2022 4th International Conference on Smart Sensors and Application (ICSSA)

Development of Blood Oxygen Level, Heart Rate

And Temperature Monitoring System by Using
ESP32
Robiah Ahmad Hazilah Mad Kaidi Muhammad Noor Nordin
Department of Engineering and Department of Engineering and Medical Engineering Technology
Technology Technology Section
2022 4th International Conference on Smart Sensors and Application (ICSSA) | 978-1-6654-9981-1/22/$31.00 ©2022 IEEE | DOI: 10.1109/ICSSA54161.2022.9870943

Universiti Teknologi Malaysia Universiti Teknologi Malaysia Universiti Kuala Lumpur British
54100 Jalan Sultan Yahya Petra 54100 Jalan Sultan Yahya Petra Malaysian Institute
Kuala Lumpur, Malaysia Selangor, Malaysia Selangor, Malaysia
[email protected] [email protected] [email protected]

Aizat Faiz Ramli Mohd Azlan Abu Yosof Kadase

Electronics Technology Section Malaysian Japan International Institute Electronic Technology Section
Universiti Kuala Lumpur British of Technology Universiti Kuala Lumpur British
Malaysian Institute Universiti Teknologi Malaysia Malaysian Institute
Selangor, Malaysia 54100 Jalan Sultan Yahya Petra Selangor, Malaysia
[email protected] Kuala Lumpur, Malaysia [email protected]
[email protected]

Abstract— Telehealth plays an even greater role in global sectors such as the fast-food industry or unhealthy lifestyles.
healthcare delivery today. Ever since the pandemic strikes the In particular, 80 per cent of the 35 million deaths associated
world, the use of healthcare monitoring systems has a massive with chronic diseases for 2019 would occur in regions with
boost significantly. This paper aims to develop a prototype of a low and medium incomes. Men and women are infected
health monitoring device that can be put at a patient home to younger than in countries with high incomes. In the next ten
replace traditional physical checkups at the hospital by using years, premature deaths are estimated to cost billions of
MAX30102 PPG module and Infrared temperature sensor with dollars in countries such as China, India and the Russian
the objective to derive PPG signal into basic health parameters Federation. In several nations, the urgency of risk mitigation
such as heart rate and oxygen saturation from the patient
and disease prevention strategies is working, and a series of
through MAX30102 and to test the accuracy of those
low-cost, high-impact initiatives can be applied step-by-step
parameters with hospital patient monitor. The experiment is
done by comparing each parameter for 30 seconds in both [8].
devices using index fingers coming from 13 participants. From Developed countries such as the United States, the United
the results obtained, although heart rate suffers motion artefact Kingdom, and Canada are experiencing a nursing shortage as
compared to the oxygen saturation, each accuracy is still the number of critically ill patients rises. The issue gets even
excellent with 89.2 and 97.6 percent. In conclusion, MAX30102 more complicated. Increasing the number of chronically sick
PPG and ECG module can get initial health reading through the
patients and the shortage of healthcare staff will increase the
PPG signal and solve the problems mentioned.
cost of services in the healthcare sector. A variable method is
Keywords—Telehealth, ESP32, monitoring required to define patient management supervision and
treatment while minimizing costs by delivering home health
monitoring and care, called home telemonitoring.
I. INTRODUCTION One of the technologies that can assist in the health
monitoring system is the IoT. Blood oxygen saturation, pulse
The Internet of Things (IoT) is a network of connected
rate, body temperature, heart rate, and body temperature are
devices and sensors that extends the internet and other
all vital indicators that should be examined on a regular basis
network connections. For example, it can connect to alarms,
[9].
gates, cars, and other objects with high-level determining and
scientific capabilities. Many industries, including automotive This paper is organized as follows: Section II reviews the
[1], agriculture [2], [3], healthcare [4]–[6] and many more previous related work on healthcare IoT systems. Section III
have embraced the IoT. The healthcare industry is one such describes the methodology of this prototype development and
trending field that provides good health services to all people Section V explains the experiment that has been conducted
worldwide. and a brief discussion of the results. Section IV finally
concludes.
Health by nature, is an illness or injury-free indication.
Advancing biomedical technology and creating a longer life II. RELATED WORKS
span is crucial to maintaining the quality of life. Healthcare is
the advancement of health in diverse medical fields through The human body consists of a signal which carries
prevention, diagnosis, and treatment offered by health specific information about the physiological system and can
practitioners. In the last 50 years, people aged 60 and above be examined by measuring the vital signs. Based on the
have tripled numbers and are predicted to be almost two analysis of several previous research, the trend is much more
billion by 2050 [7]. focused on measuring blood oxygen (SpO2) [10], heart rate
and temperature measurement. While other researchers also
In the 21st century, chronic illnesses such as coronary
include the measurement of blood pressure and ECG
conditions, heart failure (HF), high blood pressure, or obesity
measurement [11].
grew along with growth technologies in different capitalist

