Design and Development of Esp32-Based Non-Invasive Blood Sugar

Level Measurement Equipment

Dika Ari Saputra and J. Rajes Khana

Universitas 17 Agustus 1945 Jakarta Podomoro, Jl. Sunter Permai Raya Sunter Agung, Kec. Tj. Priok, Kota Jkt Utara,

Daerah Khusus Ibukota, 14350, Jakarta

Keywords: Photodiode, Blood Sugar Level, and Non-Intrusive.

Abstract: This research was conducted to produce a non-invasive blood sugar measuring device, which is a device that

can measure blood sugar levels without having to test and examine the blood directly but using a finger

attached to the sensor as a specimen, the Photoplethysmography (PPG) method is used in the design of the

device system in this work is to measure blood sugar levels. The architecture of this system uses an easy-to-

use and low-cost optical measurement technique. The ILI9488 LCD module also functions to display, in this

study the results of the calculation of the average percentage error of the tool are 3.52% with a calculation

that has an accuracy value of 95.48% from the actual tool. Judging from the calculation results, the percentage

experienced a measurement error of 0% - 3.52% in 16 trials. Thus, this GlucoTest tool can only be used as a

comparison of the sugar content of a glucometer in general. And it takes a lag time of 1 to 3 minutes for the

same patient or respondent to take measurements again.

1 INTRODUCTION

Diabetes mellitus (DM) is a disease in which the

amount of sugar in the blood is too high

(hyperglycemia) due to the inability of the pancreas

to produce insulin. When the pancreas cannot

produce insulin, the body has difficulty maintaining

blood sugar levels, resulting in high blood sugar

levels or hyperglycemia. Diabetes is divided into type

I diabetes and type II diabetes. Type I diabetes occurs

when the body is unable to produce insulin, while

type II diabetes occurs when the body is unable to

receive and process insulin efficiently. This second

can lead to obesity and lack of physical activity (F. Z.

Kamilah et al.,2021).

In blood or serum there is a concentration of

glucose called blood glucose, a normal limit for a

person who has not eaten for 3 or 4 hours and then

close to 90 milligrams or dl. Even though the

consumption of foods that contain lots of

carbohydrates, this focus does not often increase

above 140 milligrams or dl, but the person has

Diabetes Mellitus. Glucose that flows through the

blood is an important base of energy for the body's

cells. Blood glucose is sugar in the blood that is

created from carbohydrate metabolism. Checking

blood glucose is one of the checks in clinical trials (N.

F. Fahmi et al., 2021)

Currently, the most commonly used tool to

measure blood sugar levels is a blood glucose meter.

Measurement of blood glucose levels was carried out

using a glucometer which works enzymatically based

on the glucose oxidase reaction. The glucose meter

works on the principle of a biosensor. A biosensor is

a combination of a biological receptor and a

transducer. Bioreceptors are devices used to sense the

concentration of biological elements, such as

enzymes, antibodies, living cells, and other tissues.

The function of the sensor is to convert biochemical

signals into electrical signals, which then appear on

the blood glucose meter screen (J Fine et al., 2021).

However, in taking blood samples, it is necessary

to use a tool in the form of a syringe, although the

process is carried out quickly but there are still many

patients who are afraid of needles for taking blood

samples. Therefore, an accurate blood glucose meter

is needed without injuring the body first (non-

invasive).

In this study, a tool was made to measure blood

sugar levels without injuring the body. The device

does not require urine as a test material to measure the

patient's blood sugar level. This tool works on the

principle that the photodiode sensor will capture light

Saputra, D. and Khana, J.

Design and Development of Esp32-Based Non-Invasive Blood Sugar Level Measurement Equipment.

DOI: 10.5220/0012026300003582

In Proceedings of the 3rd International Seminar and Call for Paper (ISCP) UTA â

Z45 Jakarta (ISCP UTA’45 Jakarta 2022), pages 381-385

ISBN: 978-989-758-654-5; ISSN: 2828-853X

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

381

from a light source. used are LEDs. When the tool is

turned on, the light from the LED will radiate and

pass through the finger before reaching the

photodiode. The intensity of light received by the

photodiode changes from a factor of changes in the

number of blood sugar molecules. The photodiode

output signal will be processed on the ESP32. After

the signal is processed by esp32, the patient's blood

sugar value can be known. This blood sugar value is

shown on the TFT LCD screen attached to Glucotest.

The output of the tool includes the value of blood

glucose levels and a description (low/normal/high) in

units (mg/dl).

2 RESEARCH METHODS

This research was conducted in several locations, the

place is in the electrical laboratory at the University

of 17 August 1945, Jakarta to design a non-invasive

blood sugar level measuring device based on esp32.

