2023 11th International Conference on Internet of Everything, Microwave Engineering, Communication and Networks (IEMECON) | 978-1-6654-7512-9/23/$31.

00 ©2023 IEEE | DOI: 10.1109/IEMECON56962.2023.10092300

Smart Healthcare Monitoring System

Using IoT Technology
Adapaka Sai Kishore1 Gopiswara Rao Chinni3
Student Scholar Student Scholar
Information Technology Information Technology
V R Siddhartha Engineering College V R Siddhartha Engineering College
[email protected] [email protected]

Dr.G.JayaLakshmi2 Kisari Shyam Kumar Reddy3

Assistant Professor Student Scholar
Information Technology Information Technology V R Siddhartha
Engineering College V R Siddhartha Engineering College
[email protected] [email protected]
ORCID: 0000-0002-4044-4138
contemporary healthcare issue in society. The primary objective
of the initiative was to create a healthcare system. It is divided into
Abstract—Healthcare monitoring systems have grown to be two sections. The initial step is to use sensors to detect the
among the most important systems in recent years, and they patient's vitals. The second portion is the storage of essential body
have become increasingly technological. Humans are information. Python was used to perform the final analysis on
confronting an issue of sudden death owing to numerous those vitals. As a result, the suggested architecture collects sensor
illnesses, which is caused by a lack of timely medical care for data via an Arduino microcontroller, which is then subjected to
patients. The main goal was to create a reliable IoT-based various analysis.
patient monitoring system. Sensors, a data collecting unit, a
microcontroller (i.e., Arduino UNO R3), and software were The fundamental objective of the essay can be summed up as
used to create a healthcare monitoring system that can offer follows:
real- time online information about a patient's physiological  Gathering a patient's real-time medical data using IoT.
status (i.e., Arduino). The patient's temperature, heart rate, and
Spo2 readings are tracked by the system, which also displays  The patient's information is processed and categorized.
and stores them. The data will be processed using the Support
 To analyze and anticipate any disease or ailment in the
Vector Machine method, and graphs will be plotted to show
early stages of development utilizing data mining
whether the conditions are abnormal or normal.
approaches, which will also provide an optimal decision-
making strategy.
Keywords— Arduino UNO, Arduino software, SVM, temperature,
 To provide Internet of Things-based healthcare solutions
heart rate, spo2 sensors.
at any time and from any location.
I. INTRODUCTION
II. RELATED WORK
In the current situation, health care plays a significant role in
society. Many people are dying these days as a result of Numerous researchers have presented a wide range of models for
ineffective medical treatment. To solve this, Smart IoT in Healthcare and the prognosis of different diseases using
Medication could be beneficial to society. There are diverse methodologies. The research that has been done that is
currently several Internet of Things devices that can be used highlighted in this section. To get real-time medical data about a
to watch over a patient's health from a distance. Health patient, one of the papers previously mentioned suggested
practitioners are also using these smart devices to monitor adopting IoT. It also indicates that the patient's information should
their patients. Here, the terminology Smart Health be pre-processed and categorized. Additionally, using data mining
Monitoring Systems have evolved to assist comfortable techniques, it demonstrates how to comprehend and anticipate any
healthy living, more accessible contact between healthcare sickness or disorder at an early stage, simplifying the decision-
providers and patients for close monitoring, routine making process. Another paper explains how to collect data from
consultation, and overall healthy living. Internet of Things the sensors as well as how to use the algorithm to determine the
(IoT) sensors are the only way to do smart health monitoring. patient's state. We looked at a smart health monitoring system
Temperature sensors, pulse sensors, pressure sensors, and using IoT in another research. It will aid in delivering better
blood oxygen sensors are examples of health sensors that diagnostics for people living in remote areas.
collect data from persons who have ever been linked to
them. As a result, this effort ultimately aims address a

978-1-6654-7512-9/23/$31.00 ©2023 IEEE

Authorized licensed use limited to: VIT University. Downloaded on October 22,2024 at 15:56:32 UTC from IEEE Xplore. Restrictions apply.
Ahmad et al. [1] examine the evolution of healthcare, Sahoo et al. [10] examined the system for managing healthcare as
identifying the reasons, methods, and effects. The evolution well as the enormous amount of patient data produced by various
of network services and historical events are provided. reports. Then, in attempt to project the patient's or subject's future
Healthcare evolutions are presented in terms of social
health, they looked at the health parameters. They accomplish this
relevance and scientific significance.
via a large data analytics platform in the cloud that employs
Saranya et al. [2] proposed that the healthcare expert may probabilistic techniques.
forecast and diagnose disease using this approach, which is
useful for subsequent medical therapy. The paper is
Tyagi et al. [11] explored the role of IoT in Healthcare and
concluded with the IoT- based Disease Prediction and
Diagnosis System for Healthcare Using Data Mining researched its technical aspects to make it a reality and find
Technique opportunities for which they propose a cloud-based conceptual
Hamim et al. [3] proposed an android application that could framework in which patients' medical data and information can be
access the database and display a graphical depiction of the securely transferred, with the patient's and their family's
health parameters using Android Studio IoT combined with permission, by establishing a network between patients, hospitals,
health sensors may reduce the requirement for hospital visits doctors, labs, and other entities. The primary motivations for this
for basic care. are to alleviate doctor shortages, release patients from costly
clinical help, and enhance patient care and service.
Prajakta et al. [4] For the measurement and remote
monitoring of human body parameters, a new method based III. SYSTEM ARCHITECTURE
on a wearable wireless sensor network has been developed.
We will attach different health sensors to a person. From the
The suggested architecture will be able to simultaneously sensors we will collect data using Arduino software and will be
test and monitor a number of human body parameters. stored in a csv file using Ardospreadsheet extension. We will
analyze the data and will predict the conditions of the respective
Sudevan et al. [5] developed an application that provides a users as shown in the figure 1
communication interface between the patient and the
caregiver in order to convey real-time data about the patient's
condition to them.