Authorized licensed use limited to: ANURAG GROUP OF INSTITUTIONS. Downloaded on July 02,2024 at 09:09:17 UTC from IEEE Xplore. Restrictions apply.
978-1-6654-9981-1/22/$31.00 ©2022 IEEE 167
 2022 4th International Conference on Smart Sensors and Application (ICSSA)

In the field of medical science, some significant work has

been done using IoT to monitor patient health. Examples of
works related to this field are explained below.
In an IoT network, Tamilselvi et al., 2020 [12] built a
health monitoring system that can track a patient's
fundamental symptoms such as heart rate, oxygen saturation
percentage, body temperature, and eye movement. Heartbeat,
SpO2, Temperature, and Eye Blink sensors were employed as
capturing elements, while an Arduino-UNO was used as a
processing device. Although the developed system was
deployed, there are yet defined performance measures for any
of the patients.
In the IoT setting, Acharya et al. 2020 [13] presented a
healthcare monitoring kit. Heartbeat, ECG, body temperature,
and respiration were among the basic health metrics tracked
by the designed system. Pulse sensor, temperature sensor,
blood pressure sensor, ECG sensor, and Raspberry Pi are the
main hardware components employed here. Sensor data was
captured and forwarded to a Raspberry Pi for processing
before being sent back to the IoT network. The system's
biggest flaw is that no data visualisation interfaces have been
created.
Most of the suggested health monitoring systems operate
offlin[14]. The online system is preferred to boost system
efficiency. Some suggested online systems use radio
frequency (RF) [15] to monitor a patient’s condition at a
remote node, although this does not cover a large region.
The data from the sensors can be collected using a variety
of methods. The raw data from the sensors usually require an
interface to make it readable. The detection of all four sensors Fig. 2. A flow chart of the proposed system
has been interfaced with Arduino in various works [16], [17].
The Arduino generally features a Wi-Fi connection III. PROPOSED SYSTEM
through an external module, making data transmission to web
services easier. Other articles, on the other hand, have used an A health monitoring system using photoplethysmography
ESP32 as an interface to read the data. In addition, the ESP32 (PPG) and Infrared (IR) sensors is proposed for this project
microcontroller features built-in Wi-Fi and Bluetooth dual- implementation to acquire patient health information and
mode [18]. Aside from that, the ESP32 has a larger memory transfer its data to health care providers. This system utilizes
capacity than the Arduino [19]. ESP32 as the system processor, Maxim MAX30102 PPG
module as in [20] and MLX90614 IR Temperature module.
Some proposed systems do not have a warning system The output is to be used as numerical details and can be used
when the readings are out of the ordinary. Some of them lack by users when necessary in monitoring health conditions. The
a built-in application that can show real-time readings. These patient's device creation would position the index finger at the
functionalities are included in our suggested system, and the top of the photoplethysmography and forehead for the IR
various sensors are examined in the study to find the best sensor. The sensor will then use infrared and red monitoring
performance. to detect the user's heart rate, blood oxygen level, and
temperature. The data collected will be displayed in android
apps for patients or health experts.
Based on Fig. 1, the health monitoring system would
consist of two inputs which are a PPG and an IR temperature
sensor. The PPG is used for heart rate and blood oxygen
saturation. Meanwhile, the second input is used to obtain the
subject's body temperature. The power supply for this project
will be from a LIPO rechargeable battery (1100mAH).
After these data have been collected, the sensor input is in
raw signal data. Then, specific coding would be used in ESP32
using tensilica xtensa lx6 microprocessor with low noise
receive amplifier, power amplifier, filters, and power-
management modules to translate these signals into numerical
numbers. Then send those data outputs which show reading of
BPM, SpO2 and 0C.
Fig. 1. A block diagram of the proposed system

