NON-INVASIVE GLUCOMETER USING

INFRARED SPECTROSCOPY

BACHELOR OF TECHNOLOGY
In
ELECTRONICS AND COMMUNICATION ENGINEERING
A Real-Time Project Report Submitted By

D.AKSHITHA (23011A0411)
G.LAXMI KRISHNA PRIYA (23011A0426)
K.HARI VIGNESH (23011A0441)
Y.SRI DIVYA (23011A0451)
V.ASHRITHA(24015A0405)

Under the esteemed guidance of

DR.A.RAJANI

Professor in ECE Department

Department Of Electronics and Communication Engineering Jawaharlal

Nehru Technological University Hyderabad University College Of
Engineering , Science & Technology Hyderabad .
(Kukatpally -Hyderabad-500085)
2024-2025
 Department Of Electronics and Communication Engineering Jawaharlal
Nehru Technological University Hyderabad University College Of
Engineering , Science & Technology Hyderabad .
(Kukatpally -Hyderabad-500085)

CERTIFICATE BY THE SUPERVISOR

This is to certify that the real-time project entitled “NON-INVASIVE GLUCOMETER

USING INFRARED SPECTROSCOPY” being submitted by D.Akshitha
(23011A0411), G.Laxmi Krishna Priya (23011A0426), K.Hari Vignesh (23011A0441),
Y.Sri Divya (23011A0451), V.Ashritha (24015A0405) In partial fulfillment of the
requirements for the award of degree in bachelor of technology in Electronics And
Communication Engineering from Jawaharlal Nehru Technological University during
the academic year, 2024-2025 is a bonafide work carried out under my guidance and
supervision. the results have been verified and found to be satisfied.

Supervisor
Dr. A. Rajani
Professor
JNTUH UCESTH
Kukatpally Hyderabad
 Department Of Electronics and Communication Engineering Jawaharlal
Nehru Technological University Hyderabad University College Of
Engineering , Science & Technology Hyderabad .
(Kukatpally -Hyderabad-500085)

CERTIFICATE BY THE HEAD OF THE DEPARTMENT

This is to certify that the real-time project entitled “NON-INVASIVE GLUCOMETER

USING INFRARED SPECTROSCOPY” being submitted by D.Akshitha
(23011A0411), G.Laxmi Krishna Priya (23011A0426), K.Hari Vignesh (23011A0441),
Y.Sri Divya (23011A0451), V.Ashritha(24015A0405) in partial fulfillment of the
requirements for the award of degree in Bachelor of Technology in Electronics and
Communication Engineering from Jawaharlal Nehru Technological University during
the academic year, 2024-2025 is a bonafide work carried out by them.

Supervisor
Dr. T. Madhavi Kumari
Associate Professor & Head
JNTUH UCESTH
Kukatpally Hyderabad
 Department Of Electronics and Communication Engineering Jawaharlal
Nehru Technological University Hyderabad University College Of
Engineering , Science & Technology Hyderabad .
(Kukatpally -Hyderabad-500085)

DECLARATION OF THE CANDIDATES

We hereby declare The Real-Time Project entitled “NON-INVASIVE GLUCOMETER

USING INFRARED SPECTROSCOPY” is a bonafide record work done and submitted
under the esteemed guidance of Dr. A. RAJANI, Professor, Department of ECE, JNTUH,
in partial fulfillment of the requirements for Real-Time project in Electronics and
Communication Engineering at the Jawaharlal Nehru Technological University during the
academic year 2024-2025 is a bonafide work carried out by us and the results kept in the
mini project has not been reproduced. The results have not been submitted to any other
institute or university for the award of a degree or diploma.

D.AKSHITHA (23011A0411)
G.LAXMI KRISHNA PRIYA (23011A0426)
K.HARI VIGNESH (23011A0441)
Y.SRI DIVYA (23011A0451)
V.ASHRITHA(24015A0405)
 ACKNOWLEDGEMENT

This project titled “NON-INVASIVE GLUCOMETER USING INFRARED

SPECTROSCOPY” was carried out by us. We are grateful to Prof. Dr. A. Rajani,
Professor of the Department of Electronics and Communication Engineering, JNTU
Hyderabad University College of Engineering, Science and Technology Hyderabad, for
their guidance while pursuing this project.

We take this precious opportunity to acknowledge our internal project guide Dr. A. Rajani,
Professor of Electronics and Communication Engineering, JNTUH University College of
Engineering, Science and Technology Hyderabad for her timely advice, effective
guidance, and encouragement throughout the completion of our real-time project work.

