposed biosensors
till present scenario represent the advancements in terms of portability of device and precision in
measurement. But, due to some environmental and measurement limitations; these biosensors were not
taken for real-time diagnosis. The history of glucose measurement is shown in Fig. 5 [27].
1962 First glucose biosensor based on enzyme electrode
developed by Clarke and Lyons
1965 First ever glucose measurement strip developed by
1987 First ever self glucose
measurement strip
by Medisense
1971Ames Reflectance glucometer by
Anton Hubert
1991 First ever Continuous
Glucose Measuring Device
2000 First Ever
Non invasive
Fig. 5: History of Glucose Measurement.
D. Glucose Measurement Technique
Presently, the glucose monitoring is carried out either laboratory based technique or home based
monitoring. These both approaches are invasive in nature which provides discomfort by blood pricking
and it only helps to measure the glucose measurement at that point of time. It is also not very convenient
for the user to take out blood samples multiple times in a day and many patients are reluctant to opt
such type of solution. Therefore, significant changes of glycemic profile may go unnoticed because of
unanticipated side effects and low compliance from the patients. This could impact on improper insulin
dosage and unknown food ingredient. However, they are reliable solution due to their good sensitivity
and higher accuracy for glucose measurement [28], [29].
The novel approach for glucose measurement has been explored from past several years which is based
on the principle of physical detection than conventional chemical based principle. Such non-invasive based
method does not require the blood sample but uses the interstitial fluid (ISF) for glucose molecule detec-
tion. There are several attempts in the same direction for glucose measurement through sweat, saliva, tears
and skin surface [30]. However, the main challenge is to have precise measurement, good sensitivity and
reliability from such measurement. Such approach could be suitable for Continuous Glucose Measurement
(CGM) and self monitoring purpose. Such CGM techniques would provide the frequent measurement in
a day which would helpful for better glucose control and also for the necessary preventive actions for
Skin between
Light Detection
through Optical
Sensors and Signal
Data Acquisition
and Prediction of
Blood Glucose
using Regression
Preparation of
Lancet Pricking of the
Blood Contact to strip
for Monitoring
Invasive Approach (Capillary Glucose Measurement)
Non Invasive Approach
Blood Glucose
Blood Sample
Glucose Value
(mg/dl) using
Clinical Centrifuge for
Separation of Serum Prepared Serum for Glucose
Invasive Approach (Serum Glucose Measurement)
Data Acquisition and Prediction of
Blood Glucose
Sample Preparation in Lab/ Requirement of
Disposable Strips and Lancets
Pricking of the
Fig. 6: Invasive versus Noninavsive Glucose Measurement.
hyperglycemia and hypoglycemia patients. Such kind of techniques would also support for the dietician
and healthcare provider to prepare proper diet plan according to glucose fluctuation for the patient.
E. The Need for Continuous Glucose Measurement (CGM)
The measurement of glucose could be done through non-invasive, semi (or minimal) invasive and
invasive approach. The frequent measurement may not be possible using invasive method which can
cause trauma. The semi-invasive and non-invasive could be useful for Continuous Glucose Measurement
(CGM) without any pricking of the blood. However, the non-invasive glucose measurement is most
suitable technique which helps to measure the blood glucose painlessly [31].
CGM assist to have proper blood glucose level analysis at each prandial mode. It helps to measure
glucose insulin level after insulin secretion,hysical exercise or subsequent to medication. The frequent
glucose reading also helpful to endocrinologist for providing the proper prescription. It mainly helps for
type 1 diabetic patients to take care of their insulin dosage over the period of time. The proper diet
management could be possible with help of recurrent glucose monitoring and flow diagram of CGM is
shown as Fig. 7 [28], [29]. The CGM is useful for the patients for frequent glucose measurement over
the period of time. This would helpful to identify the average blood glucose value for the last 90 days,
by which glycated haemoglobin (HbA1c) can be determined.
through iGLU To avoid excessive
food intake
To avoid excessive
Insulin secretion
Analysis and prescribed
treatment for
diabetic patient
Red Blood Cells
To identify the average
blood Glucose value
Fig. 7: The objectives of continuous glucose monitoring.
This Section discusses an overview of various types of glucose-level measurement mechanisms. In
the past, many works has done for the glucose measurement. They can be invasive, non-invasive, or
minimally invasive. A lot of works has been completed based on the non-invasive technique. They are
technically based on optical and non-optical methods. Some of the optical techniques used methods based
on Raman Spectroscopy, NIR spectroscopy, and PPG method. A taxonomy of the different methods is
provided in Fig. 8 [25], [28], [29], [32].
A. Invasive Methods
Many commercial continuous blood glucose measurement devices use cost-effective electrochemical
sensors [33]. They are available to respond quickly for glucose detection in blood [34]. Lancets (for
pricking the blood) is used at the primary stage for blood glucose monitoring for various commercial
devices available in the market [35]. The frequent measurement through the process is so much panic due
to picking the blood sample from the fingertip more than 3-4 times in a day for frequent monitoring[36].
The low invasive biosensor for glucose monitoring has been developed with glucose oxidase that re-
quire around 1mm penetration inside the skin for measurement [37]. The technique of photometric was
attempted to detect glucose with help of small blood volumes [38].
B. Minimally Invasive Methods
The minimally invasive method using prototype sensor was developed to have frequent monitoring of
glucose tissue [39]. The sensor is wearable and is implanted on membrane which contains the immobilized
glucose oxidase. The glucose monitoring through implantable devices were developed [40]. The semi or
minimal invasive method using biosensors designed for diabetes patient [41]. The wearable micro system
explored for frequent measurement of glucose [42]. Similarly, there was an attempt of continuous glucose
Blood Glucose
Monitoring System
Invasive Minimally Invasive Non Invasive
In-vivo In-vitro
Spectroscopy Polarimetry
(PPG) Signal
Long Wave
Micro System
Glucose Control
System Electrochemical
iGLU 1.0
Glucose Serum
iGLU 2.0
Fig. 8: An overview of the Glucose Measurement Options [32], [28], [29], [25].
monitoring with help of microfabricated biosensor through transponder chip [43]. The signal coming out
of transponder chip was used for the calibration for semi invasive approach of Dexcom sensor [44]. The
diabetes control explored by glucose sensor with artificial pancreas system [45]. The minimal invasive
approaches have limitations mainly accuracy and may have shorter life span for monitoring.
This is a wearable microsystem for the continuous monitoring of the blood glucose. It’s a minimally
invasive method for the glucose monitoring. The main idea behind this is that it uses micro-actuator
which consists the shape memory alloy (SMA) for the extraction of the blood sample from human skin
[46]. An upgraded version of SMA is used for the implementation of PCB. Because of it’s feasibility
and performance, it can be considered as the first wearable device for the glucose monitoring but it is
large in size which makes it inconvenient.
C. Non-invasive Methods
Non-invasive measurement would mitigate all the previous issues and would provide painless and
accurate solutions [47], [48]. The non-invasive glucose measurement solution for smart healthcare had
developed through portable measurement [31]. A lot of approaches have been introduced for glucose
measurement [49]. The non-invasive measurement are more convenient for continuous glucose mea-
surement in comparison to invasive method and semi invasive [47], [48]. The glucose measurement
with help of optical method has observed more reliable and precise in the literature [50]. The popular
optical methods include non-invasive measurement such as Raman spectroscopy, near infer-red spec-
troscopy,polarimetric,scattering spectroscopy [51], photoacoustic spectroscopy [52] etc. For the develop-
ment of a non-invasive measurement device, it is considered by the researcher that the device would
be much convenient for the user’s perspective [53], [54]. Calibration of the blood glucose to interstitial
glucose dynamics have been considered for the accuracy of continuous glucose monitoring system [55],
[56]. Several calibration algorithms have been developed and implemented for portable setup [57]. There
has been several concious efforts towards the development of the self-monitoring system [58].
Blood in Vessel
NIR LED NIR Detector Voltage Acquisition
through Light Detection
from Blood Vessels
Optical Spectroscopy for
Serum Glucose Measurement
Pricking of the
Blood Contact to strip
for Monitoring
Capillary Glucose
Blood Pricking from
Epidermis Layer of Skin
Fig. 9: NIR Spectroscopy Mechanism of Serum Glucose Measurement.
D. Invasive Versus Non-invasive Glucose Measurements: The Trade-Offs
Recent glucose measurement methods for the ever-increasing the diabetic patients over the world
are invasive, time-consuming, painful and a bunch of the disposable items which constantly burden for
the household budget. The non-invasive glucose measurement technique overcomes such limitations, for
which this has become significantly researched era. Although, there is tradeoff between these two methods
which is represented in Fig. 10.
E. Capillary Glucose versus Serum Glucose for Noninvasive Measurement
The serum glucose value is precise which is always close to actual blood glucose measurement with
compare to capillary glucose level. Traditional approaches able to measure capillary glucose instantly
Non-invasive Glucose
Invasive Glucose
Measurement Detector
Fig. 10: Representation of Tradeoffs between Invasive and Non-invasive Glucose Measurement.
but the serum glucose measurement identification is difficult. It is observed that the glucose level of
capillary is always higher than serum glucose. The accurate measurement of blood glucose would help
for appropriate control actions. Therefore, it is really important to measure the serum glucose than the
capillary glucose which is more reliable for medication. Capillary blood glucose measurement has been
used widely than serum glucose estimation for medication purpose. The serum glucose is not possible for
continuous glucose measurement or frequent measurement for diabetes. The blood glucose is controlled in
much better way if one can measure serum glucose at regular interval. Laboratory analysis of glycosylated
haemoglobin (HbA1c) which provides 6-8 weeks blood glucose measurement is also being done through
the serum blood only. For the non-invasive measurement point of view, serum and capillary glucose are
being measured through the optical spectroscopy. The mechanism of blood glucose measurement is based
on received IR light after absorptions and scattering from glucose molecules which flow in blood vessels.
The methodology is quite similar for both types of glucose measurement except the post-processing
computation models which are necessary for blood glucose estimation.
F. Non-invasive Method for Glucose Level Estimation by Saliva
As the most convenient method to estimate glucose level is via saliva [59] and is used for children and
adults. This saliva has specific type of parts which can be defined as: (1) gland-specific saliva and (2)
whole saliva. The collection of the Gland-specific saliva is done by individual glands like parotid, Sub
mandibular, sublingual, and minor salivary glands. This diagnosis is done by the history of the patient in
terms of associated risk factors, family history, age, sex, duration of diabetes,and any associated illness.
Other Glucose measuring methods consist of measurement using photo-metric glucometers requiring very
small sample volumes [60]. Figure 2 represents the flow of the model. The basic approach is based on
the reaction of the chemical test strip that reacts with the blood and changes colour. Measurement is done
by capturing the reflections of the test area and then glucose level is estimated. It requires validation in
large number of patients.
This Section presents detailed discussions of various available approaches for noninvasive glucose-level
monitoring. There have been several efforts for noninvasive glucose measurement using optical techniques
[27], [29], [61], [62]. These techniques are mainly based on various spectroscopy based methods. For
the development of a non-invasive measurement device, it is considered by the researcher that the device
would be much convenient for the users perspective. Fig. 11 presents summary of various types of
noninvasive glucose measurement techniques, whereas their comparative perspectives are presented in
Fig. 12. A qualitative comparative perspective of various noninvasive methods is summarized in Table I.
Photo acoustic
spectroscopy Near Infrared
Radio frequency
(RF) technique
Fig. 11: Various spectroscopy techniques for noninvasive glucose measurement.
TABLE I: Qualitative comparison of various noninvasive glucose-level monitoring methods.
Technique Advantages Disadvantages
Near Infra-Red
(NIR) The signal intensity is directly pro-
portional to glucose molecule
The glucose detection concept would
work with other interfacing sub-
stance such as plastic or glass
The glucose signal weak compara-
tively so complex machine learning
model is required for interpretation
High scattering level
Glucose readings
vary in different
Change in position
of inductor will
provide undesired
results of glucose
Raman scattering
is a nonlinear or
scattering which
occurs when
light interacts
with a certain
Instability of
laser wavelength,
intensity and
long spectral
acquisition times
are the main
limitations of this
Because of other
present in the
interior chamber
of the eye.
The amount of
rotation is a linear
function of the
path length, the
concentration and
the specific
rotation constant
(property of the
Slight difference
in placing the
sensor at the
same location
might affect the
output of the
Effect of
pressure on the
sensor, body
temperature and
sweat on the
Fig. 12: Comparative Perspective of Various spectroscopy techniques for noninvasive glucose measure-
Technique Advantages Disadvantages
Mid Infra-Red
(MIR) The glucose molecule absorption
Low scattering
The light has limited penetration
with tissue
Noise is present in the signal so
water and other non-glucose metabo-
lites would be detected.
Far Infra-Red
Frequent Calibration is not required
Least sensitive towards scattering
The radiation intensity depends on
temprature and substance thickness
Strong absorption with water so it
is difficult to have precise glucose
Spectroscopy Less sensitivity towards temperature
and water
High specificity
Requirement of the laser radiation
source hence it can dangerous cell
for CGM
Susceptible towards noise interfer-
ence so low SNR
Technique Advantages Disadvantages
Photo acoustic
Simple and compact sensor design
Optical radiation will not harmful for
the tissue
Signal is vulnerable towards acoustic
noise, temperature,motion etc.
It carries some noise from some non-
glucose blood components
The laser intensity variation will not
change much the glucose prediction
Requirement external laser source
and requires proper alignment with
sensitive for the change in PH and
Iontophoresis Based on simple enzyme based elec-
trode system
Highly accurate as it measure glu-
cose from interstitial fluid
Difficult to have proper calibration
Not so user-friendly approach due to
passing of the current through skin
Highly sensitive for glucose
molecule detection due to immune
for light scattering
Good sensitivity because of distinc-
tive optical properties
Very much sensitive for local pH
and/or oxygen,
Suffers from foreign body reaction
Bio impedance
spectroscopy Comparatively less extensive
Easy for CGM
Prone towards sweating, motion and
Require large calibration period
Millimetre and Mi-
crowave sensing Deep penetration depth for precise
glucose measurement
No risk for ionization
Poor selectivity
Very much sensitive for physiolog-
ical parameters such as sweating,
breathing and cardiac activity
Optical Coherence
Tomography High resolution and good SNR
Not vulnerable for blood pressure
and cardiac activity
Glucose value may change as per
skin and motion
Suffers from tissue inhomogeneity
Surface Plasma
Resonance Small glucose molecule can be de-
tected due to high sensitivity
Long calibration process and size is
Glucose value changes with varia-
tion in temperature,sweat and mo-
Technique Advantages Disadvantages
Time of flight and
THz Time domain
strong absorption and dispersion for
glucose molecule
Lesser depth resolution and longer
time for measurement
Metabolic Heat
Conformation Uses the concept of well-known
various physiological parameters for
glucose prediction
Sensitive towards variation in tem-
perature and sweat
sensing low-cost and can be easily miniatur-
No risk of ionization
Lack of selectivity due to dielec-
tric constant is mainly affected with
other blood components
More sensitive for the slight change
of temperature
Technology Well established technology with not
much harm to tissue cell
Long penetration below the skin or
Limited accuracy with ultrasound
only hence mostly used with NIR as
costly technology for measurement
and not useful for CGM
Favourable technology as there is no
side-effect to skin
Based on well known enzymatic
Error prone due to environmental
A. Near-Infrared (NIR) Spectroscopy
It is well known as Infrared spectroscopy (IR spectroscopy) or vibration spectroscopy where radiation
of infrared type are incident on the matter [63], [64]. Various types of IR spectroscopy is shown in
Fig. 13. In general, IR spectroscopy includes reflection, scattering and absorption spectroscopy [65]. The
wave from IR absorption cause the molecular vibration and generate the spectrum band with wavelength
number in cm1[66].
In this case, the light in the wavelength range of 700nm to 2500nm for Near-infrared region is applied
at the object (may be finger or ear lobe) [68]. The light may interact with blood components and it
may scattered, absorbed and reflected [69], [70]. The intensity of received light varies as per glucose
concentration as per Beer-Lambert law [71], [72] The receiver would help to measure the presence glucose
molecule from the blood vessel [73].
1) Long-Wave versus Short-Wave NIR Spectroscopy: The optical detection is useful approach to have
precise glucose measurement. FIR (Far infra-red) based optical technique help to get the resonance
between OH and CH for first overtone. However, long wave NIR has good performance in vitro testing.