Authorized licensed use limited to: ANURAG GROUP OF INSTITUTIONS. Downloaded on July 02,2024 at 09:09:17 UTC from IEEE Xplore. Restrictions apply.
168
 2022 4th International Conference on Smart Sensors and Application (ICSSA)

Fig. 3. The prototype of the proposed system

Fig. 5. Typical PPG waveform including systolic and diastolic peak

The key to creating oxygen saturation in the PPG signal

lies in the infrared and red led ratio. The oxyhemoglobin
contains oxygen when attracted to the 950nm wavelength of
infrared. While non-oxyhemoglobin will be attracted to the
650nm wavelength of red light. From the ratio of both
wavelength absorbance, oxygen saturation in human body can
be calculated by using an equation
Fig. 4 Circuit diagram for the proposed system aR^2+bR+c. (1)
The a, b and c values are constant, which are a=-16.67, b=
The project flow (Fig. 2) will start when ESP32 is 8.33 and c =100. So for R-value is coming from
connected to the internet and cloud systems application. Then, R=Red/Infrared. Value a, b and c are obtained through series
MAX30102 and MLX90614 temperature sensors will start to of comparative measurement with medical grade SpO2 sensor
initialize (Fig. 3). If those sensors are unsuccessful to with reading obtain by our PPG sensor. Thus, measurement of
initialize, then the system will reboot. Once both of the sensors heart rate is conducted by counting systolic peak for each
have initialized, the system will check if the required cardiac cycle for ten second and then multiply those peaks by
condition is met by identifying whether user index finger is on 6 to get beat per minute (BPM) measurement.
the sensor. In addition, if the condition is met, both sensors
will proceed to send the information to output. If no output is
shown or zero reading is captured, the system will loop back B. Accuracy Measurement
to retain the sensor information.
Fig. 4. show a complete circuit diagram for the actual
system. The sensor including MAX30102 and MLX90614 In order to ensure sensor accuracy while collecting user
will be power by ESP32 3.3V pin. ESP32 will draw the power health information, a series of comparative measurements is
from LIPO rechargeable battery through the battery charger conducted to ensure all information recorded is on par with
module. The output system can be viewed directly to Arduino certified available healthcare products in the market. To
IDE serial monitor or through dedicated cloud systems. compare the reading of heart rate and blood oxygen level the
selected product that been chosen is Philips patient monitor
IntelliVue X3 (Fig. 6). Meanwhile, for body temperature, the
A. Measurement Algorithm selected device was a non-contact Infra-red thermometer by
Duratool (DR3055) as shown in Fig. 7. The selected devices
are certified by SIRIM and Malaysia Medical Device
Heart rate from MAX30102 is a derivation from the PPG Authority (MDA).
signal. It is based on the systolic and diastolic peaks as
indicated in Fig. 5. The measurement is done by counting the
systolic peak for each cardiac cycle for one second and then
multiplying those peaks by 60 seconds by using the formula
60 x (delta/1000).
MAX30102 provides PPG that contains infrared and red
led to detect blood volume changes in the microvascular bed
of tissue. This kind of detection has been used in SPO2
devices in clinical care. In theory, MAX30102 can detect
oxygen saturation in the human body by using infrared and red
led to measure changes in the light absorption in the human
body. This subsection explains how to read oxygen saturation
Fig. 6. Philips patient monitor model IntelliVue X3
in the human body using MAX30102.

