International Conference on Smart Computing and Electronic Enterprise.

(ICSCEE2018) ©2018 IEEE

Integrated Health Monitoring System using GSM and

IoT
Suliman Mohamed Fati Amgad Muneer Dheeraj Mungur
Faulty of Information Technology and School of Mechatronics Engineering School of Mechatronics Engineering
Science, INTI International University, Asia Pacific University of Tech. & Innovation, Asia Pacific University of Tech. & Innov.
Nilai, Negeri Sembilan, 43300, Malaysia Kuala Lumpur, Malaysia Kuala Lumpur, Malaysia
[email protected] [email protected] [email protected]

Ahmad Badawi
School of Mechatronics Engineering
Asia Pacific University of Tech. & Innovation,
Kuala Lumpur, Malaysia
[email protected]

Abstract— This paper proposes a health monitoring system It is very tough job for the paramedical staff to continuously
for the patient in a coma based on GSM and Internet of Things monitor each patient’s 24 hours since the proportion of staff to
(IoT). In this proposed system, the four health parameters patient is very low. So it is not an easy task to monitor each and
implemented are LM35 temperature sensor, heartbeat sensor, every patient regularly. According to statistics provided by
accelerometer sensor and eye blink sensor. Using these four World Health Organization (WHO) 2015 about 6.9 million
parameters simultaneously, it has been monitored the coma children under 5 die from treatable and preventable diseases
patient’s condition. Hence ruling out the use of the thermometer each year due to lack of number of health workers as well as
and other devices to check the condition of the patient. about - 55% of 57 different countries face health workforce
Consequently, there is no need for a lot of clinical staff nor
crisis. So it is important and necessary to propose health
accompanying persons to be physically present to check the
condition of the patient because there is GSM module to send
monitoring systems for a coma patient.
SMS message to the mobile phone of the person in charge Such healthcare systems are needed to continuously
(nurses/doctor) in case there is any abnormalities in health monitor and record all the vital information of a particular
parameters. Moreover, this system uses Wi-Fi for IoT wherein subject by maintaining all the records of that comatose
“Thingspeak” is used to monitor the coma patient online via manually. Patient Monitoring Systems are systems wherein a
mobile phone. Alternatively, the patient’s relative can log in surgeon can continuously monitor more than one patient, for
easily to the system to check the patient statues. The IoT shows
more than one parameter at a time in a remote place [3]. The
the results online of body temperature, heartbeat, body
technical brilliance and development in different fields has led
movement and eye blink along with the time and the date for
each parameter. Additionally, the results achieved for the body
to a drastic change in our lives, one among them is embedded
temperature and the heartbeat are 36Co and 87 BPM systems and telecommunications. The advances in information
respectively. Finally, the results obtained by the MPU-6050 and communication technologies enable technically, the
gyroscope and the eye blink sensor were very satisfactory. continuous monitoring of health related parameters with
Keywords—GSM, Internet of Things, health monitoring system, wireless sensors, wherever the user happens to be. They
body sensors, coma patient. provide valuable real time information enabling the physicians
to monitor and analyze a patient's current and previous state of
I. INTRODUCTION health. However, the current systems for this monitoring are
Coma or unconsciousness is a state of a person where an prohibitively expensive and can be only found at certain
individual cannot respond to the internal and external stimulus hospitals as well as the systems used in hospitals are complex
[1]. Coma is a deep state of unconsciousness. The coma patient and only certain people can use them.
is a person whose heart is still beating however he/she is Furthermore, in the existing healthcare systems, the
unaware of his surrounding environment and has no physical medical world today faces two basic problems when it comes
control over his entire body. Such cases require a serious to patient monitoring, firstly the need of health care providers
attention and continuous monitoring to save the patients’ lives. present bedside the patient and secondly the patient is restricted
Nowadays, having someone to watch critically ill person is to bed and wired to large machines. In order to achieve better
very expensive and takes a lot of manpower. Besides, such a quality patient care, the above cited problems have to be
continuous supervision by a paramedical assistant are error solved. As the technologies are advancing it has become
prone and may lead to difficulties due to human error. In case feasible to propose health monitoring system based on GSM.
of critically ill patients it requires to measure the vital Lastly, according to aforementioned challenges it helpful and
parameters at least for every 15 seconds until the patient’s necessary to proposed system that can eliminate the burden of
condition stabilizes. Furthermore, [2] state that monitoring of continuous supervision and send SMS message to the doctor or
coma patients is different from monitoring the normal patients. person in charge only when attention is needed. Therefore, the
 International Conference on Smart Computing and Electronic Enterprise. (ICSCEE2018) ©2018 IEEE