In similar way, the fiber-optic sensor is used along with laser based mid-infrared spectroscopy for vitro
based glucose measurement. The continuous glucose measurement has been achieved with multivariate
calibration model for error analysis [74]. The FIR approach has limitation of shallow penetration in
comparison with short wave NIR. The short NIR would help to detect the glucose molecule more
Spectroscopy Reflectance
Spectroscopy Scattering
Fig. 13: Classification of vibrational spectroscopy [67].
IR LED Object OPD Filter
1550E Finger/Ear
Light to voltage
Converter RC Low
Pass Filter
NIR Spectroscopy
Fig. 14: Block diagram representation of IR spectroscopy.
accurately [75]. The concept of NIR spectroscopy for glucose detection is shown in Fig. 15. The specific
wavelength of NIR spectroscopy has already been applied earlier for precise glucose measurement using
non-invasive measurement [76]. Some specific wavelength such as 940 nm has been considered for
the detection of glucose [77]. The vibration of CH molecule has been observed at 920 nm with NIR
spectroscopy [76]. In some other works, the glucose absorption has been validated for the range 1300 to
1350 nm and stretching of glucose has been identified in NIR region [78], [79]. The presence of glucose
component has been measured at 1300 nm in the work [80].
Epidermis Layer
Dermis layer
Subcutaneous Tissues
Near Infrared
Short wave
Long wave
Mid and Far Infrared
Fig. 15: Penetration depth various Infrared Signals in Human Skin [25], [32].
2) NIR Spectroscopy Based Methods: A method to estimate the non-invasive blood glucose with NIR
spectroscopy using PPG has been proposed in literature [81]. This method is performed using NIR LED
and photo detector with an optode pair. At NIR wavelengths(935nm, 950nm, 1070nm), PPG signal is
obtained by implementation of analog front end system. The glucose levels has been estimated using
Artificial Neural Network (ANN) running in FPGA. A microcontroller is used, for the painless and
autonomous blood extraction [82]. The ideal system Blood Glucose Measurement (BGM) in which the
microcontroller is used to display the blood glucose and for the transmission of blood glucose. A remote
device is used for the tracking of the insulin pump which is needed for diabetes management. This type
of measurement [83] method uses change in the pressure of the sensitive body part, because it generates
the sound waves. The response of the photo acoustic signals will be stronger when glucose concentration
is higher. In order to improve SNR and for the reduction of noise to transfer the signal to the computer
for further processing,the signal is then amplified. Feature extraction and glucose estimation is estimated
by photo acoustic amplitude. In order to gather the photo acoustic signals, two pulsed laser diodes and
piezoelectric transducer is used. Utilization of the LASER makes the setup costly and bulky.
3) Non-invasive Blood Glucose Measurement Device (iGLU): In this approach, “Intelligent Glucose
Meter (iGLU)” [84] has been utilized for the acquisition of data. This device works on a combination
of NIR spectroscopy and machine learning. This device has been implemented using three channels. It
uses an Internet-of-Medical-Things platform for storage and remote monitoring of data. In the proposed
device, an NIR Spectroscopy is used with multiple short wavelengths [85]. It uses three channels for data
collection. Each channel has its own emitter and detector for optical detection. Then the data collection
processed by a 16 bit ADC with the sampling rate of 128 samples/second. Regression techniques is used
to calibrate and validate the data and analyse the optimized model. The data that is stored on cloud can
be used and monitored by the patients and the doctors. Treatment can be given based on the stored data
values. This is a low cost device with more than 90 % accuracy but it does not give real time results.
4) Why NIR is Preferred Over other Noninvasive Approaches?: Glucose measurement has been done
using various non-invasive approaches such as impedance spectroscopy, NIR light spectroscopy, PPG
signal analysis and so on. But, apart from optical detection, other techniques have not be able to provide
the precise measurement. PPG is one of the promising alternative but the PPG signal varies according to
blood concentration [86], [87]. It may not be useful to have precise prediction of the blood glucose. The
saliva and sweat properties vary from one person to another person. Therefore, it could not be reliable
glucose measurement method. The other spectroscopy have been also applied for glucose measurement.
However, they are not able to provide portable, cost effective and accurate prediction of body glucose.The
glucose measurement using optical detection using long NIR wave which is not capable to detect the
glucose molecules beneath the skin as it has shallow penetration [75]. Therefore, small NIR wave has
been considered as potential solution for real-time glucose detection [77], [88].
B. Mid Infrared (MIR) Spectroscopy
The bending and stretching of glucose molecules would be observed very well with Mid Infrared (MIR)
spectroscopy [89]. The depth of skin penetration is very less because it tends to have larger absorption
of water. This technique helps to have ISF glucose value in vivo measurement. There are some attempts
for precise glucose measurement through saliva and palm samples.
C. Blood Glucose Level Measurement using PPG
The change of blood volume with absorption of the light from tissue has been detected with PPG
signal [87]. The change of the blood volume has been measured using pressure pulse with help of light
detector [86]. The change in volume of blood would result as the change of light intensity hence it may
not be occur due to glucose molecule. This may result as inaccurate glucose measurement. The difference
of NIR against PPG has been shown in Fig 16. The intelligent glucose measurement device iGLU is
mainly based on principle of NIR spectroscopy which helps to have precise glucose measurement. There
have been several work for glucose detection based on PPG signal [90]. The data from patient body has
logged to estimate the presence of glucose using PPG. Subsequently, various machine leaning models
have been used for prediction of body glucose value [91]. The different parameters from total 70 subjects
of healthy and diabetes have been considered for the prediction using Auto-Regressive Moving Average
(ARMA) models [92]. There have been also several other smart solutions for glucose estimation using
PPG signal with intelligent algorithms [93], [94], [95].
One of the optical based techniques is Photo-plethysmography (PPG) which is used in advanced health
care. It is non-invasive glucose measurement technique. In NIR spectrum a sensor similar to a pulse
oximeter is used to record the PPG signal [87]. Photo transmitter and receiver is used to build the sensor
which will operate in near infrared region at 920nm. At wavelength 920nm, by measuring changes in the
absorption of light, a PPG signal can be obtained. The veins in the finger grow and contract with every
A method of measuring blood glucose using pulse oximeter and transmission of the PPG glucose
monitoring system is available [90]. As the glucose concentration increases, there is decrease in the light
absorbance in the blood. The obtained signal is in the form of photo current, and for the filtering of this
signal is then changed into the measurable voltage values. For the processing of filtered signal, lab view
is used to estimate the blood glucose level.
A system using machine learning techniques and PPG system for the measurement of blood glucose
level non-invasively has been prototyped [86]. In this model, a PPG sensor, an activity detector, and a
signal processing module is used to extract the features of PPG waveform. It finds the shape of the PPG
waveform and the blood pressure glucose levels, the functional relationship between these two can be
obtained then.
In PPG, the change in light intensity will be varied according to changes in blood volume. PPG signal
analysis is not based on the principle of glucose molecule detection. Hence, the system has limited
accuracy [25], [32]. Fig. 16 illustrates the differences.
LED Detector Logged signal for
features extraction
Blood glucose
using machine
learning model
after extracting
features sets
from signal
Specific Wavelengths
are not required
PPG Signal Analysis Time and
analysis at
frame level
Single pulse
LED Detector Logged voltage values
after absorption and
reflectance of light from
glucose molecule
Blood Glucose estimation
using proposed and
optimized machine
learning model
Specific Wavelengths
needed for glucose
molecule detection
NIR Spectroscopy
Fig. 16: PPG Versus NIR for Non-invasive Glucose Measurement [25], [32].
D. Impedance Spectroscopy
Impedance spectroscopy (IMPS) refers to the dielectric spectroscopy [96]. The steps of impedance
spectroscopy (IMPS) is shown in Fig. 17. This technique finds the dielectric properties of skin [97].
The current is directed through the skin [98]. Due to directed small current at multiple wavelengths, the
impedance range is obtained [99]. The range lies between 100 Hz to 100 MHz [100], [101]. Change in
glucose concentration will reflect the change in sodium ions and potassium ions concentration [102]. So,
the cell membrane potential difference will be changed [103]. Thus, the dielectric value will be changed
which predicts the glucose value of human body [104].
An enzyme sensor in a flow cell has been explored for glucose measurement in saliva [105]. Polypyrrole
(PPy) supported with copper (Cu) nanoparticles on alkali anodized steel (AS) electrode for glucose
detection in human saliva is available in [106]. The high precision level cannot be possible through these
methods as sweat and saliva properties vary according to person. Hence, this approach is not suitable for
glucose measurement in smart healthcare.
E. Raman Spectroscopy
Due to the interaction of light with a glucose molecule, the polarization of the detected molecule
will change [107]. In this technique, oscillation and rotation of molecules of the solution are possible
through the incident of LASER light [108]. The vibration of the molecule affects the emission of scattered
light [109]. Due to this principle, blood glucose concentration can be predicted as [110]. This technique
provides more accuracy with compared to infra-red spectroscopy technique [111]. There has been several
research based on Raman spectroscopy to have precise glucose measurement. The validation has been
Voltage Control
Resonance Circuit
Analog to
and estimated
Digital to
and Storage
of Data
to skin
Fig. 17: The Steps of Impedance Spectroscopy (IMPS).
also carried out on using vivo testing. Fig. 18 presents basic framework of Raman spectroscopy, whereas
Fig. 19 presents its usage for noninvasive glucose measurement.
Light Beam
Expander Mirror
Beam Splitter Raman Spectrometer
Data Analyzer
Vacuum System
Fig. 18: Building blocks of Raman spectroscopy.
F. Time of Flight and THz Domain
The blood glucose estimation is adopted though Time of Flight (TOF) measurements for vitro testing
[112]. The short pulse of laser light is inserted in the sample for photon migration measurement. This
photon will experience scattering and absorption phenomenon while traveling from the sample. The
optical analysis of the photons would be useful for precise glucose measurement.
LASER Diode Band pass
Filter Spectrograph
Object (Finger
or Ear lobe)
Notch Filter
Fig. 19: Noninvasive glucose measurement using Raman spectroscopy.
G. Photo Acoustic Spectroscopy
Photoacoustic spectroscopy refers to the photoacoustic effect for the generation of the acoustic pressure
wave from an object (refer Fig. 20) [113]. In this spectroscopy technique, the absorption of modulated
optical input provides the estimation of blood glucose detection [114]. High intense optical light is
absorbed by an object according to its optical conditions [115]. This process provided excitations of
particular molecules according to its resonant frequency [116]. The absorbed light is considered as heat
which provides rising in localized temperature and thermal expansion of the sample [117]. The expansion
in volume generates pressure in acoustic form [118]. The generated photoacoustic wave can be used
to predict the glucose concentration through specific excited wavelengths which are resonant for the
vibration of glucose molecules [119]. At the specific resonance frequency, the glucose molecule changes
own characteristic. This change is in the acoustic waveform [120]. In previous work, 905 nm wavelength
optical light is used for excitation [121], [122].
H. Capacitance Spectroscopy
In the capacitance spectroscopy technique, inductor stray capacitance varies according to body ca-
pacitance (Fig. 21) [123]. The body capacitance is used to estimate body glucose concentration [124].
Flexible inductor based sensor follows the coupling capacitance principle for body glucose detection.
In this technique, there is not any interaction between the inductive sensor and body skin through the
current [125]. This is the advantage of the impedance spectroscopy technique. The stray capacitance of
the inductive sensor will vary according to body glucose. In this technique, the effect of fat and muscles
will be negligible with respect to body glucose [126].
I. Surface Plasmon Resonance (SPR)
The Surface Plasmon Resonance (SPR) utilizes electron oscillation approach at dielectric and metal
interface for glucose sensing [127]. It detects mainly the change in refractive index before as well as
after the analytes interaction. The optical fiber based SPR has been used for point of care measurement
for glucose due to its portability.
J. Radio Frequency (RF) Technique and Microwave Sensing
In the RF technique, the variation in the s-parameters response reflects the change in blood glucose
[128], [129]. Fig. 22 shows typical steps of this technique. The response is determined through the antenna
Pulse Laser
Diode Object PZT LNA ADC
Clark error
grid Analysis
LS9 Series, 100
Hz TTL ckt Finger/Ear
PWM Circuit
(905 nm)
Pressure to voltage
Converter 351A
Best fit
Over 1024
frames using
Lead zirconate
Embedded System
Fig. 20: Photo acoustic spectroscopy.
Voltage Control
Circuit with stray
and phase)
Analog to
and estimated
Digital to
and Storage
of Data
to skin
Fig. 21: The typical steps of capacitance spectroscopy.
or resonator [130], [131]. They follow the changes in dielectric constant value through the transmission
[132]. The change in dielectric constant can be found as the change in resonance frequency spectrum
through the antenna or resonator [133], [134]. The dielectric of blood varies according to blood glucose
concentration. The human finger is an appropriate measurement object but there are many factors that
play a cardinal and dominant role in the accuracy of measurement and repeatability. These are; the skin
thickness, fingerprints, the applied pressure by the fingertip during measurement and positioning of a
finger on the sensor [135].
Ferrite Antenna
Implantable Tag
Converter Temperature
Fig. 22: Glucose measurement using RF sensing technique.
K. Ocular Spectroscopy
In the Ocular Spectroscopy technique, glucose concentration is measured through the tears. A specific
lens is used to predict the body glucose concentration [136]. A hydrogel wafer is deposited to the lens.
This wafer is prepared by boronic acid with 7 µm thickness. The wafer is deposited on lens and then
optical rays are inserted on the lens. Then reflected light will change its wavelength. Change in wavelength
will refer to a change in glucose concentration in tears.
L. Iontophoresis
In the Iontophoresis or Ionization technique, a small electric current passes through the skin diffusively.
Three electrodes are used for the same [137]. A small potential is applied through the electrodes to
the different behaviour electrodes. During this process, glucose is transferred towards the cathode. The
working electrode can have the bio-sensing function by the generation of current during applied potential
through electrodes. This biosensor determines passively body glucose. The measurement is possible
through wrist frequently [138].
M. Optical Coherence Tomography
The Optical Coherence Tomography technique is based on the principle followed by reflectance
spectroscopy. In this technique, low coherent light is excited through the sample (sample is placed in
an interferometer). In an interferometer, a moving mirror is placed in reference arc. A photodetector is
placed on another side and it detects the interferometric signal. This signal contains backscattered and
reflected light. Due to this process, we could get high-quality 2-D images. The glucose concentration
increases with the increment of the refractive index in interstitial fluids. Change in the refractive index
indicates the change in the scattering coefficient [107]. So, the scattering coefficient relates to glucose
concentration indirectly.
N. Polarimetry
The Polarimetry technique is commonly used in a clinical laboratory with more accuracy. The optical
linear polarization-based technique is used for glucose monitoring [139]. This technique is usually based
on the rotation of vector due to thickness, temperature and concentration of blood glucose. Due to the
process of prediction of glucose, the polarized light is transmitted through the medium containing glucose
molecule. Due to high scattering through the skin, the depolarization of beam is possible. To overcome
this drawback, a polarimetric test has been done through the eye. The light passes through the cornea.
This technique is totally unaffected due to rotation of temperature and pH value of blood [140].
light source
Polarizer Angle
Path Length
Control Unit
and Data
Fig. 23: Non-invasive glucose measurement using Polarimetry.
O. Thermal Emission Spectroscopy
The Thermal Emission Spectroscopy based technique is based on the naturally generated IR wave
from the body. The emitted IR waves will vary according to body glucose concentration. The usual
mid-IR emission from tympanic membrane of human body is modulated with tissue emitting. The
selectivity of this technique is same as the absorption spectroscopy. Due to this technique, glucose
can be determined through the skin, fingers and earlobe. This technique is highly precise and accurate
for glucose measurement [141]. It could provide the useful solution which is precise and acceptable at
clinical with measurement of thermal emission from tympanic membrane.
P. Ultrasound
The Ultrasound method is based on low frequency components to extract the molecules from skin
similar as reverse iontophoresis method [142]. It is also alike sonophoresis and has larger skin permeability
than reverse iontophoresis. Few or several tens of minutes of ultrasound exposure are required to pull
glucose outward through the skin. There are few attempts for such technology and there is not any
commercial device with such type of technology.
Q. Metabolic Heat Conformation (MHC)
The Metabolic Heat Conformation (MHC) method helps to measure the glucose value with metabolic
heat and oxygen level along various physiological parameters considerations [143]. The mathematical
model for metabolic energy conservation has been modified by several physiological parameters consid-
eration such as pulse rate, oxyhemoglobin saturation, heat metabolic rate and the blood flow volume.