Authorized licensed use limited to: ANURAG GROUP OF INSTITUTIONS. Downloaded on July 02,2024 at 09:09:17 UTC from IEEE Xplore. Restrictions apply.
169
 2022 4th International Conference on Smart Sensors and Application (ICSSA)

TABLE I. THE HEART RATE DATA COLLECTION

Subject Age Patient Heart Rate Error

Monitor (%) Prototype (%) (%)

Subject 1 24 95 86 9

Subject 2 24 105 92 12

Subject 3 22 85 79 7

Subject 4 22 63 63.75 1
Fig. 7. Duratool DR3055 non-contact IR thermometer
Subject 5 22 78 77.5 1

IV. EXPERIMENT AND RESULT Subject 6 23 83 73.5 11

Subject 7 22 90 87.75 2.5

Subject 8 23 65 62.75 3

Subject 9 22 87 83 4.5

Subject 10 23 84 87.5 4

Subject 11 22 102 101.5 0.4

Subject 12 23 90 27.5 70

Subject 13 22 98 82 16

Fig. 8. MAX30102 raw PPG signal

B. SpO2
The raw PPG signal produced by MAX30102 is shown in Table II tabulates the SpO2 data collected from subjects.
Fig. 8. However the signal is noisy due to increasing optical The patient monitor's highest value is 100% oxygen saturation
density and path length of illuminated tissue while decreasing collected from all subjects except for subjects 4 and 5. The
the light intensity of the photodetector so that the PPG signal lowest is 97% oxygen saturation collected from subject 5. The
is correlated with blood volume. Therefore, it is hard to see spo2 data collection.
systolic and diastolic peaks from the produced signal. To solve
this issue, we create an analogue interface circuit as the op- Subject Age Patient Product Error
amp will work as a current to voltage converter. Also, series Monitor (%) Prototype (%) (%)
of filter has been developed to remove excess of noisy PPG
signal. For SpO2, the measurement conducted base on Eq (1). Subject 1 24 100 98 2
Meanwhile for the MLX90614 IR sensor, no changes are
Subject 2 24 100 96 4
required to be done on raw signal and measurement was
directly converted by microprocessor based on input voltage Subject 3 22 100 98 2
from the sensor.
A. Heart Rate Subject 4 22 99 98 1

Table I compares the heart rate parameter between the Subject 5 22 97 100 3
patient monitor and the product prototype. There are 13
subjects from ages 22 to 24 years old, two subjects with age Subject 6 23 100 97.8 2.2
24 years old, seven patients with age 22 years old and four
patients with age 23 years old for all obtained data. The patient Subject 7 22 100 98 2
monitor's highest value is 105 BPM collected from subject 2,
Subject 8 23 100 98.9 1
and the lowest is 63 BPM collected from subject 4. For the
product prototype, the highest value is 101.5 BPM from Subject 9 22 100 98.73 1.3
subject 11, and the lowest is 27.5 BPM from subject 12.
The low reading in patient 12 is suspected due to poor Subject 10 23 100 98.81 1.1
sensor placement during heart rate measurement. All test
subject has been instructed not to move their finger and in rest Subject 11 22 100 98 2
position throughout the measurement. The measurement was
Subject 12 23 100 98.9 1
conducted for 2 minutes straight to ensure reading accuracy.
Subject 13 22 100 93 7

Authorized licensed use limited to: ANURAG GROUP OF INSTITUTIONS. Downloaded on July 02,2024 at 09:09:17 UTC from IEEE Xplore. Restrictions apply.
170
 2022 4th International Conference on Smart Sensors and Application (ICSSA)