Almotiri et al. [6] provided a template for a regular

framework that ensures a regular examination of specific
wellness limits and forecasts of any disease or disruption
will prevent the patient from regularly visiting emergency
clinics.
Fig. 1. System Architecture of proposed approach
Joyia et al. [7] present an overview of IoT services and
technologies in the healthcare industry A number of research
issues have been highlighted, with key research trends IV. DESIGN APPROACH
projected in the coming years. A variety of use research
In this paper, we propose a system to track a patient's body
benefits have been identified, as well as the most suitable
temperature, heart rate, and spo2. We extend the current method to
application domains.
predict whether the patient has a chronic ailment or disease using
the many health data and additional symptoms obtained by the
Shubham et al. [8] developed Based on the Internet of system.
Things, a health monitoring system was developed. The
device used sensors to track bodily oxygen saturation and
pulse rate, which were also shown on an LCD. These sensor
results are then sent to the app, which, if required, also
creates an IP address. These computed numbers can be
entered into thewebsite to predict heart disease.

Nagavelli and Rao [9] developed a new way for predicting

the severity of an illness from a patient's medical data using a
mining-based statistical methodology dubbed "chance of
disease degree threshold." They modified an algorithm that
is typically used to determine the hyperlink weight of
websitesin order to accomplish their purpose. Fig. 2. Design Methodology

Authorized licensed use limited to: VIT University. Downloaded on October 22,2024 at 15:56:32 UTC from IEEE Xplore. Restrictions apply.
In the above methodology, the Arduino and the body should The various components used are:
both be connected to the sensors. Data that is generated
A. Arduino Board
should be stored in a CSV file. Now load the dataset in
the google colab. Now training and testing the data. The Microchip ATmega328P microprocessor serves as the
Applying the different algorithms on the dataset. Evaluating foundation for the Arduino Uno, a microcontroller board created
the Accuracy with respective algorithms used. Plotting by Arduino.cc. The board features input/output (I/O) pins for
Graphs and determining the body conditions that are normal digital and analogue signals that can be connected to various
or abnormal. expansion boards(shields) and other circuits. This board's
programming language is Arduino.
For analysis of our data, the Support Vector Machine
algorithm is employed. We import the necessary libraries
into the Support Vector Machine Algorithm before importing
the dataset and extracting the x variables and y variables
independently. The dataset should then be split into train and
test groups before the SVM classifier model is started.
Next Making predictions using the SVM classifier model
after fitting it. Finally, evaluate the model’s performance.
Fig.3. Arduino UNO
V. IMPLEMENTATION DETAILS B. Temperature Sensor

Here we implement our proposed system in two phases. A temperature sensor is a temperature measuring tool. We
They are utilize a device that measures body temperature.

Phase 1:
• First, we must install Arduino and ensure that it is
suitable for our system.

• Following Arduino setup, we must attach the

sensors to it before connecting it to our system.

• After opening the Arduino software, we must select

Arduino type using the tools.
Fig. 4. Temperature Sensor (DS18B20 Sensor)
• To obtain data from the sensors, code must be
implemented in Arduino. C. MAX30100 Sensor

• Finally, the data will be shown on the Arduino The MAX30100 is a sensor that identifies both heart rate
Software's Serial Monitor. and pulse oximetry readings. Two LEDs, a photodetector, better
optics, and low-noise analogue signal processing are used to
• We save data in a csv file using Ardo spreadsheet detectthe pulse oximetry and heart rate signals.
after displaying it in Serial Monitor.

Phase 2:
• In this phase the csv file is loaded in google colab and
then we import the required libraries.

• After importing the dataset which is real-time data we

train and test the dataset so that it can be used for
predictions.
Fig. 5. Heart Rate and Spo2
• After that we apply different algorithms and check for Sensor(MAX30100Sensor)
the best accuracy.

• After that this model can be used for predicting the

conditions of a patient.