This method has shown good reproducibility and decent accuracy in humans.
R. Fluorescence
The Fluorescence technique is based on the excitation of blood vessels by UV rays at particular
specified frequency ranges [144]. This is followed through the detection of fluorescence at a specified
wavelength. The sensing of glucose using fluorescence through tear has been done by the diffraction of
visible light. At 380 nm, an ultraviolet LASER was taken for excitation through the glucose solution
medium. Fluorescence was estimated which is directly related to glucose concentration. In this technique,
the signal is not affected by variation in light intensity through the environment.
S. Kromoscopy
The Kromoscopy technique uses the response from various spectroscopic of NIR light with four
different detectors over different wavelength [145]. It employs the multi-channel approach with overlapped
band-pass series filters to determine the glucose molecule. In this method, the radiation of IR are imparted
on the sample and this will be divided among four detectors with band-pass filter. Each detector will
detect the light of the similar structure of the tissue. Subsequently, the complex vector analysis has been
utilised to measure the glucose concentration.
T. Electromagnetic Sensing
In the Electromagnetic Sensing method, the variations in blood sample conductivity is observed by
change in blood glucose concentration [146]. The alternation of electric field would be measured by
electromagnetic sensor whenever there will be change in blood glucose concentration. This method utilizes
the dielectric parameter of the blood samples. The frequency range for electromagnetic sensing is in the
range of 2.4 to 2.9 MHz. The glucose molecule has maximum sensitivity at particular optimal frequency
for given temperature of the medium.
U. Bioimpedance Spectroscopy and Dielectric Spectroscopy
It is useful to measure the variation of the blood glucose with help of conductivity and permittivity
from red blood cells membrane [147]. The spectrum of bioimpedance spectrum is measured from 0.1
to 100 MHz frequency range. It help to find the resistance with passing through electric current which
is flowing from human biological tissue. The change of plasma glucose would allow the changes in
potassium and sodium to have the change in conductivity of the membrane of the red blood cell. The
multisensor approach is usually incorporated with this spectroscopy in order to measure sweat, moisture,
movement and temperature for precise glucose measurement.
V. Reverse Ionospheresis
The small DC current is passed from anode to cathode on the skin surface to have small interstitial fluid
(ISF). Iontophoresis is employed for ionized molecules penetration at skin surface by such low current
[148]. The electric potential is passed from anode and cathode to electroosmotic flow across the skin.
This would allow to extract the molecules through skin whereas the the molecules of glucose are moved
towards the cathode. The enzyme method helps to sense the concentration of glucose molecules through
oxidation process. the method has very widely accepted and has good potential to measure accurate
glucose value.
W. Sonophoresis
The Sonophoresis technique is based on the cutaneous permittivity of the interstitial fluid (ISF) [149].
It also uses enzyme method for glucose measurement. The low frequency ultrasound wave has been
applied in order to have glucose molcules at the skin surface. The cutaneous permittivity of the ISF is
increased to enable glucose at the epidermis surface. The contraction and expansion occurs in stratum
corneum that subsequently opens the ISF pathway. There has been some attempts with this method for
glucose detection but it has been observed that it could be helpful in drug delivery in stead of glucose
X. Occlusion Spectroscopy
The Occlusion spectroscopy based methods depend on the concept of light scattering which is of inverse
proportion of glucose concentration [150]. The flow is ceased for few seconds by applying pressure with
pneumatic cuff. The volume of blood would change due to pulse generated from the pressure excursion.
The light is transmitted through the sample and the variation of the intensity of in a received light defines
the glucose concentration. The momentary blood flow cessation helps to get higher SNR value of the
received signal. Hence, the sensitivity for glucose detection would be increases with good robustness for
accurate glucose measurement.
Y. Skin Suction Bluster Technique
The Skin Suction Bluster technique uses the concept of blister generation through vacuum suction
over limited skin area [151]. The glucose measurement is performed on fluid which is collected from the
blister. It has lower glucose molecules than plasma but it is well enough to have the glucose measurement.
This method has low risk of infection, painless and well-tolerated. It is actually useful to measure HbA1c
value which represents three month average glucose value.
Z. Multimodal approach based measurement
A two modal spectroscopy combining IMPS and mNIR spectroscopy is explored for high-level re-
producibility of non-invasive blood glucose measurement [152]. These two techniques are combined to
overcome the limitation of individual employed technique [153]. Impedance spectroscopy based circuit
measures the dielectric constant value of skin or tissue through RLC resonant frequency and impedance to
predict glucose level [154]. To improve the accuracy of NIR spectroscopy, mNIR spectroscopy technique
is used. In this technique, three wavelengths 850 nm, 950 nm and 1300 nm are used [155]. For precise
and accurate measurement, IMPS and mNIR are joined by an ANN (Artificial Neural Network) through
DSP processor [80]. Multiple techniques are used concurrently to reduce noise from the signal [156].
Optimized signal conditioning circuits are used for removing garbage part of the received signal for final
estimated value. Somewhere, optimized differential amplifiers are designed for noise cancellation. These
are useful for biomedical measurement applications [157]. These signal conditioning circuits are used
as post-processing approaches for glucose measurement [158], [159]. This device explored high-level
accuracy in clinical trials. Based on the Clarke error grid analysis, the measured values are in A zone.
This device has USB connectivity, battery charge and alerts for various stages of glucose level.
Spectroscopy IC
Impedance Spectroscopy mNIR Spectroscopy
(850, 950 & 1300nm)
Data Combining
Through ANN
Glucose Value
Fig. 24: Multimodal IC based non invasive glucose measurement.
This Section presents various post-processing and calibration techniques which are deployed in various
systems or frameworks for noninvasive glucose-level monitoring.
TABLE II: Approaches Comparison with Noninvasive Works [25], [32].
Works Spectroscopy Spectra Specific Measurement Linearity
technique wavelength range (%)
Singh, et al. [160] Optical - - 32-516 mg/dl 80
Song, et al. [80] Impedance and Reflectance NIR 850-1300 nm 80-180 mg/dl -
Pai, et al. [161] Photoacoustic NIR 905 nm upto 500 mg/dl -
Dai, et al. [162] Bioimpedance - - - -
Beach, et al. [163] Biosensing - - - -
Ali, et al. [88] Transmittance and Refraction NIR 650 nm upto 450 mg/dl -
Haxha, et al. [77] Transmission NIR 940 nm 70-120 mg/dl 96
Jain, et al. [164] Absorption and Reflectance NIR 940 nm 80-350 mg/dl 90
Jain, et al. (iGLU 1.0)
[32], [28]
Absorption and Reflectance NIR 940 and 1300 nm 80-420 mg/dl 95
Jain et al. (iGLU 2.0)
[29], [25]
Absorption and Reflectance NIR 940 and 1300 nm 80-420 mg/dl 97
TABLE III: Statistical and Parametrical Comparison with Noninvasive Works [25], [32].
Works R MARD AvgE MAD RMSE Samples Used Measurement Device
value (%) (%) (mg/dl) (mg/dl) (100%) model sample cost
Singh, et al. [160] 0.80 - - - - A&B Human Saliva Cheaper
Song, et al. [80] - 8.3 19 - - A&B Human Blood Cheaper
Pai, et al. [161] - 7.01 - 5.23 7.64 A&B in-vitro Blood Costly
Dai, et al. [162] - 5.99 5.58 - - - in-vivo Blood Cheaper
Beach, et al. [163] - - 7.33 - - - in-vitro Solution -
Ali, et al. [88] - 8.0 - - - A&B Human Blood Cheaper
Haxha, et al. [77] 0.96 - - - 33.49 A&B Human Blood Cheaper
Jain, et al. [164] 0.90 5.20 5.14 5.82 7.5 A&B Human Blood Cheaper
Jain, et al. (iGLU 1.0)
0.95 6.65 7.30 12.67 21.95 A&B Human Blood Cheaper
Jain et al. (iGLU 2.0)
0.97 4.86 4.88 9.42 13.57 Zone A Human Serum Cheaper
A. Post processing and calibration techniques
Various calibration processes have been applied for a high level of accuracy and noise reduction from
received signal. These post-processing techniques are used to design the model for errorless continuous
monitoring [165], [166].
1) Noise Minimization and Signal Conditioning: The coherent averaging technique has been adapted
to minimize the variance of random noise [167]. The impact of noise is minimized with averaging of N
number of individual samples coming from the continuous frames [168]. Frames in the maximum count
have been chosen for averaging to have SNR improvement [169]. This proposed coherent averaging has
been used frequently through MATLAB and coherent averaged signal acquired. Golay code has been
proposed as calibration of measured data. The filtering or cancellation of unusual measured data has
been achieved through the implementation of Golay code [170], [171], [172].
2) Computation Models for glucose Estimation: The regression model of regularized least square is
proposed by several researchers for measurement [173]. The estimated value is computed from photoa-
coustic signals. These photoacoustic signals are used to calibrate for estimation of glucose concentration
[174]. This can be possible through multi-variable linear regression model [175]. With the objective
of high-level accuracy, a post-processing SVM technique is proposed [176]. Support vector machine is
a better option of correct measurement in glucose monitoring system [177]. Artificial Neural Network
(ANN) has also been proposed for data combining [178]. The measured data from multiple techniques
are combined through the proposed neural network model [179]. This artificial neural network has
been implemented in DSP processor [180]. This proposed data interpretation model has been used for
combining and calibrating of data for final estimated glucose concentration [181].
B. Metrics for Model Validation
The calibration method is used to have precise blood glucose estimation for measurements [182].
The obtained glucose concentration values are used to compare with conventionally measured glucose
concentrations [183]. The Clarke error grid analysis has been considered maximum measurement for
analysis which is used to check the performance of any device for accuracy measurement [184]. The
process flow is represented in Fig. 25.
Error Analysis
from Predicted
Capillary Glucose
Concentration (mg/dl)
Prediction from Calibrated
Machine Learning Models
for Validation
Analysis of
Serum Glucose
Concentration (mg/dl)
through Machine
Dataset Dataset
Error Analysis
from Predicted
Data Acquisition System
of Non-invasive Glucose
Data Acquisition System
of Non-invasive Glucose
Fig. 25: Representation of Metrics for model Validation.
C. Clinical Accuracy Evaluations using Clarke error grid analysis
The Clarke Error Grid has been analysed as benchmark tool to examine the clinical precision for
biomedical application. It has prediction of point as well as rate accuracy, and it amends for physiologic
time lags inherent for measurement of body glucose. The exploitation of the Clarke error grid modelling
will significantly make easy the development and refinement of a precise biomedical device. In 1970,
this technique was developed by C.G. Clark to identify the accuracy of the clinical trials which helps to
find the precision of estimated blood glucose with blood glucose value through the conventional method.
A description of the Error Grid Analysis came into view in diabetes care in 1987. The grid is divided
with five different zones mainly as zone A, zone B, zone C, zone D, and zone E. If the values residing in
either zone A or zone B then it signifies satisfactory or accurate prediction of glucose results according
to Beckman analyzer. The zone C values may prompt gratuitous corrections which may lead to a poor
outcome. If the values are under zone D which actually defines a hazardous failure to sense. Zone E
reflects the “erroneous treatment” [185].
Zone A-
Clinically Accurate
Zone B-
Zone C-
Zone D-
Failure to Detect
Zone E-
Fig. 26: Clarke error grid analysis.
There have been several non-invasive glucometer at market(such as Freestyle Libre sensor, SugarBEAT
from Nemaura medical) which used for proper medication. They would be like skin-patch with daily
disposable feature and adhesive to have the continuous glucose monitoring. Such solutions may not be
consider as completely non-invasive device which would help in frequent glucose measurement. Therefore,
they may not be useful for continuous monitoring.
There are some products as DiaMon Tech, glucowise, glucotrack, glutrac and CNOGA medical device.
Glutrac is smart healthcare device but it has accuracy issues for the blood glucose measurement. It has
higher cost while precision is still not acceptable. The non-invasive stripless device known as Omelon B-2
has been used for the CGM. The fluorescent technique based Glucosense has been made for contonuous
monitoring of the glucose value. The flexible textile-based biosensor has developed from Texas University
to measure the glucose level. All the available device have accuracy issues and considerable higher cost.
A. Wearable versus Non-wearable for Glucose Monitoring
The glucose monitoring have been attempted using non-wearable and wearable solutions in the lit-
erature. Most of the non-wearable approaches are based on photoacoustic spectroscopy and Raman
spectroscopy which are also of invasive type. The implantable devices are of semi-invasive type and
are mainly of biosensors in nature. Sweat patches, Glucowatch and Smart contact lenses are of wearable
devices category. LifePlus has developed non-invasive and wearable device for CGM purpose and it is
under consideration for commercialization. Most of non-invasive device are wearable and helpful for
frequent glucose measurement.
B. Noninvasive Glucose Measurement Consumer Products
There are variety of products such as GlucoTrack®, glucometer from Labiotech [186], and similar
available solutions have accuracy issues and cost is also high. The glucowise is another non-invasive device
for continuous glucose measurement from Medical Training Initiative (MTI) . The Raman scattering
spectroscopy based non-invasive solution is also developed by 2M Engineering [187]. These devices
are not much popular because of their cost and precision. Further for the high level of accuracy of
glucose measurement, Glucotrackhas been developed by integrity applications Ltd. [188]. This non-
invasive glucose monitoring device employed three consecutive ultrasonic spectroscopy, thermal emission
and electromagnetic techniques. This device is highly precise and accurate because of a combination of
three techniques [189]. A comparative perspective of various consumer products for noninvasive glucose
measurement has been summarized in Table IV.
TABLE IV: A Comparative perspective of a selected consumer products for noninvasive glucose
Company Device Technology Object Summary Snapshot
Reverse ion-
It would be worn as watch is
used with disposable compo-
nent, autosensor which is to be
attached at back of biographer
that contact with the skin to
provide frequent glucose mon-
Finger On basis of tissue photography
analysis from fingertip capillar-
ies, this device can analyze vari-
ous bio parameters in very short
Pendra Impedance
It helps to measure the glucose
with sodium transport of ery-
throcyte membrane, The change
of fluxes of transmembraneous
sodium occur due to impedance
field which is detected by de-
vice to generate final glucose
It is based on optical con-
cept on finger which is at-
tached to a ring-shaped sensor
probe. The probe has red/near-
infrared RNIR spectral region
light source as well as detector.
It has pneumatic cuffs which
generates systolic pressure to
produce optical signal for glu-
cose monitoring
Company Device Technology Object Summary Snapshot
C8 Medis-
ensor Glu-
cose detec-
Raman Spec-
This technique is based on
monochromatic light source
passes through skin where
scattered light is detected. The
generated colors from Raman
spectra helps to exact chemical
structure of glucose molecule
Glucotrack Combination
of Electro-
ultrasonic and
Ear lobe
In this device, three differ-
ent techniques are used concur-
rently to increase the accuracy
and precision
Infra red
Finger It is helpful for Hyperglycemia
or Pre-Diabetic patients which
allow for regular monitoring of
precise blood glucose measure-
ment at every 30 seconds
Glucowise Radio Wave
This non invasive wireless de-
vice can measure glucose con-
centration in very short time.
It is based on electromagnetic
waves of specific frequencies
for blood glucose detection. It
uses a thin-film layer of meta-
material which increases the
penetration for precise glucose
SugarBeat Reverse ion-
This has been proved accu-
rate device, pain-free contin-
uous blood glucose monitor-
ing. SugarBEAT®provides real-
time, needle-free glucose mea-
surement. Generally, it needs
one time finger-prick test for
calibration. One time finger
prick is used when new patch
is required to insert
Company Device Technology Object Summary Snapshot
Free Style
Glucose oxi-
dase method
It uses enzyme glucose sens-
ing technology for the detec-
tion of glucose levels through
interstitial fluid Glucose oxidase
method is applied through sen-
sor where electrical current pro-
portional to the glucose con-
centration and glucose can be
Raman Spec-
Fore arm
Raman spectroscopy technique
based this device can detect glu-
cose in blood through returning
spectrum from the skin
Optical Tech-
Finger NA
Various models have been developed for diet control using various parameters for glucose-insulin
balance. The parameters are mainly includes net hepatic glucose balance, renal excretion rate, glucose
absorption rate and peripheral glucose utilization for the glucose consumption prediction for the diabetic
patients. These are useful parameters to calculate the glucose level by proper insulin dosage along with
scheduled diet plan. Therefore, the glucose-insulin control model was designed to balance glucose insulin
level in the body for diabetes persons using proper medication.