We also owe a deep respect of gratitude to our parents and friends for their cheerful
encouragement and valuable suggestions, without whom this work would not have been
completed in this stipulated time.

We would like to articulate our heartfelt gratitude to the authorities of JNTUH for their
help throughout our project work. A few lines of acknowledgment do not fully express our
gratitude and appreciation for all those who guided and supported us throughout this
project. Lastly, we acknowledge the help received from many journals and websites.
Finally, we thank one and all who helped us directly or indirectly throughout our project
work.

D.AKSHITHA (23011A0411)
G.LAXMI KRISHNA PRIYA (23011A0426)
K.HARI VIGNESH (23011A0441)
Y.SRI DIVYA (23011A0451)
V.ASHRITHA(24015A0405)
 INDEX

CONTENTS PAGE
NO
1
LIST OF FIGURES
1
LIST OF TABLES
2
LIST OF ABBREVIATIONS
3
ABSTRACT
4-5
CHAPTER-1
4
Intoduction
5
Aim
5
Objectives
5
Applications
6-9
CHAPTER-2
6-9
Components
10-13
CHAPTER-3
10
System Functionality and Workflow
10-12
Working of MAX30100
13
Machine Learning Integration-Regression Model
14-21
CHAPTER-4
14-21
Implementation and Results
22
CHAPTER-5
22
Conclusion and Future Scope
23
REFERENCES
 LIST OF FIGURES

FIGURE CAPTION PAGE

NO

1.1 Invasive Glucometer 4

Apparatus

2.1 Arduino Uno 6

2.2 I2C LCD Display 7

2.3 MAX-30100 Sensor 8

2.4 Jumper Wires 9

3.1 MAX-30100 Sensor Functional 11

Diagram and PPG Diagram

4.1 Hardware Connections 15

4.2 Arduino Connections 15

4.3 Output 20

LIST OF TABLES

TABLE TITLE PAGE

NO

2.1 Pin Description 8

4.1 Samples 16

4.2 Results 21

1
 LIST OF ABBREVIATIONS

NIR Near Infrared

NIRS Near Infrared Spectroscopy

SpO2 Peripheral Capillary Oxygen Saturation

BPM Beats Per Minute

PPG Photoplethysmograph

AFE Analog Front-End

ADC Analog-to-Digital Converter

I2C Inter-Integrated Circuit

LCD Liquid Crystal Display

IDE Integrated Development Environment

ICSP In-Circuit Serial Programming

ML Machine Learning

CSV Comma-Seperated Values

WiFi Wireless Fidility

IoT Internet of Things

2
 ABSTRACT

This project aims to develop a non-invasive and low-cost glucometer using Arduino and
Bluetooth technology for continuous blood glucose monitoring. Traditional glucometers
require finger pricking to measure blood glucose levels, which can be painful and
inconvenient for users. Our solution leverages non-invasive techniques, such as optical
sensors or near- infrared technology, to detect blood glucose levels through the skin
without puncturing it.

The system operates on the principle of infrared absorption spectroscopy, where glucose
molecules absorb specific wavelengths of infrared light. An infrared LED emits light that
penetrates through the skin and interacts with glucose molecules in blood vessels. The
amount of light absorbed or scattered depends on the glucose concentration.

For glucose level measurement, Beer-Lambert’s Law states that the intensity of light
detected after passing through blood is inversely proportional to the glucose concentration,
meaning higher glucose levels absorb more light, reducing the detected intensity.

The MAX30100 is a low-power, high-precision integrated biosensor module. It features

dual LEDs (red: 660 nm, infrared: 880 nm) and a high-sensitivity photodiode, enabling
accurate measurements using reflective photoplethysmography (PPG). An Arduino
microcontroller is used to process the absorption pattern and apply a precalibrated
algorithm to estimate the glucose level.The Output of BPM and SpO2 levels are shown on
the LCD Display.

3
 CHAPTER-1
INTRODUCTION
Diabetes is a type a metabolic disease in which the blood glucose (blood sugar) level in
human body increases drastically from its normal level. Diabetes is a state of a body where
it is not able to produce the quantity of insulin sufficiently required to maintain normal
level of blood glucose. Diabetes can lead to major complications like heart failure and
blindness in the human body. Hence regular monitoring of glucose level is important.
Thus, diabetic patients reg up their blood glucose levels through proper diet as well as by
injecting insulin. For the effective treatment of diabetes, patients have to measure the level
of blood glucose periodically.