aim of this paper is to develop a prototype at a low cost that initiate the communication between the computer and the GSM
monitors the health condition of a critically ill patient/person in module. The GSM modem is a modem that required a SIM too
a coma. This aim can be achieved by the following objectives: work. It is similar like a mobile phone except it is assembled
together with the power supply circuit. They also have a unique
 Design an appropriate circuit for GSM patient health IMEI for identification. The GSM module will be used to send
monitoring system. messages to the recipient needed when the input parameters
 Test, implement and establish communication between exceeded the conditions set.
the microcontroller, sensors, and GSM. Many critically ill patients died every day because they did
 Develop a GUI for live monitoring of the heart rate and not get help in a timely manner. These patients require
temperature of the patient and physical body spontaneous and accurate decision so that their life can be
movements. saved. [7] Designed a health monitoring system which is
energy efficient and reliable. Doctors are able to monitor the
 Develop a system that is able to send SMS alert parameters of the patient in real time like temperature, position
messages in case of emergency. and electro cardio gram. The proposed idea of the author would
be useful for health care providers to improve diseases
 Develop a system that can monitor patient condition
management. The system would also be beneficial for medical
online and store data on cloud.
practitioners as it would provide better treatment and better
This proposed system will be helpful in assisting the doctor monitoring. As transferring data via cable is a hassle, the
about the health condition of the unconscious patient and ZigBee is introduced to the system as it is energy efficient
alerting the doctor whenever care is required. The proposed compared the Bluetooth. It also provides higher number of
system will assist the doctor by giving an alarm and send SMS nodes and better network flexibility. This system can be
message about the health condition of the patient, when the set enhanced by the integration of GPS. The location of the remote
of vital signals recorded are out of the normal range. This patient can be detected and can provide nearest emergency
system is beneficial to relatives or guardians that are taking support. Other sensors could be introduced to the system to add
care of the patient. Where all the parameters results are more parameters that be monitored such as blood pressure,
displayed on the mobile application or in special website, in Galvanic – Skin Resistance Amenia and pulse oximeter. For
order to let the coma patient’s family to check their relative patients suffering from asthma, the SY-HS-220 sensor which is
patient online without needing to stay in the hospital or to call used to measure humidity could be used.
to the doctor for asking about their relative condition.
In the biomedical patient monitoring, the Internet of Things
The remainder of this paper is organized as follows: section is the latest trend. [8] Proposed a cardiac alerting system using
2 investigates the literature review of health monitoring IoT. The vital signs can be measured automatically through
systems, section 3 presents the system design and sensors and these data can be easily sent to a clod using Wi-Fi.
implementation of the proposed system, section 4 discusses the Such system can be made available to run 24/7 and 365 days a
findings and results, and finally section 5 concludes the paper. year. These data can be easily transmitted and analyzed using
particular software. Author’s previous work consisted of a
II. LITERATURE REVIEW AND RELATED WORKS wireless patient monitoring system which consisted of a GSM
enabling short message services, circuit switched data or
In the literature, there are several approach proposed to general packet radio service. These could provide a data rate
develop health monitoring system. For example, [4] have used ranging from 9.6kb/s to 171.2kb/s. A Bluetooth module was
in GSM based health monitoring system a method that used also used providing up to 4 channels at a rate up to 295 Hz.
GSM network and Very Large Scale Integration (VLSI) This system was not power efficient. The system was designed
technology. Another work by [5] which investigate about the in a way such that it will perform at a low power and data rate
effects seen in the patient using “image processing based coma is higher compared to previous work. [9] Proposed a heath
patient monitoring system with feedback” which is a very monitoring system for fitness related program, monitoring
advanced project related to physical changes in body chronic diseases and for elderly persons. The system will
movement of the patient and It also provide a system for monitor health parameters such as hemoglobin count, sugar
monitoring patient heartbeat rate and gives warning in form of level, blood cell count, pulse and temperature. These data are
alarm & LCD Display. extracted from respective sensors and are sent to the local
[6] Proposed to monitor the health of the patients using server using either Bluetooth or ZigBee. The health parameters
GSM as it will remove the human errors and diminish the are sent to the cloud via the ESP8266 Wi-Fi module. These
excessive burden of constantly monitoring them. With the use data can be monitored using internet or the application named
of sensors such as LM35 for temperature and pulse rate “Thing speak”.
integrated with a microcontroller, the monitoring will be Fig. 1 shows the block diagram for the IoT system. The
automatic. The health monitoring system is proposed to ESP8266 enables the Wi-Fi communication between the
provide alert for the staff when conditions are met. This will sensors and the cloud. When the module is set as host mode, it
diminish the rate of hospitalization of critically ill patients. For can boot up the application processor directly from an external
improvement, the serial output can be attached to the device so flash. To improve the performance of the system in
that the heart rate of the patient can be sent to the pc for either applications and to minimize memory, a cache is integrated.
online or offline analysis. [4] Used the GSM module to be
 International Conference on Smart Computing and Electronic Enterprise. (ICSCEE2018) ©2018 IEEE