For the product prototype, the highest value is 100% The difference between the highest and lowest value of
oxygen saturation from subject 5, and the lowest is 93% from the infrared thermometer is 3.2 °C, and for the product
subject 13. prototype is 6.89 °C. The infrared thermometer average value
Subject 5 oxygen saturation on the product prototype is is 34.83 °C and 33.33 °C for the product prototype.
excellent. Unfortunately, the value on the patient monitor is Meanwhile, the average error value is 5.2% with 95%
97% which is peculiar because the patient monitor is accuracy.
standardize medical equipment with better accuracy and filter As the discussion for results tabulated in Table I and II,
than the product prototype. Upon investigation, subject 5's the comparative measurement was conducted has shown a
finger was moving inside the patient monitor finger clip, high percentage similarity with medical grade equipment. It
creating motion artefacts that increased light level changes. calculated that for heart rate, the average similarity recorded
The inadequate signal is caused by motion artefact makes the at 89%, while for SpO2 at 98%. The reason for much lower
patient monitor difficult to calculate oxygen saturation, as heart rate measurement accuracy, although coming from the
shown in Fig. 4. same sensor MAX30102 is movement artefact sensitivity.
C. Body Temperature The heart rate measurement is very sensitive to the motion of
the test subject's finger. Another issue related to measurement
is the method of both sensors in capturing the light produced
There are plenty of places to check the body's
by a red and infrared LED. For MAX30102, the sensor uses
temperature, such as the forehead, mouth, and ear. However,
reflective method for detection of light while the patient
this experiment will use the index finger to capture body
monitor uses transmission of light through the finger [21].
temperature since the index finger has the most arteriovenous
Meanwhile, body temperature measurements indicate
anastomoses or AVAs. It regulates vasoconstriction and
below range of normal body temperature around 36.5 to 37.5
vasodilatation more than the rest of the body surface. The °
finger also correlates with corneal temperature as blood flow C. Both IR temperature sensors under performance mainly
is parallel with the finger and the eye. due to two major factor: distance of the subject with the
Table III depicts the body temperature data for 13 sensor and the environment where measurement is taken.
These two factors contribute massively on producing
subjects. The infrared thermometer's highest value is 36.7 °C
accurate body temperature measurements. Furthermore, it is
collected from the subject, and the lowest is 33.5 °C from
reported by Hsuan et al [22] that measurement of core body
subject 10. For the product prototype, the highest value is
temperature by using IR sensor probably requires strict
35.6 °C from subject 3, and the lowest is 28.71°C from standard operating procedure (SOP) due to this issue. They
subject 13. also suggest that all measurement related to IR sensor
temperature has to be undergo the black box method to ensure
TABLE II. THE BODY TEMPERATURE DATA COLLECTION
accuracy.

Subject Age Infrared Product Error V. CONCLUSION

thermometer Prototype (%) In this system, user health monitoring can be achieved by
(oC) (oC) collecting the physiological information from the various
sensor, microcontroller ESP32 and output. The MAX30102
Subject 1 24 35.7 34.5 3.3 module has shown excellence in measuring heart rate and
blood oxygen level parameters, scoring accuracy around 89%
Subject 2 24 35 34.7 0.8
and 98%, respectively. While for IR temperature, the
Subject 3 22 36 35.6 1 comparative studies show MLX90614 accuracy recorded at
95%. The system can accurately collect user health
Subject 4 22 35.8 34.8 2.7 information as a guide for better health assessment and
lifestyle. This system can be further improved with additional
Subject 5 22 36.7 35.3 3.8 health information detection such as blood pressure, which is
a more complex algorithm. In the near future, this particular
Subject 6 23 34.3 34.71 1
system will change the healthcare monitoring system
Subject 7 22 34.7 33.4 4.8 worldwide.