Traditional glucose monitoring involves the use of glucometers which require pricking
skin to draw blood. While accurate, this method is invasive, can be painful, and may
discourage patients from performing frequent tests. This project addresses the need for a
non- invasive, affordable, and user-friendly method of glucose monitoring by leveraging
biometric signals obtained from the GY-MAX30100 sensor.
The system uses SpO2 (blood oxygen saturation) and BPM (heart rate) to estimate glucose
levels via polynomial regression model. This machine learning model is deployed onto a
microcontroller (Arduino UNO), making the entire solution affordable and real-time.
Although it is not a replacement For Medically approved devices, it holds promise for
preliminary and frequent monitoring.

4
AIM
To design and develop a non-invasive, and affordable glucometer using Arduino and GY-
MAX30100 sensor for continuous blood glucose monitoring without the need for blood
samples.

OBJECTIVES
To develop an non-invasive glucose sensing system using GY-MAX30100 sensor.
To measure SpO2 and BPM in real-time using Arduino.
To collect and use this data to train a machine learning model for real-time glucose
prediction.
To demonstrate a cost-effective, non-invasive method for continuous glucose monitoring.

APPLICATIONS
Diabetes Monitoring: Regular glucose tracking without pricking, making it painless for
diabetic patients.
Preventive Healthcare: Early detection of high or low glucose levels helps in preventing
diabetes complications.
Remote Patient Monitoring: Enables continuous glucose tracking from home, useful for
telemedicine and elderly care.
Fitness and Lifestyle Management: Athletes and health-conscious individuals can
monitor sugar levels for optimized diet and performance.
Research and Clinical Trials: Allows data collection without invasive methods,
improving subject comfort and compliance.

5
 CHAPTER-2
HARDWARE COMPONENTS

ARDUINO UNO
Arduino UNO is a microcontroller board based on the ATmega328P. It has 14 digital
input/output pins (of which 6 can be used as PWM outputs), 6 analog inputs, a 16 MHz
ceramic resonator, a USB connection, a power jack, an ICSP header and a reset button.
The software used for Arduino devices is called IDE (Integrated Development
Environment) and it can be programmed using C and C++ language.
The Arduino Uno plays a central role in a non-invasive glucometer project. It acts as the
brain of the system, handling all the communication between the sensors, display, and
machine learning model.

Figure-2.1
LCD WITH I2C INTERFACE
A typical I2C LCD display consists of two main parts: an HD44780-based character LCD
display and an I2C LCD adapter. 16X2 Liquid Crystal Display (LCD) with I2C interface
is an alphanumeric display that can show up to 32 characters on a single screen. it can
display 16 characters per line and there are 2 such lines. In this LCD each character is
displayed in 5×7 pixel matrix. This LCD has two registers, namely, Command and
Data.The key component of I2C adapter is an 8-bit I/O expander chip called PCF8574.
This clever chip converts the I2C data from your Arduino into the parallel data that the
LCD display needs to function.
6
The I2C address of your LCD depends on which company manufactured the chip. The
16x2 LCD with I2C module plays a crucial role in your non-invasive glucometer
project by displaying the outputs like SpO2, BPM, and predicted glucose levels. The I2C
interface makes it simpler to connect and control the LCD using only 2 pins from Arduino
(SDA & SCL).

Figure-2.2

GY-MAX30100
The MAX30100 is an integrated pulse oximetry and heart rate monitor sensor solution. It
combines two LEDs, a photodetector, optimized optics, and low-noise analog signal
processing to detect pulse oximetry and heart-rate signals.
Oxygenated blood absorbs more infrared light and passes more red light while
deoxygenated blood absorbs red light and passes more infrared light. This is the main
function of the of this sensor, it reads the absorption levels for both light sources and
stored them in a buffer that can be read via I2C.

The MAX30100 also includes a built-in temperature sensor. This sensor helps improve the
accuracy of heart rate and blood oxygen readings by adjusting for changes in surrounding
temperature.

7
 Figure-2.3
Pin Description

Table 2.1
8
JUMPER WIRES
A jump wire (also known as jumper, jumper wire, DuPont wire) is an electrical wire, or
group of them in a cable, with a connector or pin at each end (or sometimes without them
– simply "tinned"), which is normally used to interconnect the components of a
breadboard or other prototype or test circuit, internally or with other equipment or
components, without soldering.

