Clusterization of Clinical Symptoms of Disease COVID-19 Using K:
Means Algorithm Based on Arduino Uno and Sensor Max 30100
Maria Ulfah and Andi Sri Irtawaty
Politeknik Negeri Balikpapan, Kec. Balikpapan Utara, Kota Balikpapan, Kalimantan Timur, Indonesia
Keywords: Covid-19, MAX30100, Arduino-Uno, Disease, K-Means.
Abstract: Symptoms of Covid-19 other than fever, which must be watched out by the public at this time is the condition
of shortness of breath due to a decrease in oxygen levels in the blood. This condition can make a person
experience respiratory problems such as shortness of breath or dyspnea which can be fatal, especially for
people who can interfere with vital organs, namely the heart and oxygen levels in the body to cause death.
This study designed a tool that can detect clinical symptoms of COVID-19 disease, namely shortness of breath
through measuring oxygen saturation levels and heart rate for shortness of breath symptoms using the
MAX30100 sensor and Arduino Uno as a controller. Tests for oxygen saturation levels and heart rate were
carried out by measuring the thumb and followed by testing the accuracy of the output using the clustering
technique using the K-Means Algorithm using the Rapid Miner Application. Comparison of readings of
clinical symptoms of shortness of breath through oxygen saturation levels. heart rate from the output of the
device compared with the use of the K-Means clustering technique showed the same results, which was 100%.
Tests 1-10,19,20 showed no symptoms of covid-19 while testing 11-18 showed symptoms of covid-19.
1 INTRODUCTION
Coronaviruses are a large family of viruses that cause
disease in humans and animals. In humans, it usually
causes respiratory tract infections, ranging from the
common cold to serious illnesses such as Middle East
Respiratory Syndrome (MERS) and acute respiratory
syndrome.
Severe Acute Respiratory Syndrome (SARS). A
new type of coronavirus found in humans since an
extraordinary event appeared in Wuhan, China, in
December 2019, was later named Coronavirus
Disease-2019 (COVID-19)
(infectionemerging.kemkes.go.id, 2021). On March
2, 2020, for the first time, the government announced
two cases of positive Covid-19 patients in Indonesia.
The spread of the Covid-19 virus has become a
worldwide concern in the past year. According to data
from information worldometers, as of February 26,
2021, there have been 113,516,853 cases of Covid-19
worldwide. the total number of people exposed to
COVID-19 in the country reached 1,527,524 cases
with a total of 41,242 cases died as of April 3, 2021
(covid19.go.id, 2021)
One of the symptoms that people have to watch
out for today is Happy hypoxia, which is a decrease
in oxygen levels in the blood, this condition can make
a person experience respiratory problems such as
shortness of breath or dyspnea which can be fatal,
especially for people who can interfere with vital
organs, namely the heart and lungs. oxygen levels in
the body to cause death.
Symptoms of Covid-19 are divided into 3 levels,
namely mild symptoms, moderate symptoms and
severe symptoms as follows (PDPI,2020)
1. Not Complicated / Mild:
1) Fever > 38
0
C
2) Cough.
3) Sore Throat.
4) Nasal congestion
5) Malaise
6). Headache
7).Muscle Pain
2. Mild Pneumonia:
1) Fever > 38
0
C.
2) Cough
3) Shortness of breath, persistent cough and sore
throat.
4) Children with mild pneumonia have a cough or
difficulty breathing and rapid breathing, respiratory
rate: <2 months, 60x/min; 2–11 months, 50x/ minute
Ulfah, M. and Irtawaty, A.
Clusterization of Clinical Symptoms of Disease COVID-19 Using K: Means Algorithm Based on Arduino Uno and Sensor Max 30100.
DOI: 10.5220/0011843900003575
In Proceedings of the 5th International Conference on Applied Science and Technology on Engineering Science (iCAST-ES 2022), pages 597-600
ISBN: 978-989-758-619-4; ISSN: 2975-8246
Copyright © 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
597
; 1–5 years, 40 x/min and no signs of severe
pneumonia.
3. Severe pneumonia. In adult patients:
1) Symptoms that appear include fever or suspicion
of respiratory tract infection.
2) Signs that appear are tachypnea (respiratory rate:
> 30x/minute), severe respiratory distress or patient
oxygen saturation <90% of the air outside.
The heart pumps blood throughout the body through
the arteries, which causes the arteries to contract or in
the vessels to stretch and constrict. So to measure the
heart rate can be done on the arteries by using
photoplethysmography. (Lukman, 2020)
The following is the classification of heart rate based
on age (Arthana, 2017)
Figure 1: Classification of heart rate by age.
2 METHODS
The research apparatus included an Arduino uno,
MAX30100 sensor, LCD , buzzer.
Starting at the initial state of all existing devices. then
measured oxygen saturation levels and heart rate
using the MAX30100 sensor. Result conditions:
1. Condition, if oxygen saturation level <95% and
heart rate >100 bpm indicates shortness of breath
Then: *There are Symptoms*
2. Condition, if oxygen saturation level >95% and
heart rate <100 bpm Then: *No Symptoms*
Based on the needs and implementation to help detect
clinical symptoms of COVID-19 disease using
Arduino uno, MAX30100 sensor.