Subject 8 23 35.6 35.31 0.8

ACKNOWLEDGEMENT
Subject 9 22 33.7 32.6 3
The authors would like to express their gratitude to Universiti
Subject 10 23 33.5 32.45 3
Teknologi Malaysia and Universiti Kuala Lumpur, which
Subject 11 22 34.1 33.4 2 have provided research funding through Collaborative
Research Grant Vot R.K130000.7356.4B416 and
Subject 12 23 34.3 28.71 16 CRGS/2019/UniKL/03/1.

Subject 13 22 34 30 11

Authorized licensed use limited to: ANURAG GROUP OF INSTITUTIONS. Downloaded on July 02,2024 at 09:09:17 UTC from IEEE Xplore. Restrictions apply.
171
 2022 4th International Conference on Smart Sensors and Application (ICSSA)

REFERENCES [16] R. Sameh, M. Genedy, A. Abdeldayem, and M. H. Abdel azeem,

“Design and Implementation of an SPO2 Based Sensor for Heart
Monitoring Using an Android Application,” Journal of Physics:
[1] D. Pejic, V. Krizanovic, and K. Grgic, “Detection of attacks and Conference Series, vol. 1447, no. 1, p. 012004, Jan. 2020, doi:
intrusions on automotive engine IoT sensors,” in Proceedings of 10.1088/1742-6596/1447/1/012004.
the 16th International Conference on Telecommunications, [17] Md. S. Mahmud, H. Wang, A. M. Esfar-E-Alam, and H. Fang, “A
ConTEL 2021, Jun. 2021, pp. 134–141. doi: Wireless Health Monitoring System Using Mobile Phone
10.23919/ConTEL52528.2021.9495981. Accessories,” IEEE Internet of Things Journal, vol. 4, no. 6, pp.
[2] H. M. Kaidi et al., “Internet of Things: A Monitoring and Control 2009–2018, Dec. 2017, doi: 10.1109/JIOT.2016.2645125.
System for Rockmelon Farming,” International Journal of [18] M. Islam, A. Rahaman, and R. Islam, “Development of Smart
Integrated Engineering, vol. 12, no. 6, pp. 55–61, 2020, doi: Healthcare Monitoring System in IoT Environment,” vol. c, 2020.
10.30880/ijie.2020.12.06.007. [19] S. K., P. A. S., P. U., P. S., and S. P. V., “Patient health monitoring
[3] A. N. Harun, R. Ahmad, N. Mohamed, A. R. A. Rahim, and H. M. system using IoT,” Materials Today: Proceedings, Jun. 2021, doi:
Kaidi, “Morphological and physiological responses of brassica 10.1016/j.matpr.2021.06.188.
chinensis on different far-red (FR) light treatments using internet- [20] K. Rajguru, P. Tarpe, V. Aswar, K. Bawane, S. Sorte and R.
of-things (IoT) technology,” Agriculture (Switzerland), vol. 11, Agrawal, "Design and Implementation of IoT based sleep
no. 8, Aug. 2021, doi: 10.3390/agriculture11080728. monitoring system for Insomniac people," 2022 Second
[4] R. Hegde, S. Ranjana, and C. D. Divya, “Survey on Development International Conference on Artificial Intelligence and Smart
of Smart Healthcare Monitoring System in IoT Environment,” in Energy (ICAIS), 2022, pp. 1215-1221, doi:
Proceedings - 5th International Conference on Computing 10.1109/ICAIS53314.2022.9742803.
Methodologies and Communication, ICCMC 2021, Apr. 2021, pp. [21] F. Guede-Fernández, V. Ferrer-Mileo, M. Mateu-Mateus, J.
395–399. doi: 10.1109/ICCMC51019.2021.9418405. Ramos Castro, and M. Fernández-Chimeno, “A
[5] K. Anjali, R. Anand, S. D. Prabhu, and R. S. Geethu, “IoT based photoplethysmography smartphone-based method for heart rate
Smart Healthcare System to Detect and Alert Covid Symptom,” in variability assessment: Device model and breathing influences,”