Figure-2.4

9
 CHAPTER-3
SYSTEM FUNCTIONALITY AND WORK-FLOW

SYSTEM ARCHITECTURE
Near-infrared spectroscopy (NIRS) is a spectroscopic method that uses the near-infrared
region of the electromagnetic spectrum (from 780 nm to 2500 nm). Glucose has several
absorption peaks in the NIR region. where light possesses its maximum penetration depth
in tissue is referred as Near Infrared window. Glucose has light absorption peaks at
wavelengths of 940 nm, 970 nm, 1197 nm, 1408nm, 1536nm, 1688nm, 1925 nm, 2100nm,
2261nm and 2326nm, but at 940 nm wavelength the attenuation of optical signals by other
constituents of the blood like water, platelets, red blood cells etc. is minimum, hence a
desired depth of penetration can be achieved, and actual glucose concentration can be
predicted.

WORKING OF MAX30100
The signal flow of the MAX30100 sensor begins when the LED driver activates the Red
and Infrared (IR) LEDs, allowing light to pass through the skin.
The blood absorbs and reflects light depending on oxygen levels.
A photodetector then captures the reflected light and generates a PPG waveform and
converts it into an analog electrical signal.
This signal is processed by the Analog Front-End (AFE), which amplifies the signal,
filters out noise, and cancels ambient light interference.
The cleaned analog signal is then converted to digital format by the ADC and stored in
data registers (32 FIFO), where each register holds both Red and IR values using 3 bytes
(24 bits).
The Digital Processing Unit (DPU) further processes the data and removes artifacts.
SpO2 value is determined by the ratio of absorption of Red and IR light, based on Beer-
Lambert’s law while BPM value is determined by the time between peaks in IR PPG signal.

10
 Figure-3.1

Time(seconds)

Figure-3.2

The image shows a Photoplethysmograph (PPG) waveform, which represents changes in

blood volume in tissue as measured by light absorption.
The red signal corresponds to the RED LED (660 nm wavelength), which is more
sensitive to deoxygenated hemoglobin.
The purple/blue signal corresponds to the IR LED (typically 940 nm), which is more
sensitive to oxygenated hemoglobin.
Absorption of Red and IR light is obtained using the following law,

11
Beer-Lambert’s Law:
A=εcl
Where,
A = Absorbance
ε = Molar absorptivity (L mol−1 cm−1)
c = Concentration of the solution (mol L−1)
l = Path length of the sample (cm )

The BPM and SpO2 values are calculated using the following equations:

6000
BPM=
Average Time between peaks (ms)

SpO2=110−( 25× R )

Where,

AC Red / DC Red
R=
AC IR/ DC IR

Finally, the processed data is transmitted via the I²C interface to a microcontroller (such as
an Arduino) for display or storage.

12

MACHINE LEARNING - REGRESSION MODEL

An intermediate stable reading is selected from the list of values and is given as input for
the polynomial regression model. The BPM and SPO2 values are not sufficient to build
linear model as the data set contains less information. It necessitates the use of the
Polynomial regression model to accurately predict the information. The inputs to the
polynomial regression model are SPO2 and BPM which are the digital output from the
sensor. As there are only two input vectors, to predict the values of glucose more features
are required which is obtained by using polynomial regression model. The squared input
vectors and the product of the input vectors are used as new polynomial features. These
polynomial features are fit into the polynomial regression model using the built in function
in Python. The polynomial regression equation for glucose prediction is given in equation
2 2
Y =ax+ bz+ c x +d z + exz+ f
Where,
a,d,c,d,e,f are constant coefficients
x = SpO2
z = BPM
Y = Glucose (mg/dl)

The polynomial regression model is trained using predetermined datasets and regression
co- efficient values are obtained from the trained model. The co-efficient values are used
to predict the new required parameter value for the new sample inputs from the sensor.
Thus, BPM and SPO2 values are used to predict the Glucose level concentration in the
blood.
The collected dataset in Table – 4.1 consists of 30 data samples and its respective SpO2,
BPM and Glucose values.

13
 CHAPTER-4
IMPLEMENTATION AND RESULT
The development of a non-invasive glucometer utilizing the MAX30100 sensor involves
a series of critical steps beginning with the meticulous selection and setup of the hardware.
The finger is placed on the sensor properly. According to the amount of infrared light
absorbed, the sensor estimates the values of SPO2 and BPM.
These data collected from the sensor are further fed to polynomial regression model (2nd
order polynomial equation) stored in Arduino UNO to process and estimate the glucose
levels.
The coefficients of the above equation are estimated by the collected data samples.
This processed data is then displayed to the user, typically on an LCD screen, providing
real- time feedback on their glucose levels.