Authorized licensed use limited to: VIT University. Downloaded on October 22,2024 at 15:56:32 UTC from IEEE Xplore. Restrictions apply.
 VI. HARDWARE CONNECTIONS C. Graph Results
First, we connected the sensors to the Arduino using jumper
wires and a breadboard. We used a resistor also. It can be The data will be analyzed using SVM algorithm and the
accuracy achieved through SVM is high i.e., 96.72 against
easily used anywhere and it is easily carriable. By just placing
alternative algorithms and the graphs displays the maximum
the finger we can get the data from the sensors and it is easily possible value in the dataset by which the we will determine
accessible. whether user is normal or abnormal.

Fig. 6. Hardware Connections

Fig. 9. Boxplot Results
VII. RESULTS
The error rates are shown in below table for the developed system
from actual data and observed data.
A. Serial Monitor Results
Table1: Heart rate & Body temperature error rates for actual and
The proposed system used sensors attached to the patient observed data
to capture real-time physiological indicators like as
temperature, heart rate, and spo2. Subjects Error Error Rate
Rate for for Body
Heart Temperature
Rate
S1 1.48 0.52
S2 4.31 0.71
S3 4.06 0.49
S4 2.56 0.61
S5 1.37 0.42

The error rate of the system for heart rate and body temperature is
depicted in below figure 10

Fig. 7. Serial Monitor Results 5

4 Error Rate
B. Storing the Data for Heart
3 Rate

The data which is collected from the sensors will be 2

stored in a csv file using Ardo Spreadsheet. Error Rate
1 for Body
Temperatur
0 e
S1 S2 S3 S4 S5

Fig. 10. The error rate of the system

Fig. 8. Storing the data

Authorized licensed use limited to: VIT University. Downloaded on October 22,2024 at 15:56:32 UTC from IEEE Xplore. Restrictions apply.
 VIII. COMPARITIVE ANALYSIS [5] Smiju Sudevan, Mani Joseph, “Internet of Things:
Incorporation into HealthcareMonitoring”, IEEE, 2019.

[6] S.H. Almotiri, M. A. Khan, and M. A. Alghamdi. Mobile

health (m-health) system in the context of iot. In 2016IEEE
4th International Conference on Future Internet of Things and
Cloud Workshops (FiCloudW), pages 39–42, Aug 2016.

[7] Gulraiz Joyia, Rao M. Liaqat, Aftab Farooq, and

SaadRehman, Internet of Medical Things (IOMT):
Applications, Benefits and Future Challenges in Healthcare
Domain, Journal of CommunicationsVol. 12, No. 4, April
2017.
We used Support Vector Machine algorithm for our
[8] Shubham Banka, IshaMadan and S.S. Saranya, Smart
prediction which has achieved the highest accuracy of 96.72
Healthcare Monitoring using IoT.International Journal of
when compared to other algorithms. Applied Engineering Research ISSN 0973-4562 Volume 13,
Number 15,pp. 11984-11989, 2018.

IX. CONCLUSION AND FUTURE WORK

[9] R. Nagavelli and C. V. Guru Rao. Degree of disease
The primary objective of our paper is that the patient possibility (ddp): A mining based statistical measuring
can’t visit the doctor directly. So, our project is very helpful approach for disease prediction in health care data mining. In
in this pandemic situation. Arduino is used to powering the International Conference on Recent Advances and
entire system. Even if the patient is at home, their health state Innovations in Engineering (ICRAIE-2014), pages 1–6, May
can be tracked in real-time using various sensors, and we can 2014.
determine whether they are normal or abnormal. It is also able
[10] P. K. Sahoo, S. K. Mohapatra, and S. L. Wu. Analyzing
to know the patient’s health situation without consulting of healthcare big data with prediction for future health
doctor physically. condition. IEEE Access, 4:9786–9799, 2016. ISSN 2169-
3536.
The future study of this project is that based on this
system we can develop a real-time hardware component [11] S. Tyagi, A. Agarwal, and P. Maheshwari. A conceptual
where everyone can use it and can easily access the framework for iot-based healthcare system using cloud
component and get the health information continuously and computing. In 2016 6th International Conference - Cloud
System and Big Data Engineering (Confluence), pages 503–
can monitor their health also, we can develop a software app
507, Jan 2016.
which is either mobile or web application.

REFERENCES

[1] Emerging trends and evolutions for smart city healthcare

systems Mohamad nazir, Khairol Amali Bin Ahmad,
2022.

[2] Saranya. E, Maheswaran. T, “IoT Based Disease

Prediction and Diagnosis System for Healthcare”,
International Journal of Engineering Development and
Research (IJEDR) | Volume 7, Issue 2 | ISSN: 2321-
9939, pp.232-237, 2019.

[3] M. Hamim, S. Paul, S. I. Hoque, M. N. Rahman and I.

Baqee, "IoT Based Remote Health Monitoring System
for Patients and Elderly People", 2019 International
Conference on RoboticsElectrical and Signal Processing
Techniques (ICREST), pp. 533- 538, 2019.

[4] M.M.E. Mahmoud et al., “Enabling technologies on a

cloud of things for smart healthcare,” IEEE Access, vol.
6, no. c, pp. 31950-31967, 2018.

Authorized licensed use limited to: VIT University. Downloaded on October 22,2024 at 15:56:32 UTC from IEEE Xplore. Restrictions apply.