And for sampling, the trial was carried out in the

author's home area.

2.1 System Block Diagram Design Tool

The overall planning of the hardware block diagram

made in this research is as follows:

Figure 1. System Block Diagram.

In the block diagram, the output system of the

photodiode will enter the microcontroller for

processing and display

2.2 Sensor System Design

The design of this sensor system circuit uses a red

LED as a light source and a Photodiode sensor as a

receiver. With a schematic of the circuit as shown in

Figure 2

Figure 2: Sensor Circuit.

The resistance value of the resistor can be

calculated using the formula:

R =





 





R =





 





R =

 

 

= 100Ω

)

So the value of the resistor is considered to be

100Ω

where

R = current limiting resistor

𝑣



= 5v (tegangan Input)

𝑣



= 2v (Input voltage)

I = 50 mA = 0,05 A

2.3 Voltage Buffer Design

The voltage buffer circuit design stage is needed to

stabilize the sensor reading value. In this circuit the

components used are 538 op-amp and 10kΩ trimmer

Figure 3. Voltage buffer Circuit.

This arrangement is very profitable because we can

get an amplifier with very large input resistance

(input impedance) and with very small output

resistance (output impedance), which is close to

ISCP UTA’45 Jakarta 2022 - International Seminar and Call for Paper Universitas 17 Agustus 1945 Jakarta

382

perfect conditions. As a result, an operational

amplifier with a shape like the one in the picture

above acts as a buffer with a gain = 1. The application

of a good buffer arrangement made of a transistor

amplifier or operational amplifier (Op-Amp) is

usually used as a signal stabilizer.

2.4 Program Design

The software design stage begins with making a

flowchart followed by making a lasting program

using Arduino Uno. The software design flowchart is

as follows:

Figure 4: Flowchart Program.

When the start button is pressed, the program will

start port initialization, if the sensor detects a finger,

the program will read the sensor value, if the sensor

does not detect a finger, the program will again ask

for finger input, then if the value is obtained, the value

will be converted and determined whether the blood

sugar level is low, normal or height is then displayed

on the LCD.

3 RESULTS AND ANALYSIS

This chapter discusses the results and systems that

have been designed and built. It is designed to

determine if the system meets the standards the author

wants. Each block is tested with the aim of Knowing

the work of each circuit block, making it possible to

know whether each circuit module performs its

function properly. Proceed with the overall test. The

test of this tool consists of 7 main parts, namely:

1. Sensor

Testing

2. Measurement Condition Testing

3. LCD

Testing

4. Overall Tool Testing

5. Reading Value

6. Research Results.

7. Error Percentage.

3.1 Sensor Testing

In sensor testing, it is done by measuring the output

voltage of the sensor when it is not blocked by a

finger and when the sensor is blocked by a finger. The

sensor measurement results can be seen in table 1.

Table 1: Sensor Measurement Results.

No Condition ADC Out

ut Volta

1 Not blocked by

fingers

470 1,51 V

2 Blocked b

fin

878 2,83 V

In this test, the adc value is 470 when the sensor

is not blocked by a finger and will produce an adc

value of more than 470 when the sensor is blocked by

a finger. This shows the sensor is functioning

properly. According to the way the photodiode sensor

works, the brighter the light, the smaller the sensor

reading value. Conversely, the darker the light, the

greater the sensor reading value.

3.2 Measurement Condition Testing

In this test, it is carried out by simulating two

measurement conditions, namely when measuring in

the dark (the instrument is in a closed condition) and

when the instrument is in a light condition (the

instrument is in an open condition). The results of this

test can be seen from table 2.

Table 2. Testing Measurement Conditions

Condition ADC LUX Tool

Results

Glukometer

Results

Dark

conditions

788 281 94 mg/dL

98 mg/dl

Light

Condition

920 680

165

mg/dL

The results of these tests show that the darker the

measurement conditions, the more accurate the

measurement results, on the contrary, the brighter the

measurement conditions, the more error the reading

value will be.

Design and Development of Esp32-Based Non-Invasive Blood Sugar Level Measurement Equipment

383

3.3 LCD Testing

The ILI9488 TFT LCD test aims to find out whether

the lcd can function properly. The use of pins on the

ILI9488 TFT LCD, there are several pins that are

used as data sender and data receiver pins, namely

CS, RESET, DC/RS, SDI(MOSI), SCK pins. For the

touchscreen pins, namely T_CLK, T_CS, T_DIN,

TDO. and for power use the VCC, GND, and LED

pins as the backlight power which is connected

directly to the esp32 pin.