SENSING UNIT

Obtain final

Glucose Value

IR Absorbance

value to

ARDUINO
Send the data to

LCD
Input value in the

Regression

equation

FLOW CHART

Figure-4.1

14
HARDWARE CONNECTIONS

Figure-4.2

These connections are made in CIRKIT IDE online simulator.

PIN CONNECTIONS FROM ARDUINO TO SENSOR AND

LCD
MAX30100 Arduino Uno I2C LCD Arduino Uno
VIN ─────► 3.3V VIN ───► 5V
GND ─────► GND GND ───► GND
SCL ─────► A5 SCL ───► A5
SDA ─────► A4 SDA ───► A4
INT ─────► (optional)

Figure-4.3
15
 IMPLEMENTED CODE BASE
STEP-1: Collect various data samples of SpO2, BPM and its corresponding glucose values
and export it as glucose.csv to train the model.

SpO2 BPM Glucose

98 72 110
95 80 125
93 90 140
99 65 100
96 78 118
97 74 112
94 85 132
92 88 142
97 68 102
98 82 123
98 74 114
92 61 110
98 78 117
98 99 131
98 71 113
94 76 102
97 71 104
97 78 106

Table-4.1

STEP-2: Now train the polynomial regression model in python with the obtained SpO2 and
BPM values from the dataset. Run train_model.py once to train the model.

16
 STEP-3: Create predict_glucose.py to start live prediction of glucose from SpO2 and
BPM values.

STEP-4: Extract the coefficients from the above dataset by using the following code in python.

STEP-5: Formulate a code in ARDUINO IDE to send the SpO2 and BPM values obtained
from the sensor to ML model and also receive Glucose values from ML model to Arduino.

17
#include <Wire.h>
#include <MAX30100_PulseOximeter.h>
#include <LiquidCrystal_I2C.h>

#define REPORTING_PERIOD_MS 250 // Faster update interval

#define BUZZER_PIN 6 // Buzzer connected to pin 6

#define BUTTON_PIN 7 // Reset button

PulseOximeter pox;
LiquidCrystal_I2C lcd(0x27, 16, 2); // Adjust address if needed

uint32_t tsLastReport = 0;
float spo2_sum = 0;
float bpm_sum = 0;
int count = 0;
bool fingerDetected = false;
bool measurementDone = false;

void onBeatDetected() {
// Optional heartbeat callback
}

void setup() {
pinMode(BUZZER_PIN, OUTPUT);
pinMode(BUTTON_PIN, INPUT_PULLUP); // Button input with pull-up

lcd.init();
lcd.backlight();
lcd.clear();
lcd.setCursor(0, 0);
lcd.print("Place finger...");

if (!pox.begin()) {
lcd.clear();
lcd.print("Sensor error!");
while (1);
}

pox.setIRLedCurrent(MAX30100_LED_CURR_11MA); // Faster signal detection

pox.setOnBeatDetectedCallback(onBeatDetected);
}

void loop() {
// ✅ Reset with button
if (measurementDone && digitalRead(BUTTON_PIN) == LOW) {
measurementDone = false;
count = 0;
spo2_sum = 0;
bpm_sum = 0;
fingerDetected = false;
lcd.clear();
lcd.setCursor(0, 0);
lcd.print("Place finger...");
delay(500); // Debounce
return;
}

if (measurementDone) return;

pox.update();

if (millis() - tsLastReport > REPORTING_PERIOD_MS) {

tsLastReport = millis();

float spo2 = pox.getSpO2();

float bpm = pox.getHeartRate();

if ((spo2 == 0 || bpm == 0) || isnan(spo2) || isnan(bpm) || spo2 < 90 || bpm < 40) {

if (fingerDetected) {
lcd.clear();
lcd.setCursor(0, 0);
lcd.print("Finger removed!");
lcd.setCursor(0, 1);
 lcd.print("Resetting...");

fingerDetected = false;
spo2_sum = 0;
bpm_sum = 0;
count = 0;

delay(1000);
lcd.clear();
lcd.setCursor(0, 0);
lcd.print("Place finger...");
} else {
lcd.setCursor(0, 0);
lcd.print("Waiting... ");
lcd.setCursor(0, 1);
lcd.print("No finger ");
}
return;
}