Proceedings of the 6th International Conference on Biomedical Signal Processing and Control, vol. 57, p. 101717,
Communication and Electronics Systems, ICCES 2021, Jul. 2021, 2020
pp. 685–692. doi: 10.1109/ICCES51350.2021.9488952. [22] Chen, H., Chen, A. and Chen, C., 2020. Investigation of the Impact
[6] N. Zholdas, O. Postolache, and M. Mansurova, “Health of Infrared Sensors on Core Body Temperature Monitoring by
Monitoring System Using Internet of Things,” Apr. 2021. doi: Comparing Measurement Sites. Sensors, 20(10), p.2885.
10.1109/SIST50301.2021.9465928.
[7] G. Paré, M. Jaana, and C. Sicotte, “Systematic Review of Home
Telemonitoring for Chronic Diseases: The Evidence Base,”
Journal of the American Medical Informatics Association, vol. 14,
no. 3. pp. 269–277, May 2007. doi: 10.1197/jamia.M2270.
[8] World Health Organization. and Public Health Agency of Canada.,
Preventing chronic diseases : a vital investment. World Health
Organization, 2005.
[9] S. Nookhao, V. Thananant, and T. Khunkhao, “Development of
IoT Heartbeat and Body Temperature Monitoring System for
Community Health Volunteer,” in DAMT & NCON 2020 :
International Conference on Digital Arts, Media and Technology
(DAMT) and ECTI Northern Section Conference on Electrical,
Electronics, Computer and Telecommunications Engineering
(NCON), 2020, pp. 106–109.
[10] C. K. Tseng et al., “Application of a minimized wearable device
combined with SpO2 and ECG sensors to detect stenosis or
occlusion of arteriovenous fistula/graft, progression of
arteriosclerosis and arrhythmia,” in Proceedings - 2018 7th
International Symposium on Next-Generation Electronics, ISNE
2018, Jun. 2018, pp. 1–4. doi: 10.1109/ISNE.2018.8394718.
[11] A. N. N. Chamim, J. Rinaldi, Y. Ardiyanto, I. Iswanto, and A.
Ma’Arif, “Heart Rate and Body Temperature Monitoring Based on
Android Operating System,” in Proceeding - 2020 2nd
International Conference on Industrial Electrical and Electronics,
ICIEE 2020, Oct. 2020, pp. 143–148. doi:
10.1109/ICIEE49813.2020.9276750.
[12] V. Tamilselvi, S. Sribalaji, P. Vigneshwaran, P. Vinu, and J.
Geetha Ramani, “IoT Based Health Monitoring System” in 2020
6th International Conference on Advanced Computing &
Communication Systems (ICACCS), 2020, pp. 386–389.
[13] A. D. Acharya and S. N. Patil, “IoT based Health Care Monitoring
Kit,” in Proceedings of the 4th International Conference on
Computing Methodologies and Communication, ICCMC 2020,
Mar. 2020, pp. 363–368. doi:
10.1109/ICCMC48092.2020.ICCMC-00068.
[14] H. F. Nweke, Y. W. Teh, G. Mujtaba, and M. A. Al-garadi, “Data
fusion and multiple classifier systems for human activity detection
and health monitoring: Review and open research directions,”
Information Fusion, vol. 46, pp. 147–170, Mar. 2019, doi:
10.1016/j.inffus.2018.06.002.
[15] M. F. Ahmed, M. O. Ali, M. H. Rahman, and Y. M. Jang, “Real-
time health monitoring system design based on optical camera
communication,” in International Conference on Information
Networking, Jan. 2021, vol. 2021-January, pp. 870–873. doi:
10.1109/ICOIN50884.2021.9334018.

Authorized licensed use limited to: ANURAG GROUP OF INSTITUTIONS. Downloaded on July 02,2024 at 09:09:17 UTC from IEEE Xplore. Restrictions apply.
172