A. Glucose Controls Approaches
The mathematical models for insulin delivery have been presented to determine the coefficients of
blood regulation. The model has been proposed for insulin secretion with glycemic profile for type 2
diabetic person [190], [191]. The non-linear model is developed using differential equation with delay
model with help of non-diabetic subjects [192]. Most poplar “Uva/Padova Simulator” was also explored
which was approved from FDA to have the proper clinical trials. The parameter are extracted with type
1 diabetic virtual patients [193]. The intravenous test for glucose tolerance with Hovorka maximal model
has explored for non-diabetic subjects [194]. The samples from type 1 diabetic persons were collected to
explain the model with help of time monitoring. The model is proposed mathematically for blood glucose
value prediction in the postprandial mode for type 1 diabetes patients [195], [196]. The mathematical
model for glucose-insulin balance for longer period is explored using two days clinical information
[197]. A algorithm was developed for T1DM patient meal detection for the purpose of frequent glucose
measurement. The work has integrated bolus meal mathematical model for glucose-insulin delivery model
[198]. Diabetic and healthy people were considered to acquire the values for the variable state dimension
algorithm. The diet plan was examined in the absence of meal profile to have the glycemic profile balance,
an intelligent PID controller (iPID) was developed to type 1 diabetic person [199], [200].
B. Glucose Controls Consumer Products
Type-1 diabetic patients aren’t able to produce insulin. Insulin is a hormone that can balance body
sugar (glucose) which is a prime source of energy that obtains from carbohydrates. If anybody has type 1
diabetes, it is necessary to be ready for insulin therapy. Insulin may be injected by self-injection, patients
who take multiple injections daily of insulin may also think about use of an insulin pump. An insulin
pump gives short-acting insulin all day long continuously. The insulin pump replaces the requirement of
long-acting insulin. A pump also substitutes the requirement of multiple injections per day along with
continuous insulin infusion and also serves to improve the glucose levels. Various types of insulin pumps
are already available in the market as consumable product mainly as Animas, Medtronic, Roche, Tandem
and Omnipod insulin pump are consumables. These insulin pumps are advanced to each other in terms of
their upgraded features. A comparative perspective of a selected consumer products for glucose control
is presented in Table V.
TABLE V: A comparative perspective of a selected consumer products for glucose control.
Work Technology Object Findings Observation
[86] photoplethy-
Finger It helps to extract the fea-
tures of PPG signal through
machine learning models to
estimate Systolic and di-
astolic blood pressure and
blood glucose values
machine learning models ap-
plied where random forest tech-
nique has best prediction results
as R2
SB P = 0.91, R2
DBP = 0.89
and R2
BGL = 0.90. CEG has
87.7% observation in Zone A,
10.3 % in Zone B, and 1.9% in
Zone D
[201] mid-infrared
attenuated total
and trapezoidal
ATR prism
oral mucosa
inner lips
Using a multi-reflection
prism brought about higher
sensitivity, and the flat and
wide contact surface of the
prism resulted in higher
measurement reproducibility
& spectra around 1155
cm1for different blood
glucose levels for fasting
and before fasting
the coefficient of determination
R2is 0.75. The standard error is
12 mg/dl, and all the measured
values are in Region A
[202] Optical Coher-
ence Tomogra-
Fingertip It measures the optical ro-
tation angle and depolariza-
tion index of aqueous glu-
cose solutions with low and
high scattering, respectively.
The value of angle increases
while depolarization index
decreases with glucose value
The correlation factor has a
value of R20.9101, the average
deviation is found around 0.027.
Work Technology Object Findings Observation
[203] Contactlenses
Tears r the fabrication of a soft,
smart contact lens in which
glucose sensors, wireless
power transfer circuits, and
display pixels to visualize
sensing signals in real time
are fully integrated using
transparent and stretchable
[204] transmission
Sliva After completely absorbing
the sufficient amount of
saliva on the strip, the sam-
ple would reach detection
zone via paper microfluidic
movement and enzymatic re-
action between GOx and
salivary glucose would ini-
tiate a pH change, resulting
in a change in strip color
that was recorded by using
RGB detector on the hand-
held instrument which helps
for glucose detection
The developed biosensor had
a wide detection range of de-
tection between 32- and 516-
mg/dL glucose concentration
while the sensitivity of detec-
tion was 1.0 mg/dL/count at a
limit of detection (LOD) of 32
mg/dL within a response time
of 15 s
[205] impedance
and multi-
wrist Hand
IMPS and mNIRS use the
indirect dielectric character-
istics of the surrounding tis-
sue around blood and the
optical scattering character-
istics of glucose itself in
blood, respectively, the pro-
posed IC can remove various
systemic noises to enhance
the glucose level estimation
mean absolute relative
differences (mARD) to 8.3%
from 15.0% of the IMPS and
15.0–20.0% of the mNIRS
in the blood glucose level
range of 80–180 mg/dL. From
the Clarke grid error (CGE)
analysis, all of the measurement
results are clinically acceptable
and 90% of total samples can
be used for clinical treatment
Work Technology Object Findings Observation
[84] NIR
Fingertip short NIR waves with ab-
sorption and reflectance of
light using specific wave-
lengths (940 and 1,300 nm)
has been introduced
The Pearson’s correlation coef-
ficient (R) is 0.953 and MAD is
09.89 which is RMSE 11.56
[206] Microwave De-
earlobe The absorption spectrum of
microwave signal helps to
measure using two antenna.
The sine wave of 500 MHz
is for blood glucose mea-
It can measure blood glucose
from 0 to 500 mg/dl with step
size of 200 mg/dl used for
the experiment for testing the
resolution. It obtained 0.5226
mean standard deviation while
the minimum value of standard
deviation is 0.04119.
[94] PPG Finger The prediction of blood glu-
cose was with machine-
learning using a smartphone
camera. First the invalid
data was separated from the
valid signals for two blood-
glucose groups. The system
did not require any type of
The device was able to mea-
sure glucose only 70-130 mg/dl
range. The results show accu-
racy of 98.2% for invalid single-
period classification and and the
overall accuracy is 86.2%.
[46] MEMS Finger It is minimally invasive tech-
nique known as e-Mosquito
which extracts blood sam-
ple with shape memory al-
loy (SMA)-based microactu-
ator. It considered as first
ever wearable device which
performs the automatic situ
blood extraction and per-
forms the glucose analysis.
The method provided linear cor-
relation (R2= 0.9733) between
standard measurements and the
e-Mosquito prototype.
[207] Visible NIR Wrist The paper developed biosen-
sor which helps to exploit
pulsation of arterial blood
volume from the wrist tis-
sue. The visible NIR spec-
troscopy was used for re-
flected optical signal to es-
timate blood glucose.
The correlation coefficient (Rp)
value after averaging all obser-
vation is 0.86, whereas the stan-
dard prediction error is around
6.16 mg/dl.
Work Technology Object Findings Observation
[201] mid-infrared
attenuated total
inner lip
Novel optical fiber probe
was introduced using mul-
tireflection prism with help
of ATR spectroscopy. The
sensitivity increases with the
number of reflections while
measurement reproducibility
was higher due to prism’s
wide and flat and wide con-
tact surface.
The practical and sustainable mechanisms are the prime factors of smart and automated healthcare
system. These are being optimized to support the population migration and quality of life in smart cities
and smart villages [208], [209]. The features of smart healthcare system are continuous monitoring for
critical care, ambient intelligence and quality of service for proper point of care mechanism [210], [211].
The non-invasive and precise glucose measurement is requirement for diabetic person and would also
needed to store the information using IoMT for proper treatment [26]. The traditional method for glucose
measurement has limited capability and is not able to assist the remotely located healthcare provider. The
diabetic person would like to monitor their glycemic profile frequently in a day with support of storing
at cloud server. The smart health care system would allow the point of care treatment for diabetes person
with frequent monitoring.
The internet of Medical Things (IoMT) has allowed to connect the patients with doctors remotely
for rapid treatment and special assistance using smart healthcare [208]. The continuous monitoring of
vital parameters have provided to awareness about the diet plan and routine activity management with
contemporary healthcare consumers devices. With the active support of remote healthcare solution, the
smart healthcare has potential to ameliorate the quality of service at reduced cost. The smart sensors
would capture the patient data continuously and help to store the data on cloud data centre. It is also
useful for the analysing the data and easy exchange of the information through mobile applications to
doctors as well as patients. The healthcare Cyber-Physical System (H-CPS) has been used successfully
to address the various challenges of healthcare sector with intelligent algorithms.
The continuous glucose monitoring would certainly help the diabetic patients to plan their diet for
the purpose of glucose control. The solution should be precise, low cost and easy to operate for rapid
diagnosis [32], [28]. The serum glucose would always consider as accurate than capillary measurement.
Therefore, the rapid serum glucose measurement solution with continuous monitoring is desired for the
smart healthcare. The novel serum glucometer is portable device and is also integrated with IoMT to
store the glucose values continuously at cloud. It would be useful for the healthcare provider to track
the health records of remote located diabetes person. The smart healthcare management of continuous
glucose measurement is defined in Fig. 27.
A detailed example of a closed-loop system that presents glucose-level monitoring and insulin release
to control it is illustrated in Fig. 28 [2]. This IoMT framework can provide a better solution for evaluation
of insulin doses through the closed-loop automated insulin secretion diabetes control. Such an integrated
IoMT framework can be implemented to diagnose and for the treatment of diabetic patients in terms of
controlling their blood glucose level in smart healthcare and be effective in smart village and smart cities
for healthcare with limited medical personnel.
The security and privacy issues of the medical devices are paramount aspect in any IoT network. The
hardware security of wearable device is very crucial because control actions mainly occur in wireless
Smart healthcare
For Glycemic
Profile balance
Data storage
Analysis of
glucose insulin
Fig. 27: Blood glucose diagnosis and Control in smart health care system.
media. The security vulnerabilities are defined for glucose measurement device and its control are shown in
Fig. 29. The devices security are important due to connected health system in an insecure and unreliable
IoMT framework [212]. The integrity of useful medical information is also crucial security aspect of
smart healthcare. All patients medical records are stored over the server therefore the security of such
data are also really important. The controlled access with proper authentication is required to have secure
monitoring with proper patient treatment.
This Section outlines the shortcomings and discusses some open problems of glucose level measure-
ments and control.
A. Limitations of the Existing Approaches and Products
1) Photoacoustic spectroscopy has been implemented for glucose measurement. Real-time testing and
validation have not been done from human blood. The artificial solution was prepared in the
laboratory for glucose measurement. The prototype module with LASER and corresponding detector
is costly and at the same time requires considerable bigger area and does not provide portable
solution. Therefore, it is not much popular solution for continuous glucose monitoring.
2) Raman spectroscopy is a nonlinear scattering which occurs when monochromatic light interacts
with a certain sample. Raman spectroscopy based solution is applicable for a laboratory test and
also occupies the significant larger area. Hence, the system based on this approach will not be
applicable for frequent glucose measurement.
Prescribed Schedule
for Diet and Insulin Secretion
Glucose-Insulin Model
(Artificial Pancreas System)
for Test
Output for
Prescribed Schedule
Internet Connectivity
and Cloud Storage Diabetologist
Blood Glucose
Level (mg/dl)
Plasma Insulin
Level (mU/L)
Other Glucose
Blood Glucose
Level (below
70 mg/dl)
Blood Glucose
Level (between
70-150 mg/dl)
Blood Glucose
Level (above
150 mg/dl)
No Need
of Insulin
Fig. 28: A closed-loop automated insulin secretion diabetes control system in an IoMT framework [2].
Display of Parameters
Wearable for Noninvasive
Serum Glucose
Monitoring iGLU 2.0
Hospital Doctor
Cloud Storage
Medical Device Security
Integrity of
Channel Data
Access Control
of Personal Data
Fig. 29: Our Long-Term Vision of Security-Assured Non-invasive Glucose-Level Measurement and
Control through our Proposed iGLU.
3) The retina based glucose measurement is also one of the alternate non-invasive glucose detection
approach, data has also been collected through retina for glucose measurement. Such technique is
not useful for the glucose measurement all the time.
4) In case of bio-capacitance spectroscopy, the slight difference in placing the sensor at the same
location might affect the output of the sensor. Effect of pressure on the sensor, body temperature
and sweat on the skin may also affect the output of the sensor.
5) Glucose detection is performed with the impedance spectroscopy (IMPS) by electrodes connection
to the skin which is affected with skin. The accuracy is always an issue as the saliva and sweat
could change for each individual and that may reflect to the precision of glucose. Therefore, this
technique is not best for reliable glucose measurement in smart healthcare.
6) PPG signal has been used to extract features for blood glucose level prediction. But the PPG may
be precise blood glucose measurement technique where the output value would vary according the
blood volume only. Therefore, the glucose molecule has not been detected precisely in the blood
sample using this technique.
B. The Open Problems in Non-invasive Glucose Measurement
There are lots of challenges for commercialization of non-invasive glucose measurement device. But,
some open problems have been discussed which are prime challenges for precise non-invasive glucose
measurement. These challenges have been represented in Fig. 30. The precise glucose measurement of
hypoglycemic patient and long-time continuous glucose measurement without instantaneous error are the
open problems which are focussed by the researchers recently.
Challenges for
Precision in
Measurement of
Overcome the
and Skin
Constraints for
Device with
advanced and
Fig. 30: Open Challenges in Noninvasive Glucose-Level Measurement.
The effect of blood pressure, body temperature and humidity have not been considered in the
literature which affect the values of glucose measurement.
The cost effective and portable solution of continuous glucose measurement device has also not been
addressed properly.
The accurate glucose measurement has been also been open challenge for full rage from 40 mg/dl
to 450 mg/dl.
The effective integration of glucometer with internet of medical things for continuously data logging
to the cloud has still not potentially resolved.
The mathematical model for automatic insulin secretion according to measured glucose value has
to be address in better manner with internet framework.
The privacy and security issues of insulin and blood glucose measurement system is still not resolved
The efficient power management mechanism has to be developed for continuous glucose measure-
ment with insulin delivery system.
The paper presents survey of glucose measurement approaches along with overview of glucose control
mechanism. Many techniques available in literature are only a proof of concept, showing good correlation
between device estimated result and reference value of blood glucose. However, they are neither accurate
nor cost effective solutions and not available for commercial purpose. The optical detection using short
NIR has been potential solution to mitigate the drawbacks of all previous methods. In future, the multi-
model approaches could be considered for precise glucose estimation. The device or prototype model
should be more effective in different zones to support the continuous health monitoring. It should be
implemented as a portable device for real time application with more frequently. This device should be
developed as continuous health monitoring with minimum cost.
The future research for upcoming noninvasive glucose monitoring device is mentioned in Fig. 31. The
device is required to be integrated with advanced IoMT framework. This advanced IoMT framework will
alow to connect the device with all nearest diabetic centers to get best treatment. Unification of glucose-
level measurement and automatic diet quantification can have strong impact on smart healthcare domain
[213]. The durability, portability and user-friendly device is also the future vision in this era. The device
should have the feature of border-line cross indication. Because of this feature, any person will be aware
to take own blood glucose level. A secured device with end to end users control and authentication is also
necessary for future advancement. Physical Unclonable Function (PUF) based security of IoMT-devices
can be effective for IoMT-devices which are intrinsically resource and battery constrained [212], [214].
Unified healthcare Cyber-Physical System (H-CPS) with blockchain based data and device management
can be effective and needs research [215], [216].
Wide Range
(30- 600 mg/dl)
Wearable and
Operation based
Device with free
Device with
Advanced IoMT
Durable and
Highly Precise
Device for
Diabetic Patients
A Secured
Device with end
to end users
A Low Cost
Device with
Cross Indication
Fig. 31: Our Future Vision for Non-invasive Glucose-Level Measurement.
[1] Diabetestalk, “What is the normal fasting blood sugar range for adults,” 2018, last Accessed on 18 Jan 2021. [Online].
Available: sugar-range-for-adults
[2] P. Jain, A. M. Joshi, and S. P. Mohanty, “iGLU 1.1: Towards a glucose-insulin model based closed loop iomt framework
for automatic insulin control of diabetic patients,” in 2020 IEEE 6th World Forum on Internet of Things (WF-IoT). IEEE,
2020, pp. 1–6.
[3] A. M. Joshi, U. P. Shukla, and S. P. Mohanty, “Smart healthcare for diabetes during COVID-19,” IEEE Consumer
Electronics Magazine, vol. 10, no. 1, pp. 66–71, January 2020.