// Valid finger data

fingerDetected = true;
spo2_sum += spo2;
bpm_sum += bpm;
count++;

lcd.setCursor(0, 0);
lcd.print("Reading ");
lcd.print(count);
lcd.print("/5 ");

lcd.setCursor(0, 1);
lcd.print("Hold steady... ");

if (count >= 5) {
float avg_spo2 = spo2_sum / count;
float avg_bpm = bpm_sum / count;

// ML regression formula
float glucose = 11301.8481
- 238.7789 * avg_spo2
+ 6.0215 * avg_bpm
+ 1.2543 * avg_spo2 * avg_spo2
- 0.0280 * avg_spo2 * avg_bpm
- 0.0142 * avg_bpm * avg_bpm;

lcd.clear();
lcd.setCursor(0, 0);
lcd.print("S:");
lcd.print(avg_spo2, 1);
lcd.print(" B:");
lcd.print(avg_bpm, 1);

lcd.setCursor(0, 1);
lcd.print("G:");
lcd.print(glucose, 1);
lcd.print(" mg/dL");

// ✅ Beep buzzer
digitalWrite(BUZZER_PIN, HIGH);
delay(200);
digitalWrite(BUZZER_PIN, LOW);

// ✅ Freeze sensor
measurementDone = true;
}
}
}

19
STEP-5: The predicted glucose value is sent to Arduino via serial communication which
displays the SpO2, BPM and glucose values on the LCD.

Figure-4.4

20
RESULTS

SpO2 BPM Actual Predicted Accuracy

Glucose Glucose

Table-4.2

21
 CHAPTER-5
CONCLUSION AND FUTURE SCOPE
The research represents a significant advancement towards personalized and proactive
healthcare, promising improved management and outcomes for individuals living with
diabetes. To enhance the accuracy of our non-invasive glucose monitoring prototype, we
will expand the dataset to improve the model’s learning from diverse glucose levels and
skin types. Advanced machine learning algorithms, including deep learning and recurrent
neural networks, will aid in better predictions. High-resolution sensors, especially near-
infrared spectroscopy, will improve signal quality. Robust, individualized calibration
protocols will ensure precise readings. Integration of multimodal sensor data will provide
a comprehensive view of glucose levels. Extensive clinical trials with diverse participants
will refine and validate the prototype, ensuring its reliability.
Besides, the fact that the system can estimate the glucose levels, it can also be integrated
to desired devices like personal mobile phones, smartwatches using Bluetooth monitoring.
This can also be altered using wifi connections. It helps the users to monitor their glucose
levels effectively and detect the disease early to adjust their daily routine (diet plans).
Devices can also learn from user data to create custom calibration profiles, improving long
term accuracy.

22
 REFERENCES
Zheng T, Li W, Liu Y, Ling BW, A noninvasive blood glucose measurement system by
arduino and near-infrared, Proceedings of2016 IEEE International Conference on
Consumer Electronics-China (ICCE-China), China, pp. 1-3, 2016.
Yusoff IM, Yahya R, Omar WR, Ku LC, Non invasive cholesterol meter using Near
Infrared sensor, Proceedings of 2015 Innovation & Commercialization of Medical
Electronic Technology Conference (ICMET),Malaysia, pp. 100-104, 2015.
Daarani P, Kavithamani A, Blood glucose level monitoring by noninvasive method using
near infrared sensor, International Journal of Latest Trends in Engineering and
Technology,Vol. IRES 1, pp. 141-147, 2017.
Anis SN, Alias R, A Portable Non-Invasive Blood Glucose Monitoring Device with IoT,
Evolution in Electrical and Electronic Engineering, Vol. 2, pp. 36-44, 2021.
N. H. Ahniar, G. R. Lestari and R. N. Hidayati, A Non-Invasive Cholesterol Measuring
Device Using a Photodiode Sensor With a BLYNK Interface, Proceedings of 2022 5th
International Conference on Information and Communications Technology
(ICOIACT),Yogyakarta, Indonesia, pp. 473-478, 2022.
Gayathri B, Sruthi K, Menon KU, Non-invasive blood glucose monitoring using near
infrared spectroscopy, Proceedings of 2017 International Conference on Communication
and Signal Processing (ICCSP) , Chennai, India, pp. 1139-1142, 2017.
Shinde AA, Prasad RK, Non invasive blood glucose measurement using NIR technique
based on occlusion spectroscopy, International Journal of Engineering Science and
Technology (IJEST), Vol. 3, pp. 8325-8333, 2017.

23