Figure 2: Block Diagram.
3 RESULTSAND DISCUSSION
In testing the tools, the results of the research were
carried out on 20 people by measuring through the
thumbs of the hands and in parallel with the tools on
the market.
Table 1: Tool Test Results.
No
Name
Level
SP02
(%)
Level
Heart
Rate
(bpm)
LCD
Display
Buzzer
1
Arjun
98
82
No
symptoms
OFF
2
Aldy
98
69
No
symptoms
OFF
3
Arif
96
82
No
symptoms
OFF
4
Rahmat
95
80
No
symptoms
OFF
5
Figo
97
84
No
symptoms
OFF
6
Dea
95
84
No
symptoms
OFF
7
Sri
96
83
No
symptoms
OFF
8
Sigit
97
75
No
symptoms
OFF
9
Winda
97
77
No
symptoms
OFF
10
Ragil
96
78
No
symptoms
OFF
11
Nasrul
101
105
symptoms
ON
12
Heri G.
97
110
symptoms
ON
13
Rahel
99
120
symptoms
ON
14
Joko
95
115
symptoms
ON
15
Mahi
96
102
symptoms
ON
16
Heri
95
122
symptoms
ON
17
Agung
95
118
symptoms
ON
18
Fitri
97
102
symptoms
ON
19
Ahmad
98
86
No
symptoms
OFF
20
Chandra
96
90
No
symptoms
OFF
From testing 20 samples of SpO2 levels and heart
rate on the thumb finger, it can be seen on tests 1-10,
19, 20 indicating normal conditions (without
symptoms), where oxygen saturation values (SP02
and heart rate) are still in normal levels. While the 11-
18th test shows abnormal conditions (There are
iCAST-ES 2022 - International Conference on Applied Science and Technology on Engineering Science
598
Symptoms), where the oxygen saturation value (SP02
and heart rate) is above the normal value.
3.1 Implemented Tools
Figure 3: Inner View.
Figure 4: Top View.
Figure 5: Testing Directly with tools on the market. Desain
Tools.
3.2 K-Means Clustering
Figure 6: K-Means . Algorithm Process Design.
Figure 7: Clustering Results.
From the test results on the K-Means application,
the membership of Cluster 0 is 12 items and Cluster
1 is 8 items with a total of 20 data tested. Each
membership in detail will be shown in Figure 7
The following table compares the results of the
clusterisasu test directly from the output of the
designed tool and further testing using the K-Means
algorithm.
Furthermore, further clusterization testing was
carried out on 20 people who were tested, with the
aim of confirming the test results in Table 1 using the
K-Means algorithm. The test is carried out using the
Neat Miner Application with a value of k = 2
Table 2: Comparison of Clustering Results.
Test Type
Clustering,without
symptoms
Clustering,with
symptoms
No
Algoritma
K-Means
1-10, 19,20
11-18
Using
Algoritma
K-Menas
1-10, 19,20
11-18
From the results of a direct comparison of the tool
output with the results of data processing using the K-
Means Algorithm with the Rapid Miner Application,
the results of clustering (grouping) are the same in
determining symptomatic or asymptomatic people by
100%
Clusterization of Clinical Symptoms of Disease COVID-19 Using K: Means Algorithm Based on Arduino Uno and Sensor Max 30100
599
Figure 8: Membership of Each Cluster.
4 CONCLUSIONS
Based on the results of the research conducted, it can
be concluded as follows: Tests for oxygen saturation
levels and heart rate were carried out on the thumb
and clustering using the K-Means Algorithm with the
Rapid Miner Application was obtained on tests 1-10,
19, 20, no symptoms of covid-19 were found. because
oxygen saturation levels and heart rate are still within
normal limits. Meanwhile, in the 11-18 test, the
symptoms of covi d-19 were found where the oxygen
saturation value and pulse heart rate that exceeds
normal values. The buzzer works well as a warning
indicator if the oxygen saturation and heart rate
values exceed normal values. Comparison of readings
of clinical symptoms of shortness of breath through
oxygen saturation levels. heart rate shows the same
results between the output of the tool with testing
using the K-Means algorithm.
REFERENCES
Https://infectionemerging.kemkes.go.id/.(2021) , accessed
April (2021).
Https://covid19.go.id/. (2021). accessed April
PDPI. (2020). Indonesian Lung Doctors Association
“Clinical Practical Guide: Pneumococcal 2019-nCoV
PDPI”, Jakarta.
Lukman, A., Riska. (2020). "Design of a Non-Invasive
Oxygen Level Measuring Device Using a Max30100
Sensor". West Java: Department of Electrical
Engineering. Faculty of Engineering, Krisnadwipayana
University West Java.
Arthana IKR, et al. (2017). “Design of a Heartbeat Detector
and Notification Via SMS”.
iCAST-ES 2022 - International Conference on Applied Science and Technology on Engineering Science
600