[4] Amit M Joshi, U. P. Shukla, and S. P. Mohanty, “Smart healthcare for diabetes: A COVID-19 perspective,arXiv preprint
arXiv:2008.11153, 2020.
[5] A. M. Alsamman and H. Zayed, “The transcriptomic profiling of COVID-19 compared to SARS, MERS, Ebola, and
H1N1,” bioRxiv, 2020.
[6] I. D. Federation, “IDF Diabetes Atlas - Diabetes is rising worldwide... and is set to rise even further,” 2019, last
Accessed on 21 March 2020. [Online]. Available:
[7] N. H. Cho, J. E. Shaw, S. Karuranga, Y. Huang, J. D. da Rocha Fernandes, A. W. Ohlrogge, and B. Malanda, “IDF
diabetes atlas: Global estimates of diabetes prevalence for 2017 and projections for 2045,Diabetes Research and Clinical
Practice, vol. 138, pp. 271–281, April 2018.
[8] P. Saeedi, I. Petersohn, P. Salpea, B. Malanda, S. Karuranga, N. Unwin, S. Colagiuri, L. Guariguata, A. A. Motala,
K. Ogurtsova et al., “Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045:
Results from the international diabetes federation diabetes atlas,” Diabetes research and clinical practice, vol. 157, p.
107843, 2019.
[9] Clevelandclinic, “Diabetes Mellitus: An Overview,” 2020, last Accessed on 18 Jan 2021. [Online]. Available:
[10] Drugs, “Type 1 diabetes mellitus,” 2020, last Accessed on 18 Jan 2021. [Online]. Available:
[11] L. M. Leontis and A. Hess-Fischl, “Diabetes mellitus: An overview,” 2019, last Accessed on 18 Jan 2021. [Online].
[12] A. Pietrangelo, “What are the different types of diabetes?” 2018, last Accessed on 18 Jan 2021. [Online]. Available:
[13] M. J. Fowler, “Diabetes: Magnitude and mechanisms,Clinical Diabetes, vol. 28, no. 1, pp. 42–46, 2010. [Online].
[14] H. Yin, B. Mukadam, X. Dai, and N. Jha, “DiabDeep: Pervasive Diabetes Diagnosis based on Wearable Medical Sensors
and Efficient Neural Networks,IEEE Transactions on Emerging Topics in Computing, pp. 1–1, 2019.
[15] P. Zhang, “Global healthcare expenditure on diabetes for 2010 and 2030.Diabetes Research and Clinical Practice, 2011.
[16] J. Venkataraman and B. Freer, “Feasibility of non-invasive blood glucose monitoring: In-vitro measurements and phantom
models,” in 2011 IEEE International Symposium on Antennas and Propagation (APSURSI), July 2011, pp. 603–606.
[17] S. H. Wild, G. Roglic, A. Green, R. Sicree, and H. King, “Global prevalence of diabetes: Estimates for the
year 2000 and projections for 2030,” Diabetes Care, vol. 27, no. 10, pp. 2569–2569, 2004. [Online]. Available:
[18] D. R. Whiting, L. Guariguata, C. Weil, and J. Shaw, “Idf diabetes atlas: Global estimates of the prevalence of diabetes
for 2011 and 2030,” Diabetes Research and Clinical Practice, vol. 94, no. 3, pp. 311 – 321, 2011.
[19] P. H. Siegel, A. Tang, G. Virbila, Y. Kim, M. C. F. Chang, and V. Pikov, “Compact non-invasive millimeter-wave glucose
sensor,” in 2015 40th International Conference on Infrared, Millimeter, and Terahertz waves (IRMMW-THz), Aug 2015,
pp. 1–3.
[20] S. M. Alavi, M. Gourzi, A. Rouane, and M. Nadi, “An original method for non-invasive glucose measurement: preliminary
results,” in 2001 Conference Proceedings of the 23rd Annual International Conference of the IEEE Engineering in
Medicine and Biology Society, vol. 4, 2001, pp. 3318–3320 vol.4.
[21] X. Li and C. Li, “Study on the application of wavelet transform to non-invasive glucose concentration measurement
by nirs,” in 2015 Fifth International Conference on Instrumentation and Measurement, Computer, Communication and
Control (IMCCC), Sept 2015, pp. 1294–1297.
[22] P. P. Pai, P. K. Sanki, S. K. Sahoo, A. De, S. Bhattacharya, and S. Banerjee, “Cloud computing-based non-invasive glucose
monitoring for diabetic care,” IEEE Transactions on Circuits and Systems I: Regular Papers, vol. PP, no. 99, pp. 1–14,
[23] P. S. Reddy and K. Jyostna, “Development of smart insulin device for non invasive blood glucose level monitoring,” in
2017 IEEE 7th International Advance Computing Conference (IACC), Jan 2017, pp. 516–519.
[24] P. Jain, A. M. Joshi, and S. P. Mohanty, “iGLU: An Intelligent Device for Accurate Non-Invasive Blood Glucose-Level
Monitoring in Smart Healthcare,” IEEE Consumer Electronics Magazine, vol. 9, no. 1, p. Accepted, January 2020.
[25] A. M. Joshi, P. Jain, S. P. Mohanty, and N. Agrawal, “iGLU 2.0: A new wearable for accurate non-invasive
continuous serum glucose measurement in IoMT framework,IEEE Transactions on Consumer Electronics, no.
10.1109/TCE.2020.3011966, p. in Press, 2020.
[26] S. P. Mohanty and E. Kougianos, “Biosensors: A Tutorial Review,IEEE Potentials, vol. 25, no. 2, pp. 35–40, March
[27] N. A. Salam, W. H. M. Saad, Z. Manap, and F. Salehuddin, “The evolution of non-invasive blood glucose monitoring
system for personal application,” Journal of Telecommunication, Electronic and Computer Engineering, vol. 8, pp. 59–65,
[28] P. Jain, A. M. Joshi, and S. P. Mohanty, “iGLU 1.0: An Accurate Non-Invasive Near-Infrared Dual Short Wave-
lengths Spectroscopy based Glucometer for Smart Healthcare,” arXiv Electrical Engineering and Systems Science, no.
arXiv:1911.04471, November 2019.
[29] P. Jain, A. M. Joshi, N. Agrawal, and S. P. Mohanty, “iGLU 2.0: A New Non-invasive, Accurate Serum Glucometer for
Smart Healthcare,” arXiv Electrical Engineering and Systems Science, vol. abs/2001.09182, 2020. [Online]. Available:
[30] Q. Liu, Y. Liu, F. Wu, X. Cao, Z. Li, M. Alharbi, A. N. Abbas, M. R. Amer, and C. Zhou, “Highly Sensitive and Wearable
In2O3 Nanoribbon Transistor Biosensors with Integrated On-Chip Gate for Glucose Monitoring in Body Fluids,ACS
Nano, vol. 12, no. 2, pp. 1170–1178, 2018, pMID: 29338249.
[31] P. H. Siegel, W. Dai, R. A. Kloner, M. Csete, and V. Pikov, “First millimeter-wave animal in vivo measurements of
l-glucose and d-glucose: Further steps towards a non-invasive glucometer,” in 2016 41st International Conference on
Infrared, Millimeter, and Terahertz waves (IRMMW-THz), Sept 2016, pp. 1–3.
[32] P. Jain, A. M. Joshi, and S. P. Mohanty, “iGLU: An Intelligent Device for Accurate Non-Invasive Blood Glucose-Level
Monitoring in Smart Healthcare,” IEEE Consumer Electronics Magazine, vol. 9, no. 1, pp. 35–42, January 2020.
[33] N. M. Zhilo, P. A. Rudenko, and A. N. Zhigaylo, “Development of hardware-software test bench for optical non-
invasive glucometer improvement,” in 2017 IEEE Conference of Russian Young Researchers in Electrical and Electronic
Engineering (EIConRus), Feb 2017, pp. 89–90.
[34] S. I. Gusev, A. A. Simonova, P. S. Demchenko, M. K. Khodzitsky, and O. P. Cherkasova, “Blood glucose concentration
sensing using biological molecules relaxation times determination,” in 2017 IEEE International Symposium on Medical
Measurements and Applications (MeMeA), May 2017, pp. 458–463.
[35] M. W. Sari and M. Luthfi, “Design and analysis of non-invasive blood glucose levels monitoring,” in 2016 International
Seminar on Application for Technology of Information and Communication (ISemantic), Aug 2016, pp. 134–137.
[36] S. Lekha and M. Suchetha, “Non- invasive diabetes detection and classification using breath analysis,” in 2015 International
Conference on Communications and Signal Processing (ICCSP), April 2015, pp. 0955–0958.
[37] J. Li, P. Koinkar, Y. Fuchiwaki, and M. Yasuzawa, “A fine pointed glucose oxidase immobilized electrode for low-invasive
amperometric glucose monitoring,” Biosensors and Bioelectronics, vol. 86, pp. 90–94, 2016.
[38] N. Demitri and A. M. Zoubir, “Measuring blood glucose concentrations in photometric glucometers requiring very small
sample volumes,IEEE Transactions on Biomedical Engineering, vol. 64, no. 1, pp. 28–39, 2017.
[39] J. Y. Lucisano, T. L. Routh, J. T. Lin, and D. A. Gough, “Glucose monitoring in individuals with diabetes using a long-
term implanted sensor/telemetry system and model,” IEEE Transactions on Biomedical Engineering, vol. 64, no. 9, pp.
1982–1993, 2017.
[40] A. Sun, A. G. Venkatesh, and D. A. Hall, “A multi-technique reconfigurable electrochemical biosensor: Enabling personal
health monitoring in mobile devices,IEEE Transactions on Biomedical Circuits and Systems, vol. 10, no. 5, pp. 945–954,
Oct 2016.
[41] A. Gani, A. V. Gribok, Y. Lu, W. K. Ward, R. A. Vigersky, and J. Reifman, “Universal glucose models for predicting
subcutaneous glucose concentration in humans,” IEEE Transactions on Information Technology in Biomedicine, vol. 14,
no. 1, pp. 157–165, Jan 2010.
[42] G. Wang, M. D. Poscente, S. S. Park, C. N. Andrews, O. Yadid-Pecht, and M. P. Mintchev, “Wearable microsystem for
minimally invasive, pseudo-continuous blood glucose monitoring: The e-mosquito,” IEEE Transactions on Biomedical
Circuits and Systems, vol. 11, no. 5, pp. 979–987, Oct 2017.
[43] M. M. Ahmadi and G. A. Jullien, “A wireless-implantable microsystem for continuous blood glucose monitoring,IEEE
Transactions on Biomedical Circuits and Systems, vol. 3, no. 3, pp. 169–180, June 2009.
[44] G. Acciaroli, M. Vettoretti, A. Facchinetti, G. Sparacino, and C. Cobelli, “Reduction of blood glucose measurements to
calibrate subcutaneous glucose sensors: A bayesian multiday framework,IEEE Transactions on Biomedical Engineering,
vol. 65, no. 3, pp. 587–595, 2018.
[45] I. Pagkalos, P. Herrero, C. Toumazou, and P. Georgiou, “Bio-inspired glucose control in diabetes based on an analogue
implementation of a β-cell model,” IEEE Transactions on Biomedical Circuits and Systems, vol. 8, no. 2, pp. 186–195,
April 2014.
[46] G. Wang, M. D. Poscente, S. S. Park, C. N. Andrews, O. Yadid-Pecht, and M. P. Mintchev, “Wearable microsystem for
minimally invasive, pseudo-continuous blood glucose monitoring: The e-mosquito,” IEEE Transactions on Biomedical
Circuits and Systems, vol. 11, no. 5, pp. 979–987, 2017.
[47] T. Kossowski and R. Stasinski, “Robust ir attenuation measurement for non-invasive glucose level analysis,” in 2016
International Conference on Systems, Signals and Image Processing (IWSSIP), May 2016, pp. 1–4.
[48] L. P. Pavlovich and D. Y. Mynziak, “Noninvasive method for blood glucose measuring and monitoring,” in 2013 IEEE
XXXIII International Scientific Conference Electronics and Nanotechnology (ELNANO), April 2013, pp. 255–257.
[49] Y. Liu, W. Li, T. Zheng, and W. K. Ling, “Overviews the methods of non-invasive blood glucose measurement,” in 2016
IEEE International Conference on Consumer Electronics-China (ICCE-China), Dec 2016, pp. 1–2.
[50] N. K. Sharma and S. Singh, “Designing a non invasive blood glucose measurement sensor,” in 2012 IEEE 7th International
Conference on Industrial and Information Systems (ICIIS), Aug 2012, pp. 1–3.
[51] X. Zhao, Q. Zheng, and Z. M. Yang, “Two types of photonic crystals applied to glucose sensor,” in 2016 IEEE International
Nanoelectronics Conference (INEC), May 2016, pp. 1–2.
[52] Y. Tanaka, C. Purtill, T. Tajima, M. Seyama, and H. Koizumi, “Sensitivity improvement on cw dual-wavelength
photoacoustic spectroscopy using acoustic resonant mode for noninvasive glucose monitor,” in 2016 IEEE SENSORS,
Oct 2016, pp. 1–3.
[53] I. Gouzouasis, H. Cano-Garcia, I. Sotiriou, S. Saha, G. Palikaras, P. Kosmas, and E. Kallos, “Detection of varying glucose
concentrations in water solutions using a prototype biomedical device for millimeter-wave non-invasive glucose sensing,”
in 2016 10th European Conference on Antennas and Propagation (EuCAP), April 2016, pp. 1–4.
[54] Y. Nikawa and D. Someya, “Non-invasive measurement of blood sugar level by millimeter waves,” in 2001 IEEE MTT-S
International Microwave Sympsoium Digest (Cat. No.01CH37157), vol. 1, May 2001, pp. 171–174 vol.1.
[55] J. Shao, F. Yang, F. Xia, Q. Zhang, and Y. Chen, “A novel miniature spiral sensor for non-invasive blood glucose
monitoring,” in 2016 10th European Conference on Antennas and Propagation (EuCAP), April 2016, pp. 1–2.
[56] P. H. Siegel, Y. Lee, and V. Pikov, “Millimeter-wave non-invasive monitoring of glucose in anesthetized rats,” in 2014
39th International Conference on Infrared, Millimeter, and Terahertz waves (IRMMW-THz), Sept 2014, pp. 1–2.
[57] D. Wang, “An improved integration sensor of non-invasive blood glucose,” in The 7th IEEE/International Conference on
Advanced Infocomm Technology, Nov 2014, pp. 70–75.
[58] N. Bayasi, H. Saleh, B. Mohammad, and M. Ismail, “The revolution of glucose monitoring methods and systems: A
survey,” in 2013 IEEE 20th International Conference on Electronics, Circuits, and Systems (ICECS), Dec 2013, pp.
[59] R. Agrawal, N. Sharma, M. Rathore, V. Gupta, S. Jain, V. Agarwal, and S. Goyal, “Noninvasive method for glucose level
estimation by saliva,J Diabetes Metab, vol. 4, no. 5, pp. 2–5, 2013.
[60] N. Demitri and A. M. Zoubir, “Measuring blood glucose concentrations in photometric glucometers requiring very small
sample volumes,IEEE Transactions on Biomedical Engineering, vol. 64, no. 1, pp. 28–39, 2016.
[61] M. Shokrekhodaei and S. Quinones, “Review of Non-invasive Glucose Sensing Techniques: Optical, Electrical and Breath
Acetone,” Sensors, vol. 20, no. 5, p. 1251, 2020.
[62] S. Delbeck, T. Vahlsing, S. Leonhardt, G. S. G, and H. M. Heise, “Non-invasive monitoring of blood glucose using optical
methods for skin spectroscopy-opportunities and recent advances,Anal Bioanal Chem., vol. 411, no. 1, pp. 63–77, 2019.
[63] N. K. Madzhi, S. A. Shamsuddin, and M. F. Abdullah, “Comparative investigation using gaas(950nm), gaaias (940nm)
and ingaasp (1450nm) sensors for development of non-invasive optical blood glucose measurement system,” in 2014 IEEE
International Conference on Smart Instrumentation, Measurement and Applications (ICSIMA), Nov 2014, pp. 1–6.
[64] N. A. M. Aziz, N. Arsad, P. S. Menon, A. R. Laili, M. H. Laili, and A. A. A. Halim, “Analysis of difference light sources
for non-invasive aqueous glucose detection,” in 2014 IEEE 5th International Conference on Photonics (ICP), Sept 2014,
pp. 150–152.