Figure 5: Tool Preview.

3.4 Overall Tool Testing

Overall system testing is carried out after testing in

each part of the tool. The purpose of this test is to

understand how the esp32-based non-invasive blood

glucose meter works as shown in Figure 5. The whole

system of the whole system

Figure 6: Overall System Circuit.

Block diagram A shows the sensor network, the

output from the photodiode sensor is passed to pin A0

on the esp32,. Block B shows the overall picture of

the esp32 network and input from the tft lcd module,

.In block C there is a battery as a voltage source, and

in block D is the power supply network.suplay.

Figure 7: Overall View of the Tool.

Figure 6 shows the overall shape of the finished

tool, where all the components needed have been

assembled into a single unit and the tool is ready to

be used.

3.5 Reading Value Converter

The test between the 3 in 1 autocheck medical device

and the tool made has a different display of

measurement results so that conversion must be done

in order to get good results, the adc value is obtained

when taking measurements and then converted so that

the measurement results displayed show the mg/dl

value. this conversion is done withdengan

Nilai Gula Darah =





– 295

(2)

The test results after conversion are obtained

that are close to the 3 in 1 autocheck comparison tool.

As shown in Figure 8.

Figure 8: Measurement results.

3.6 Research Result

In this study, it was carried out by measuring ten

respondents who would be measured using a tool that

was made and simultaneously it would also be

measured with a 3-in autocheck tool, the

measurement data would be written into a table to

find out the measurement error between the tool made

and the tool used as a comparison. shown in table 3.

ISCP UTA’45 Jakarta 2022 - International Seminar and Call for Paper Universitas 17 Agustus 1945 Jakarta

384

Table 3: Blood Sugar Level Measurement Results

No ADC

Conversion

Value

(mg/dL)

Standard Value

(mg/dL)

776 93 96

740 75 78

765 87,5 91

779 94,5 97

810 110 113

706 58 62

768 89 91

804 107 110

713 61,5 65

780 95 99

878 144 148

868 139 147

847 128,5 135

770 90 93

720 65 69

790 100 103

3.7 Persentasi Error

Table 4: Presentation Error.

No ADC

Nilai Konversi

(mg/ dL)

Nilai

Standart

(mg/dL)

Keakurasian

1 776 93 96 3,13%

2 740 75 78 3,13%

3 765 87,5 91 3,65%

4 779 94,5 97 2,60%

5 810 110 113 3,13%

6 706 58 62 4,17%

7 768 89 91 2,08%

8 804 107 110 3,13%

9 713 61,5 65 3,65%

10 780 95 99 4,17%

11 878 144 148 4,17%

12 868 139 147 4,17%

13 847 128,5 135 4,69%

14 770 90 93 3,13%

15 720 65 69 4,17%

16 790 100 103 3,13%

The results obtained from the overall instrument

testing will be compared with the results obtained

from the invasive blood sugar measuring instrument

namely the glucometer so that the percentage error is

obtained using the equation.

4 CONCLUSIONS

After carrying out the design and manufacture of the

tool, which is then followed by the testing and

analysis phase, the following conclusions can be

drawn: Measurable blood sugar levels of 25-

300mg/dLd, The resistance sensor used is sensitive to

the state of the blood when it is fresh and accurate.

reading persists 30 seconds after sensor reading,

Judging from the results of the calculation of the

percentage experienced a measurement error of 0% -

4.62% in 16 trials. Thus the tool cannot be used as a

reference for measuring blood sugar values in the

body. Judging from the results of the research, the

tools made show that the 11th to 16th measurements

show a decreasing accuracy. It takes a lag time of 1 to

3 minutes for the same patient or respondent to take

measurements again.

REFERENCES

F. Z. Kamilah et al., “Analysis of the Determinants of

Diabetes Mellitus in Indonesia: A Case Study of the

2014 Indonesian Family Life Survey,” Disease

Prevention and Public Health Journal, vol. 15, no. 2,

2021, doi: 10.12928/dpphj.v15i2.3079.

N. F. Fahmi, N. Firdaus, and N. Putri, “Article

PENGARUH WAKTU PENUNDAAN TERHADAP

KADAR GLUKOSA DARAH SEWAKTU DENGAN

METODE POCT PADA MAHASISWA.”

J. Fine et al., “Sources of inaccuracy in

photoplethysmography for continuous cardiovascular

monitoring,” Biosensors, vol. 11, no. 4. 2021. doi:

10.3390/bios11040126.

Design and Development of Esp32-Based Non-Invasive Blood Sugar Level Measurement Equipment

385