[65] S. Tommasone, “Infrared spectroscopy: An overview,” 2018, last Accessed on 18 Jan 2021. [Online]. Available:
[66] A. A. Muley and R. B. Ghongade, “Design and simulate an antenna for aqueous glucose measurement,” in 2014 Annual
IEEE India Conference (INDICON), Dec 2014, pp. 1–6.
[67] K. U. Menon, D. Hemachandran, and A. T. Kunnath, “Voltage intensity based non-invasive blood glucose monitoring,” in
2013 Fourth International Conference on Computing, Communications and Networking Technologies (ICCCNT). IEEE,
2013, pp. 1–5.
[68] J. L. Lai, S. Y. Huang, R. S. Lin, and S. C. Tsai, “Design a non-invasive near-infrared led blood glucose sensor,” in 2016
International Conference on Applied System Innovation (ICASI), May 2016, pp. 1–4.
[69] M. Tamilselvi and G. Ramkumar, “Non-invasive tracking and monitoring glucose content using near infrared spectroscopy,
in 2015 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC), Dec 2015, pp.
[70] K. Lawand, M. Parihar, and S. N. Patil, “Design and development of infrared led based non invasive blood glucometer,
in 2015 Annual IEEE India Conference (INDICON), Dec 2015, pp. 1–6.
[71] J. Yadav, A. Rani, V. Singh, and B. M. Murari, “Near-infrared led based non-invasive blood glucose sensor,” in 2014
International Conference on Signal Processing and Integrated Networks (SPIN), Feb 2014, pp. 591–594.
[72] M. T. B. Z. Abidin, M. K. R. Rosli, S. A. Shamsuddin, N. K. Madzhi, and M. F. Abdullah, “Initial quantitative comparison
of 940nm and 950nm infrared sensor performance for measuring glucose non-invasively,” in 2013 IEEE International
Conference on Smart Instrumentation, Measurement and Applications (ICSIMA), Nov 2013, pp. 1–6.
[73] Y. Nikawa and T. Michiyama, “Non-invasive measurement of blood-sugar level by reflection of millimeter-waves,” in
2006 Asia-Pacific Microwave Conference, Dec 2006, pp. 47–50.
[74] M. Goodarzi and W. Saeys, “Selection of the most informative near infrared spectroscopy wavebands for continuous
glucose monitoring in human serum,” Talanta, vol. 146, pp. 155–165, 2016.
[75] S. Sharma, M. Goodarzi, L. Wynants, H. Ramon, and W. Saeys, “Efficient use of pure component and interferent spectra
in multivariate calibration,Analytica chimica acta, vol. 778, pp. 15–23, 2013.
[76] Y. Uwadaira, N. Adachi, A. Ikehata, and S. Kawano, “Factors affecting the accuracy of non-invasive blood glucose
measurement by short-wavelength near infrared spectroscopy in the determination of the glycaemic index of foods,
Journal of Near Infrared Spectroscopy, vol. 18, no. 5, pp. 291–300, 2010.
[77] S. Haxha and J. Jhoja, “Optical based noninvasive glucose monitoring sensor prototype,” IEEE Photonics Journal, vol. 8,
no. 6, pp. 1–11, 2016.
[78] W. Zhang, R. Liu, W. Zhang, H. Jia, and K. Xu, “Discussion on the validity of nir spectral data in non-invasive blood
glucose sensing,” Biomedical optics express, vol. 4, no. 6, pp. 789–802, 2013.
[79] M. Golic, K. Walsh, and P. Lawson, “Short-wavelength near-infrared spectra of sucrose, glucose, and fructose with respect
to sugar concentration and temperature,” Applied spectroscopy, vol. 57, no. 2, pp. 139–145, 2003.
[80] K. Song, U. Ha, S. Park, J. Bae, and H. J. Yoo, “An impedance and multi-wavelength near-infrared spectroscopy ic for
non-invasive blood glucose estimation,” IEEE Journal of Solid-State Circuits, vol. 50, no. 4, pp. 1025–1037, April 2015.
[81] S. Ramasahayam, K. S. Haindavi, and S. R. Chowdhury, “Noninvasive estimation of blood glucose concentration using
near infrared optodes,” in Sensing Technology: Current Status and Future Trends IV. Springer, 2015, pp. 67–82.
[82] A. Heller, “Integrated medical feedback systems for drug delivery,AIChE journal, vol. 51, no. 4, pp. 1054–1066, 2005.
[83] P. P. Pai, P. K. Sanki, A. De, and S. Banerjee, “NIR photoacoustic spectroscopy for non-invasive glucose measurement,
in 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). IEEE,
2015, pp. 7978–7981.
[84] P. Jain, A. M. Joshi, and S. P. Mohanty, “iGLU: An intelligent device for accurate noninvasive blood glucose-level
monitoring in smart healthcare,” IEEE Consumer Electronics Magazine, vol. 9, no. 1, pp. 35–42, 2019.
[85] P. Jain, R. Maddila, and A. M. Joshi, “A precise non-invasive blood glucose measurement system using NIR spectroscopy
and Huber’s regression model,Optical and Quantum Electronics, vol. 51, no. 2, p. 51, 2019.
[86] E. Monte-Moreno, “Non-invasive estimate of blood glucose and blood pressure from a photoplethysmograph by means
of machine learning techniques,” Artificial intelligence in medicine, vol. 53, no. 2, pp. 127–138, 2011.
[87] S. Habbu, M. Dale, and R. Ghongade, “Estimation of blood glucose by non-invasive method using photoplethysmography,”
a, vol. 44, no. 6, p. 135, 2019.
[88] H. Ali, F. Bensaali, and F. Jaber, “Novel approach to non-invasive blood glucose monitoring based on transmittance and
refraction of visible laser light,” IEEE Access, vol. 5, pp. 9163–9174, 2017.
[89] C. Vranˇ
c, A. Fomichova, N. Gretz, C. Herrmann, S. Neudecker, A. Pucci, and W. Petrich, “Continuous glucose
monitoring by means of mid-infrared transmission laser spectroscopyin vitro,” Analyst, vol. 136, pp. 1192–1198, 2011.
[Online]. Available:
[90] B. Paul, M. P. Manuel, and Z. C. Alex, “Design and development of non invasive glucose measurement system,” in 2012
1st International Symposium on Physics and Technology of Sensors (ISPTS-1). IEEE, 2012, pp. 43–46.
[91] L. A. Philip, K. Rajasekaran, and E. S. J. Jothi, “Continous monitoring of blood glucose using photophlythesmograph
signal,” in 2017 International Conference on Innovations in Electrical, Electronics, Instrumentation and Media Technology
(ICEEIMT). IEEE, 2017, pp. 187–191.
[92] H. Karimipour, H. T. Shandiz, and E. Zahedi, “Diabetic diagnose test based on ppg signal and identification system,”
Journal of Biomedical Science and Engineering, vol. 2, no. 06, p. 465, 2009.
[93] F. R. G. Cruz, C. C. Paglinawan, C. N. V. Catindig, J. C. B. Lamchek, D. D. C. Almiranez, and A. F. Sanchez, “Application
of reflectance mode photoplethysmography for non-invasive monitoring of blood glucose level with moving average filter,
in Proceedings of the 2019 9th International Conference on Biomedical Engineering and Technology. ACM, 2019, pp.
[94] Y. Zhang, Y. Zhang, S. A. Siddiqui, and A. Kos, “Non-invasive blood-glucose estimation using smartphone ppg signals
and subspace knn classifier,Elektrotehniski Vestnik, vol. 86, no. 1/2, pp. 68–74, 2019.
[95] Y. Yamakoshi, K. Matsumura, T. Yamakoshi, J. Lee, P. Rolfe, Y. Kato, K. Shimizu, and K.-i. Yamakoshi, “Side-scattered
finger-photoplethysmography: experimental investigations toward practical noninvasive measurement of blood glucose,
Journal of biomedical optics, vol. 22, no. 6, p. 067001, 2017.
[96] O. Olarte, W. V. Moer, K. Barb´
e, Y. V. Ingelgem, and A. Hubin, “Influence of the type and position of the sensor on the
precision of impedance glucose measurements,” in 2013 IEEE International Instrumentation and Measurement Technology
Conference (I2MTC), May 2013, pp. 1750–1754.
[97] S. K. Dhar, P. Biswas, and S. Chakraborty, “Dc impedance of human blood using eis: An appraoch to non-invasive blood
glucose measurement,” in 2013 International Conference on Informatics, Electronics and Vision (ICIEV), May 2013, pp.
[98] Y. Khawam, M. Ali, H. Shazada, S. Kanan, and H. Nashash, “Non-invasive blood glucose measurement using transmission
spectroscopy,” in 2013 1st International Conference on Communications, Signal Processing, and their Applications
(ICCSPA), Feb 2013, pp. 1–4.
[99] M. N. Anas and P. K. Lim, “A bio-impedance approach,” in 2013 IEEE International Conference on Smart Instrumentation,
Measurement and Applications (ICSIMA), Nov 2013, pp. 1–5.
[100] C. E. F. Amaral and B. Wolf, “Effects of glucose in blood and skin impedance spectroscopy,” in AFRICON 2007, Sept
2007, pp. 1–7.
[101] P. Jain and A. M. Joshi, “Low leakage and high cmrr cmos differential amplifier for biomedical application,Analog
Integrated Circuits and Signal Processing, pp. 1–15, 2017.
[102] B. Paul, M. P. Manuel, and Z. C. Alex, “Design and development of non invasive glucose measurement system,” in 2012
1st International Symposium on Physics and Technology of Sensors (ISPTS-1), March 2012, pp. 43–46.
[103] M. Hofmann, M. Bloss, R. Weigel, G. Fischer, and D. Kissinger, “Non-invasive glucose monitoring using open
electromagnetic waveguides,” in 2012 42nd European Microwave Conference, Oct 2012, pp. 546–549.
[104] Y. Liu, M. Xia, Z. Nie, J. Li, Y. Zeng, and L. Wang, “In vivo wearable non-invasive glucose monitoring based on dielectric
spectroscopy,” in 2016 IEEE 13th International Conference on Signal Processing (ICSP), Nov 2016, pp. 1388–1391.
[105] M. Yamaguchi, M. Mitsumori, and Y. Kano, “Noninvasively measuring blood glucose using saliva,IEEE Engineering
in Medicine and Biology Magazine, vol. 17, no. 3, pp. 59–63, May 1998.
[106] M. S. Prasad, R. Chen, Y. Li, D. Rekha, D. Li, H. Ni, and N. Y. Sreedhar, “Polypyrrole supported with copper nanoparticles
modified alkali anodized steel electrode for probing of glucose in real samples,” IEEE Sensors Journal, vol. 18, no. 13,
pp. 5203–5212, July 2018.
[107] G. Yoon, K. J. Jeon, A. K. Amerov, Y.-J. Kim, D. Y. Hwang, J. B. Kim, and H. S. Kim, “Non-invasive monitoring of
blood glucose,” in Lasers and Electro-Optics, 1999. CLEO/Pacific Rim ’99. The Pacific Rim Conference on, vol. 4, Aug
1999, pp. 1233–1234 vol.4.
[108] Y. Yamakoshi, M. Ogawa, T. Yamakoshi, M. Satoh, M. Nogawa, S. Tanaka, T. Tamura, P. Rolfe, and K. Yamakoshi, “A
new non-invasive method for measuring blood glucose using instantaneous differential near infrared spectrophotometry,”
in 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Aug 2007, pp.
[109] G. Yoon, A. K. Amerov, K. J. Jeon, J. B. Kim, and Y.-J. Kim, “Optical measurement of glucose levels in scattering
media,” in Proceedings of the 20th Annual International Conference of the IEEE Engineering in Medicine and Biology
Society. Vol.20 Biomedical Engineering Towards the Year 2000 and Beyond (Cat. No.98CH36286), vol. 4, Oct 1998, pp.
1897–1899 vol.4.
[110] H. Ishizawa, A. Muro, T. Takano, K. Honda, and H. Kanai, “Non-invasive blood glucose measurement based on atr
infrared spectroscopy,” in 2008 SICE Annual Conference, Aug 2008, pp. 321–324.
[111] T. Harada, K. Yamamoto, M. Kondo, K. Gesho, and I. Ishimaru, “Spectroscpy optical coherence tomography of biomedical
tissue,” in SICE Annual Conference 2007, Sept 2007, pp. 3056–3059.
[112] A. Popov, A. Bykov, S. Toppari, M. Kinnunen, A. Priezzhev, and R. Myllyl¨
a, “Glucose sensing in flowing blood and
intralipid by laser pulse time-of-flight and optical coherence tomography techniques,” IEEE Journal of Selected Topics
in Quantum Electronics, vol. 18, pp. 1335–1342, 07 2012.
[113] J. Y. Sim, C. G. Ahn, E. Jeong, and B. K. Kim, “Photoacoustic spectroscopy that uses a resonant characteristic of a
microphone for in vitro measurements of glucose concentration,” in 2016 38th Annual International Conference of the
IEEE Engineering in Medicine and Biology Society (EMBC), Aug 2016, pp. 4861–4864.
[114] P. P. Pai, P. K. Sanki, and S. Banerjee, “A photoacoustics based continuous non-invasive blood glucose monitoring system,”
in 2015 IEEE International Symposium on Medical Measurements and Applications (MeMeA) Proceedings, May 2015,
pp. 106–111.
[115] P. P. Pai, P. K. Sanki, A. De, and S. Banerjee, “Nir photoacoustic spectroscopy for non-invasive glucose measurement,
in 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Aug
2015, pp. 7978–7981.
[116] Y. Tanaka, Y. Higuchi, and S. Camou, “Noninvasive measurement of aqueous glucose solution at physiologically relevant
blood concentration levels with differential continuous-wave laser photoacoustic technique,” in 2015 IEEE SENSORS,
Nov 2015, pp. 1–4.
[117] L. Xiaoli and L. Chengwei, “Research on glucose concentration sensing with single wavelength laser,” in 2015 12th IEEE
International Conference on Electronic Measurement Instruments (ICEMI), vol. 03, July 2015, pp. 1547–1551.
[118] H. A. A. Naam, M. O. Idrees, A. Awad, O. S. Abdalsalam, and F. Mohamed, “Non invasive blood glucose measurement
based on photo-acoustic spectroscopy,” in 2015 International Conference on Computing, Control, Networking, Electronics
and Embedded Systems Engineering (ICCNEEE), Sept 2015, pp. 1–4.
[119] S. Camou, Y. Ueno, and E. Tamechika, “New cw-photoacoustic-based protocol for noninvasive and selective determination
of aqueous glucose level: A potential alternative towards noninvasive blood sugar sensing,” in 2011 IEEE SENSORS
Proceedings, Oct 2011, pp. 798–801.
[120] N. Wadamori, R. Shinohara, and Y. Ishihara, “Photoacoustic depth profiling of a skin model for non-invasive glucose
measurement,” in 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society,
Aug 2008, pp. 5644–5647.
[121] S. Koyama, Y. Miyauchi, T. Horiguchi, and H. Ishizawa, “Non-invasive measurement of blood glucose of diabetic based
on ir spectroscopy,” in Proceedings of SICE Annual Conference 2010, Aug 2010, pp. 3425–3426.
[122] P. Domachuk, M. Hunter, R. Batorsky, M. Cronin-Golomb, F. G. Omenetto, A. Wang, A. K. George, and J. C. Knight, “A
path for non-invasive glucose detection using mid-ir supercontinuum,” in 2008 Conference on Lasers and Electro-Optics
and 2008 Conference on Quantum Electronics and Laser Science, May 2008, pp. 1–2.
[123] R. Periyasamy and S. Anand, “A study on non-invasive blood glucose estimation- an approach using capacitance
measurement technique,” in 2016 International Conference on Signal Processing, Communication, Power and Embedded
System (SCOPES), Oct 2016, pp. 847–850.
[124] T. Yilmaz, R. Foster, and Y. Hao, “Towards accurate dielectric property retrieval of biological tissues for blood glucose
monitoring,” IEEE Transactions on Microwave Theory and Techniques, vol. 62, no. 12, pp. 3193–3204, Dec 2014.
[125] M. Gourzi, A. Rouane, M. B. McHugh, R. Guelaz, and M. Nadi, “New biosensor for non-invasive glucose concentration
measurement,” in Proceedings of IEEE Sensors 2003 (IEEE Cat. No.03CH37498), vol. 2, Oct 2003, pp. 1343–1347 Vol.2.
[126] V. Turgul and I. Kale, “On the accuracy of complex permittivity model of glucose/water solutions for non-invasive
microwave blood glucose sensing,” in 2015 E-Health and Bioengineering Conference (EHB), Nov 2015, pp. 1–4.
[127] D. Li, D. Yang, J. Yang, Y. Lin, Y. Sun, H. Yu, and K. Xu, “Glucose affinity measurement by surface plasmon resonance
with borate polymer binding,” Sensors and Actuators A: Physical, vol. 222, pp. 58 – 66, 2015. [Online]. Available:
[128] R. Kaul and U. P. Khot, “Design of microstrip antennas for glucometer application,” in 2016 IEEE International Conference
on Advances in Electronics, Communication and Computer Technology (ICAECCT), Dec 2016, pp. 352–357.
[129] V. Turgul and I. Kale, “Influence of fingerprints and finger positioning on accuracy of rf blood glucose measurement
from fingertips,” Electronics Letters, vol. 53, no. 4, pp. 218–220, 2017.
[130] H. Cano-Garcia, I. Gouzouasis, I. Sotiriou, S. Saha, G. Palikaras, P. Kosmas, and E. Kallos, “Reflection and transmission
measurements using 60 ghz patch antennas in the presence of animal tissue for non-invasive glucose sensing,” in 2016
10th European Conference on Antennas and Propagation (EuCAP), April 2016, pp. 1–3.
[131] S. Saha, I. Sotiriou, I. Gouzouasis, H. Cano-Garcia, G. Palikaras, P. Kosmas, and E. Kallos, “Evaluation of the sensitivity
of transmission measurements at millimeter waves using patch antennas for non-invasive glucose sensing,” in 2016 10th
European Conference on Antennas and Propagation (EuCAP), April 2016, pp. 1–4.
[132] V. Turgul and I. Kale, “A novel pressure sensing circuit for non-invasive rf/microwave blood glucose sensors,” in 2016
16th Mediterranean Microwave Symposium (MMS), Nov 2016, pp. 1–4.
[133] M. S. Ali, N. J. Shoumy, S. Khatun, L. M. Kamarudin, and V. Vijayasarveswari, “Non-invasive blood glucose measurement
performance analysis through uwb imaging,” in 2016 3rd International Conference on Electronic Design (ICED), Aug
2016, pp. 513–516.
[134] H. Choi, J. Nylon, S. Luzio, J. Beutler, and A. Porch, “Design of continuous non-invasive blood glucose monitoring
sensor based on a microwave split ring resonator,” in 2014 IEEE MTT-S International Microwave Workshop Series on
RF and Wireless Technologies for Biomedical and Healthcare Applications (IMWS-Bio2014), Dec 2014, pp. 1–3.
[135] V. Turgul and I. Kale, “Simulating the effects of skin thickness and fingerprints to highlight problems with non-invasive
rf blood glucose sensing from fingertips,” IEEE Sensors Journal, vol. 17, no. 22, pp. 7553–7560, Nov 2017.
[136] Y. Miyauchi, T. Horiguchi, H. Ishizawa, S. i. Tezuka, and H. Hara, “Blood glucose level measurement by confocal
reflection photodetection system,” in SICE Annual Conference 2011, Sept 2011, pp. 2686–2689.
[137] D. W. Kim, H. S. Kim, D. H. Lee, and H. C. Kim, “Importance of skin resistance in the reverse iontophoresis-based
noninvasive glucose monitoring system,” in The 26th Annual International Conference of the IEEE Engineering in
Medicine and Biology Society, vol. 1, Sept 2004, pp. 2434–2437.
[138] M. Hofmann, T. Fersch, R. Weigel, G. Fischer, and D. Kissinger, “A novel approach to non-invasive blood glucose
measurement based on rf transmission,” in 2011 IEEE International Symposium on Medical Measurements and
Applications, May 2011, pp. 39–42.
[139] K. Mitsubayashi, “Novel biosensing devices for medical applications soft contact-lens sensors for monitoring tear sugar,”
in 2014 International Conference on Simulation of Semiconductor Processes and Devices (SISPAD), Sept 2014, pp.
[140] B. D. Cameron and G. L. Cote, “Polarimetric glucose sensing in aqueous humor utilizing digital closed-loop control,” in
Proceedings of 18th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, vol. 1,
Oct 1996, pp. 204–205 vol.1.
[141] R. J. Buford, E. C. Green, and M. J. McClung, “A microwave frequency sensor for non-invasive blood-glucose
measurement,” in 2008 IEEE Sensors Applications Symposium, Feb 2008, pp. 4–7.
[142] S. Lee, V. Nayak, J. Dodds, M. Pishkou, and N. B. Smith, “Glucose measurements with sensors and ultrasound,
Ultrasound in Medicine and Biology, vol. 31, no. 7, pp. 971–977, 2005.
[143] O. K. Cho, Y. O. Kim, H. Mitsumaki, and K. Kuwa, “Noninvasive Measurement of Glucose by Metabolic
Heat Conformation Method,” Clinical Chemistry, vol. 50, no. 10, pp. 1894–1898, 10 2004. [Online]. Available:
[144] A. B. Blodgett, R. K. Kothinti, I. Kamyshko, D. H. Petering, S. Kumar, and N. M. Tabatabai, “A fluorescence method
for measurement of glucose transport in kidney cells,Diabetes Technology & Therapeutics, vol. 13, no. 7, pp. 743–751,
2011, pMID: 21510766.
[145] A. K. Amerov, Y. Sun, G. W. Small, and M. A. Arnold, “Kromoscopic measurement of glucose in the first overtone
region of the near-infrared spectrum,” in Optical Diagnostics and Sensing of Biological Fluids and Glucose and
Cholesterol Monitoring II, A. V. Priezzhev and G. L. Cote, Eds., vol. 4624, International Society for Optics and
Photonics. SPIE, 2002, pp. 11 – 19. [Online]. Available:
[146] R. Zhang, S. Liu, H. Jin, Y. Luo, Z. Zheng, F. Gao, and Y. Zheng, “Noninvasive electromagnetic wave sensing of glucose,
Sensors, vol. 19, no. 5, p. 1151, 2019.
[147] P. Bertemes-Filho, R. Weinert, T. Barato, and T. Baratto de Albuquerque, “Detection of glucose by using impedance
spectroscopy,” in Proc. International Conference on Electrical Bio-Impedance, 06 2016.
[148] R. O. Potts, J. A. Tamada, and M. J. Tierney, “Glucose monitoring by reverse iontophoresis,Diabetes/Metabolism
Research and Reviews, vol. 18, no. S1, pp. S49–S53, 2002. [Online]. Available:
[149] R. Rao and S. Nanda, “Sonophoresis: recent advancements and future trends,Journal of Pharmacy and Pharmacology,
vol. 61, no. 6, pp. 689–705, 2009. [Online]. Available:
[150] O. Amir, D. Weinstein, S. Zilberman, M. Less, D. Perl-Treves, H. Primack, A. Weinstein, E. Gabis, B. Fikhte, and
A. Karasik, “Continuous noninvasive glucose monitoring technology based on “occlusion spectroscopy”,” Journal of
Diabetes Science and Technology, vol. 1, no. 4, pp. 463–469, 2007, pMID: 19885108.
[151] B. M. Jensen, P. Bjerring, J. S. Christiansen, and H. ørskov, “Glucose content in human skin: relationship with blood
glucose levels,Scandinavian Journal of Clinical and Laboratory Investigation, vol. 55, no. 5, pp. 427–432, 1995.
[152] T. Kossowski and R. Stasi´
nski, “Multi-wavelength analysis of substances levels in human blood,” in 2017 International
Conference on Systems, Signals and Image Processing (IWSSIP), May 2017, pp. 1–4.
[153] A. Ficorella, A. D’Amico, M. Santonico, G. Pennazza, S. Grasso, and A. Zompanti, “A multi-frequency system for
glucose detection with optical sensors,” in 2015 XVIII AISEM Annual Conference, Feb 2015, pp. 1–3.
[154] K. Song, U. Ha, S. Park, and H.-J. Yoo, “An impedance and multi-wavelength near-infrared spectroscopy ic for non-
invasive blood glucose estimation,” in 2014 Symposium on VLSI Circuits Digest of Technical Papers, June 2014, pp.
[155] E. L. Litinskaia, N. A. Bazaev, K. V. Pozhar, and V. M. Grinvald, “Methods for improving accuracy of non-invasive
blood glucose detection via optical glucometer,” in 2017 IEEE Conference of Russian Young Researchers in Electrical
and Electronic Engineering (EIConRus), Feb 2017, pp. 47–49.
[156] P. Jain and S. Akashe, “An innovative design: Mos based full-wave centre-tapped rectifier,Wireless Personal Communi-
cations, vol. 90, no. 4, pp. 1673–1693, 2016.
[157] P. JAIN and S. AKASHE, “Performance analysis of analog to digital converter with augmented voltage swing boost logic
cum schmitt trigger mos switches configuration.” Journal of Active & Passive Electronic Devices, vol. 11, 2016.
[158] P. Jain and S. Akashe, “Analyzing the impact of bootstrapped adc with augmented nmos sleep transistors configuration
on performance parameters,” Circuits, Systems, and Signal Processing, vol. 33, no. 7, pp. 2009–2025, 2014.
[159] P.Jain and S. Akashe, “Design and optimization of flash type analog to digital converter using augmented sleep transistors
with current mode logic,” Radioelectronics and Communications Systems, vol. 56, no. 10, pp. 472–480, 2013.
[160] A. K. Singh and S. K. Jha, “Fabrication and validation of a handheld non-invasive, optical biosensor for self-monitoring
of glucose using saliva,IEEE Sensors Journal, vol. 19, no. 18, pp. 8332–8339, Sep. 2019.
[161] P. P. Pai, A. De, and S. Banerjee, “Accuracy enhancement for noninvasive glucose estimation using dual-wavelength
photoacoustic measurements and kernel-based calibration,” IEEE Transactions on Instrumentation and Measurement,
vol. 67, no. 1, pp. 126–136, 2018.
[162] T. Dai and A. Adler, “In vivo blood characterization from bioimpedance spectroscopy of blood pooling,IEEE Transactions
on Instrumentation and Measurement, vol. 58, no. 11, p. 3831, 2009.
[163] R. D. Beach, R. W. Conlan, M. C. Godwin, and F. Moussy, “Towards a miniature implantable in vivo telemetry
monitoring system dynamically configurable as a potentiostat or galvanostat for two- and three-electrode biosensors,
IEEE Transactions on Instrumentation and Measurement, vol. 54, no. 1, pp. 61–72, Feb 2005.
[164] P. Jain, S. Pancholi, and A. M. Joshi, “An iomt based non-invasive precise blood glucose measurement system,” in 2019
IEEE International Symposium on Smart Electronic Systems (iSES)(Formerly iNiS). IEEE, 2019, pp. 111–116.
[165] B. A. Malik, A. Naqash, and G. M. Bhat, “Backpropagation artificial neural network for determination of glucose
concentration from near-infrared spectra,” in 2016 International Conference on Advances in Computing, Communications
and Informatics (ICACCI), Sept 2016, pp. 2688–2691.
[166] D. Yotha, C. Pidthalek, S. Yimman, and S. Niramitmahapanya, “Design and construction of the hypoglycemia monito
wireless system for diabetic,” in 2016 9th Biomedical Engineering International Conference (BMEiCON), Dec 2016, pp.
[167] S. Sarangi, P. P. Pai, P. K. Sanki, and S. Banerjee, “Comparative analysis of golay code based excitation and coherent
averaging for non-invasive glucose monitoring system,” in 2014 IEEE 27th International Symposium on Computer-Based
Medical Systems, May 2014, pp. 485–486.
[168] S. R. Naqvi, N. Z. Azeemi, A. Hameed, R. Baddar, and T. Rasool, “Improving accuracy of non-invasive glucose monitoring
through non-local data denoising,” in 2008 Cairo International Biomedical Engineering Conference, Dec 2008, pp. 1–4.
[169] Y. Yamakoshi, M. Ogawa, T. Yamakoshi, T. Tamura, and K. i. Yamakoshi, “Multivariate regression and discreminant
calibration models for a novel optical non-invasive blood glucose measurement method named pulse glucometry,” in
2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Sept 2009, pp. 126–
[170] C. Z. Ming, P. Raveendran, and P. S. Chew, “A comparison analysis between partial least squares and neural network
in non-invasive blood glucose concentration monitoring system,” in 2009 International Conference on Biomedical and
Pharmaceutical Engineering, Dec 2009, pp. 1–4.
[171] S. Sarangi, P. P. Pai, P. K. Sanki, and S. Banerjee, “Comparative analysis of golay code based excitation and coherent
averaging for non-invasive glucose monitoring system,” in 2014 IEEE 27th International Symposium on Computer-Based
Medical Systems, May 2014, pp. 485–486.
[172] H. M. Heise, “Technology for non-invasive monitoring of glucose,” in Proceedings of 18th Annual International
Conference of the IEEE Engineering in Medicine and Biology Society, vol. 5, Oct 1996, pp. 2159–2161 vol.5.
[173] P. P. Pai, S. Bhattacharya, and S. Banerjee, “Regularized least squares regression for calibration of a photoacoustic
spectroscopy based non-invasive glucose monitoring system,” in 2015 IEEE International Ultrasonics Symposium (IUS),
Oct 2015, pp. 1–4.
[174] M. Stemmann, F. St˚
ahl, J. Lallemand, E. Renard, and R. Johansson, “Sensor calibration models for a non-invasive blood
glucose measurement sensor,” in 2010 Annual International Conference of the IEEE Engineering in Medicine and Biology,
Aug 2010, pp. 4979–4982.
[175] D. K. Rollins, K. Kotz, and C. Stiehl, “Non-invasive glucose monitoring from measured inputs,” in UKACC International
Conference on Control 2010, Sept 2010, pp. 1–5.
[176] B. A. Malik, “Determination of glucose concentration from near infrared spectra using least square support vector
machine,” in 2015 International Conference on Industrial Instrumentation and Control (ICIC), May 2015, pp. 475–478.
[177] M. Ogawa, Y. Yamakoshi, M. Satoh, M. Nogawa, T. Yamakoshi, S. Tanaka, P. Rolfe, T. Tamura, and K. i. Yamakoshi,
“Support vector machines as multivariate calibration model for prediction of blood glucose concentration using a new non-
invasive optical method named pulse glucometry,” in 2007 29th Annual International Conference of the IEEE Engineering
in Medicine and Biology Society, Aug 2007, pp. 4561–4563.
[178] Z. T. Dag, E. Koklukaya, F. Temurtas, H. M. Saraoglu, and S. Altikat, “Detection of the blood glucose and haemoglobin
a1c with palm perspiration by using artificial neural networks,” in 2011 7th International Conference on Electrical and
Electronics Engineering (ELECO), Dec 2011, pp. II–302–II–305.
[179] O. Olarte, W. V. Moer, K. Barb´
e, Y. V. Ingelgem, and A. Hubin, “Using random phase multisines to perform non-invasive
glucose measurements,” in 2011 IEEE International Symposium on Medical Measurements and Applications, May 2011,
pp. 300–304.
[180] M. B. Savage, S. Kun, H. Harjunmaa, and R. A. Peura, “Development of a non-invasive blood glucose monitor: application
of artificial neural networks for signal processing,” in Proceedings of the IEEE 26th Annual Northeast Bioengineering
Conference (Cat. No.00CH37114), 2000, pp. 29–30.
[181] R. Baghbani, M. A. Rad, and A. Pourziad, “Microwave sensor for non-invasive glucose measurements design and
implementation of a novel linear,IET Wireless Sensor Systems, vol. 5, no. 2, pp. 51–57, 2015.
[182] J. S. Parab, R. S. Gad, and G. M. Naik, “Influence of pca components on glucose prediction using non-invasive technique,”
in 2016 International Conference on Advances in Electrical, Electronic and Systems Engineering (ICAEES), Nov 2016,
pp. 473–476.
[183] T. R. J. C. Lekha and C. S. Kumar, “Nir spectroscopic algorithm development for glucose detection,” in 2015 International
Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS), March 2015, pp. 1–6.
[184] O. Olarte, W. V. Moer, K. Barb´
e, S. Verguts, Y. V. Ingelgem, and A. Hubin, “Using the best linear approximation as
a first step to a new non-invasive glucose measurement,” in 2012 IEEE International Instrumentation and Measurement
Technology Conference Proceedings, May 2012, pp. 2747–2751.
[185] W. L. Clarke, “The original Clarke error grid analysis (EGA),” Diabetes Technology & Therapeutics, vol. 7, no. 5, pp.
776–779, 2005.
[186] C. Fernandez, “Needle-free diabetes care: 7 devices that painlessly measure blood glucose,Labiotech, vol. 23, 2018.
[187] M. F. Schemmann and T. O’brien, “Blood glucose sensor,” Apr. 11 2013, uS Patent App. 13/646,721.
[188] A. Gal, I. Harman-Boehm, E. Naidis, Y. Mayzel, and L. Trieman, “Validity of glucotrack®, a non-invasive glucose
monitor, for variety of diabetics,Age, vol. 1, no. 295, p. 61, 2011.
[189] D. Huber, L. Falco-Jonasson, M. Talary, F. Dewarrat, A. Caduff, W. Stahel, and N. Stadler, “Multi-sensor data fusion for
non-invasive continuous glucose monitoring,” in 2007 10th International Conference on Information Fusion, July 2007,
pp. 1–10.
[190] V. W. Bolie, “Coefficients of normal blood glucose regulation,Journal of applied physiology, vol. 16, no. 5, pp. 783–788,
[191] R. N. Bergman, L. S. Phillips, and C. Cobelli, “Physiologic evaluation of factors controlling glucose tolerance in man:
measurement of insulin sensitivity and beta-cell glucose sensitivity from the response to intravenous glucose.” The Journal
of clinical investigation, vol. 68, no. 6, pp. 1456–1467, 1981.
[192] A. De Gaetano and O. Arino, “Mathematical modelling of the intravenous glucose tolerance test,Journal of mathematical
biology, vol. 40, no. 2, pp. 136–168, 2000.
[193] C. Cobelli, C. Dalla Man, G. Sparacino, L. Magni, G. De Nicolao, and B. P. Kovatchev, “Diabetes: models, signals, and
control,” IEEE reviews in biomedical engineering, vol. 2, pp. 54–96, 2009.
[194] R. Hovorka, F. Shojaee-Moradie, P. V. Carroll, L. J. Chassin, I. J. Gowrie, N. C. Jackson, R. S. Tudor, A. M. Umpleby,
and R. H. Jones, “Partitioning glucose distribution/transport, disposal, and endogenous production during ivgtt,American
Journal of Physiology-Endocrinology and Metabolism, vol. 282, no. 5, pp. E992–E1007, 2002.
[195] A. Haidar, M. E. Wilinska, J. A. Graveston, and R. Hovorka, “Stochastic virtual population of subjects with type 1
diabetes for the assessment of closed-loop glucose controllers,” IEEE Transactions on Biomedical Engineering, vol. 60,
no. 12, pp. 3524–3533, 2013.
[196] H. Kirchsteiger, G. C. Estrada, S. P¨
olzer, E. Renard, and L. del Re, “Estimating interval process models for type 1
diabetes for robust control design,IFAC Proc. Volumes, vol. 44, no. 1, pp. 11 761–11 766, 2011.
[197] N. Magdelaine, L. Chaillous, I. Guilhem, J.-Y. Poirier, M. Krempf, C. H. Moog, and E. Le Carpentier, “A long-term
model of the glucose–insulin dynamics of type 1 diabetes,” IEEE Transactions on Biomedical Engineering, vol. 62, no. 6,
pp. 1546–1552, 2015.
[198] K. Turksoy, S. Samadi, J. Feng, E. Littlejohn, L. Quinn, and A. Cinar, “Meal detection in patients with type 1 diabetes:
a new module for the multivariable adaptive artificial pancreas control system,” IEEE journal of biomedical and health
informatics, vol. 20, no. 1, pp. 47–54, 2015.
[199] J. Xie and Q. Wang, “A variable state dimension approach to meal detection and meal size estimation: in silico evaluation
through basal-bolus insulin therapy for type 1 diabetes,” IEEE Transactions on Biomedical Engineering, vol. 64, no. 6,
pp. 1249–1260, 2016.
[200] T. MohammadRidha, M. A¨
ıt-Ahmed, L. Chaillous, M. Krempf, I. Guilhem, J.-Y. Poirier, and C. H. Moog, “Model free
ipid control for glycemia regulation of type-1 diabetes,IEEE Transactions on Biomedical Engineering, vol. 65, no. 1,
pp. 199–206, 2017.
[201] S. Kino, S. Omori, T. Katagiri, and Y. Matsuura, “Hollow optical-fiber based infrared spectroscopy for measurement of
blood glucose level by using multi-reflection prism,Biomedical optics express, vol. 7, no. 2, pp. 701–708, 2016.
[202] T.-L. Chen, Y.-L. Lo, C.-C. Liao, and Q.-H. Phan, “Noninvasive measurement of glucose concentration on human fingertip
by optical coherence tomography,Journal of biomedical optics, vol. 23, no. 4, p. 047001, 2018.
[203] J. Park, J. Kim, S.-Y. Kim, W. H. Cheong, J. Jang, Y.-G. Park, K. Na, Y.-T. Kim, J. H. Heo, C. Y. Lee et al., “Soft, smart
contact lenses with integrations of wireless circuits, glucose sensors, and displays,Science Advances, vol. 4, no. 1, p.
eaap9841, 2018.
[204] A. K. Singh and S. K. Jha, “Fabrication and validation of a handheld non-invasive, optical biosensor for self-monitoring
of glucose using saliva,IEEE Sensors Journal, vol. 19, no. 18, pp. 8332–8339, 2019.
[205] K. Song, U. Ha, S. Park, J. Bae, and H.-J. Yoo, “An impedance and multi-wavelength near-infrared spectroscopy ic for
non-invasive blood glucose estimation,” IEEE Journal of solid-state circuits, vol. 50, no. 4, pp. 1025–1037, 2015.
[206] Q. Li, X. Xiao, and T. Kikkawa, “Absorption spectrum for non-invasive blood glucose concentration detection by
microwave signals,Journal of Electromagnetic Waves and Applications, vol. 33, no. 9, pp. 1093–1106, 2019.
[207] V. P. Rachim and W.-Y. Chung, “Wearable-band type visible-near infrared optical biosensor for non-invasive blood glucose
monitoring,” Sensors and Actuators B: Chemical, vol. 286, pp. 173–180, 2019.
[208] S. P. Mohanty, U. Choppali, and E. Kougianos, “Everything you wanted to know about smart cities: The Internet of
things is the backbone,” IEEE Consumer Electronics Magazine, vol. 5, no. 3, pp. 60–70, July 2016.
[209] P. Chanak and I. Banerjee, “Internet of things-enabled smart villages: Recent advances and challenges,IEEE Consumer
Electronics Magazine, pp. 1–1, 2020.
[210] C. P. Antonopoulos, G. Keramidas, N. S. Voros, M. Huebner, F. Schwiegelshohn, D. Goehringer, M. Dagioglou,
G. Stavrinos, S. Konstantopoulos, and V. Karkaletsis, “Toward an ict-based service oriented health care paradigm,” IEEE
Consumer Electronics Magazine, vol. 9, no. 4, pp. 77–82, 2020.
[211] M. Aazam, S. Zeadally, and K. A. Harras, “Health fog for smart healthcare,” IEEE Consumer Electronics Magazine,
vol. 9, no. 2, pp. 96–102, 2020.
[212] A. M. Joshi, P. Jain, and S. P. Mohanty, “Secure-iGLU: A secure device for noninvasive glucose measurement and
automatic insulin delivery in iomt framework,” in 2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI).
IEEE, 2020, pp. 440–445.
[213] L. Rachakonda, S. P. Mohanty, and E. Kougianos, “iLog: An intelligent device for automatic food intake monitoring and
stress detection in the iomt,” IEEE Transactions on Consumer Electronics, vol. 66, no. 2, pp. 115–124, 2020.
[214] V. P. Yanambaka, S. P. Mohanty, E. Kougianos, and D. Puthal, “PMsec: Physical unclonable function-based robust and
lightweight authentication in the internet of medical things,” IEEE Transactions on Consumer Electronics, vol. 65, no. 3,
pp. 388–397, 2019.
[215] L. Rachakonda, A. K. Bapatla, S. P. Mohanty, and E. Kougianos, “SaYoPillow: Blockchain-integrated privacy-assured
iomt framework for stress management considering sleeping habits,IEEE Transactions on Consumer Electronics, pp.
1–1, 2020.
[216] S. P. Mohanty, V. P. Yanambaka, E. Kougianos, and D. Puthal, “PUFchain: A hardware-assisted blockchain for sustainable
simultaneous device and data security in the internet of everything (IoE),IEEE Consumer Electronics Magazine, vol. 9,
no. 2, pp. 8–16, 2020.
Amit M. Joshi (Member, IEEE) received the Ph.D. degree from the NIT, Surat, India. He is currently an
Assistant Professor in Department of ECE, MNIT, Jaipur. He is an author of 15 peer-reviewed publications.
His current research interests include biomedical signal processing, VLSI DSP Systems and embedded
system design. He is a regular reviewer of 10 journals and 30 conferences. He has advised 06 Ph.D. and
15 Masters thesis.
Prateek Jain (Member, IEEE) earned his B.E. degree in Electronics Engineering from Jiwaji University,
India in 2010 and Master degree from ITM University Gwalior. Currently, he is an Assistant Professor in
SENSE, VIT University, Amaravati (A.P.). His current research interest includes VLSI design, Biomedical
Systems and Instrumentation. He is an author of 14 peer-reviewed publications. He is a regular reviewer
of 12 journals and 10 conferences.
Saraju P. Mohanty (Senior Member, IEEE) received the bachelor’s degree (Honors) in electrical engi-
neering from the Orissa University of Agriculture and Technology, Bhubaneswar, in 1995, the master’s
degree in Systems Science and Automation from the Indian Institute of Science, Bengaluru, in 1999, and
the Ph.D. degree in Computer Science and Engineering from the University of South Florida, Tampa, in
2003. He is a Professor with the University of North Texas. His research is in “Smart Electronic Systems”
which has been funded by National Science Foundations (NSF), Semiconductor Research Corporation
(SRC), U.S. Air Force, IUSSTF, and Mission Innovation. He has authored 350 research articles, 4 books,
and invented 4 granted and 1 pending patents. His Google Scholar h-index is 39 and i10-index is 149 with
6600 citations. He is regarded as a visionary researcher on Smart Cities technology in which his research
deals with security and energy aware, and AI/ML-integrated smart components. He introduced the Secure Digital Camera (SDC)
in 2004 with built-in security features designed using Hardware-Assisted Security (HAS) or Security by Design (SbD) principle.
He is widely credited as the designer for the first digital watermarking chip in 2004 and first the low-power digital watermarking
chip in 2006. He is a recipient of 12 best paper awards, Fulbright Specialist Award in 2020, IEEE Consumer Technology Society
Outstanding Service Award in 2020, the IEEE-CS-TCVLSI Distinguished Leadership Award in 2018, and the PROSE Award for
Best Textbook in Physical Sciences and Mathematics category in 2016. He has delivered 10 keynotes and served on 11 panels
at various International Conferences. He has been serving on the editorial board of several peer-reviewed international journals,
including IEEE Transactions on Consumer Electronics (TCE), and IEEE Transactions on Big Data (TBD). He is the Editor-in-
Chief (EiC) of the IEEE Consumer Electronics Magazine (MCE). He has been serving on the Board of Governors (BoG) of
the IEEE Consumer Technology Society, and has served as the Chair of Technical Committee on Very Large Scale Integration
(TCVLSI), IEEE Computer Society (IEEE-CS) during 2014-2018. He is the founding steering committee chair for the IEEE
International Symposium on Smart Electronic Systems (iSES), steering committee vice-chair of the IEEE-CS Symposium on
VLSI (ISVLSI), and steering committee vice-chair of the OITS International Conference on Information Technology (ICIT). He
has mentored 2 post-doctoral researchers, and supervised 12 Ph.D. dissertations, 26 M.S. theses, and 12 undergraduate projects.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Considering today’s lifestyle, people just sleep forgetting the benefits sleep provides to the human body. Smart-Yoga Pillow (SaYoPillow) is proposed to help in understanding the relationship between stress and sleep and to fully materialize the idea of “Smart-Sleeping” by proposing an edge device. An edge processor with a model analyzing the physiological changes that occur during sleep along with the sleeping habits is proposed. Based on these changes during sleep, stress prediction for the following day is proposed. The secure transfer of the analyzed stress data along with the average physiological changes to the IoT cloud for storage is implemented. A secure transfer of any data from the cloud to any third party applications is also proposed. A user interface is provided allowing the user to control the data accessibility and visibility. SaYoPillow is novel, with security features as well as consideration of sleeping habits for stress reduction, with an accuracy of up to 96%.
Full-text available
The diabetic patients are at higher risk from novel coronavirus disease 2019 (COVID-19) that spreads through Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). There are around 20% to 50% of COVID-19 cases had diabetes across the different regions from the world. This article discusses recommendations and associated risk for diabetic patients to balance glyncemic profile during COVID-19 outbreak. It also discusses the case study of various countries with impact of COVID-19 for diabetic patients. It presents emerging smart healthcare that can potentially safeguard against COVID-19.
Full-text available
People around the globe rely on their blood samples for their glucose level measurement. There is a demand for non-invasive, precise and cost-effective solutions to monitor blood glucose level and control of diabetes. Serum glucose is an accurate blood glucose measurement method in comparison to capillary glucose measurement. Presently, the serum glucose is measured through laboratory setup with an invasive approach. The invasive method is painful and is not suitable for continuous glucose measurement. In this paper, we propose a novel wearable non-invasive consumer device (called iGLU 2.0) which can be used by consumers for accurate continuous blood glucose monitoring. This device uses a novel short near infrared (NIR) spectroscopy developed by us. It is incorporated with Internet-of-Medical-Things (IoMT) for smart healthcare where the healthcare data is stored on the cloud and is accessible to the users and caregivers. Analysis of the optimized regression model is performed and the system is calibrated and validated through healthy, prediabetic and diabetic patients. The robust regression models of serum glucose level is then deployed as the mechanism for precise measurement in iGLU 2.0. The performance of iGLU 2.0 is validated with the prediction of capillary blood glucose using Average Error (AvgE) and Mean Absolute Relative Difference (mARD) which are calculated as 6.09% and 6.07%, respectively, whereas for serum glucose, AvgE and mARD are estimated as 4.88% and 4.86%, respectively.
Full-text available
This article presents the first-ever blockchain that can simultaneously handle device and data security, which is important for the emerging Internet-of-Everything (IoE). It presents a unique concept of blockchain that integrates hardware security primitives called physical unclonable functions (PUFs) to solve scalability, latency, and energy requirement challenges and is called PUFchain. This article also introduces a new consensus algorithm called “Proof of PUF-Enabled Authentication” (PoP) for deployment in PUFchain. PoP is 1000 times faster than the well-established proof-of-work (PoW) and 5 times faster than proof-of-authentication (PoAh).
Full-text available
Annual deaths in the U.S. attributed to diabetes are expected to increase from 280,210 in 2015 to 385,840 in 2030. The increase in the number of people affected by diabetes has made it one of the major public health challenges around the world. Better management of diabetes has the potential to decrease yearly medical costs and deaths associated with the disease. Non-invasive methods are in high demand to take the place of the traditional finger prick method as they can facilitate continuous glucose monitoring. Research groups have been trying for decades to develop functional commercial non-invasive glucose measurement devices. The challenges associated with non-invasive glucose monitoring are the many factors that contribute to inaccurate readings. We identify and address the experimental and physiological challenges and provide recommendations to pave the way for a systematic pathway to a solution. We have reviewed and categorized non-invasive glucose measurement methods based on: (1) the intrinsic properties of glucose, (2) blood/tissue properties and (3) breath acetone analysis. This approach highlights potential critical commonalities among the challenges that act as barriers to future progress. The focus here is on the pertinent physiological aspects, remaining challenges, recent advancements and the sensors that have reached acceptable clinical accuracy.
This paper proposes a blood glucose concentration detection method, which is non-invasive and based on the microwave detection technique. This method is harmless to the patients’ body. And the procedure of this method is fast and convenient. The feasibility of this method is verified by obtaining the relationship between the blood glucose level and the absorption spectrum features of the received signals with the range of blood glucose level 0–500 mg/dl. The validity of this method is testified by establishing a simulation model to test the feasibility of this method. The robustness is also checked by achieving the relationship between the peaks of the received signals at 500 MHz in the experiment earlobe phantom with a group of the human blood glucose level ranges from 0-500 mg/dl with the step of 20 mg/dl.