The device also includes features such as sleep/wake switching implements the four health parameters which are LM35
which makes it power efficient. The cardiac arrest warning temperature sensor, heartbeat sensor, accelerometer sensor and
system consists of a sensor which measures the heart beat at a eye blink sensor together. Besides, the GSM module and
constant time interval. This can be integrated with the alerting Internet of Things (IoT) they did not implement together in any
system which can save a person’s life immediately. The cardiac project. Thus, we introduce an integrated health monitoring
alerting system consists of a microcontroller (Arduino), a system that considers all these features.
GSM/GSPRS module and the ESP8266 is used for the
transmission of the sensed values to be monitored online. The III. SYSTEM DESIGN AND IMPLEMENTATION
readings is done from all sensors and then are transferred to the
cloud using TCP IP protocols. In case the person got a heart A. Proposed methodology
attack, a message will be sent to the doctor or fried or relatives.
The Wi-Fi module will keep the details about the rate of the This proposed system consists of GSM, IoT, heartbeat,
heartbeat. This will enable the doctor or the monitoring staff to temperature, and eye blink and accelerometer sensor. The
see when the patient goes the heart attack. The rate will be heartbeat sensor used to monitor the pulse rate of the patient.
displayed as well as the time thus the doctor will know the The LM35 sensor measured the body temperature. The
exact time when the patient experienced the heart attack. accelerometer sensor is used to sense changes in the patient and
it placed on the hand of the patient. The Eye blink is self-made
The authors in [10] proposed a health monitoring system sensor and developed using image processing techniques in
integrated with IoT and with Matlab analysis. The system was order to detect any blinks in a comatose. For the eye blink a
implemented using an open API IoT platform known as Thing camera placed on the bed and it focused only on the patient
peak. With the help of an internet connection acting as a carrier face. Additionally, IoT used to monitor a coma patient
between the connected ‘things’ and the cloud of things peak, condition online and store data on cloud. Lastly, process the
data can be stored, analyzed and displayed in forms of graphs. result and display to a LCD. In case of any abnormal
The author used sensors such as light dependent resistor, air parameters, SMS will be sent to the medical staff and
quality sensor, rain sensor, pressure sensor and humidity and respective relatives.
temperature sensor. These sensors were used for IoT which
will be transmitted using the ESP8266 Wi-Fi module.

Fig. 2. Proposed methodology

B. Flow chart
As depicted in Fig. 3, the general flow chart of the
proposed health monitoring system, all four sensors are used to
acquire various different health parameters such as beats per
minute of the heart, body temperature, pitch and roll of the
coma patient in case he is awake and does any physical
Fig. 1. Architecture of IoT HMS movement as well as an eye blink detection sensor to detect
whether the person has blinked or opened his eyes while under
The Wi-Fi module can either host the function or to divest monitoring.
every Wi-Fi networking function. When programming of the
system is done, it can just be booted from an external flash. All four sensors’ output values are then checked whether
[11] Proposed a monitoring system using HTTP and MQTT they are above normal preset conditions or not. In case of any
protocols. This enable lot of functions such as a web platform abnormality in any of the health parameters of the patient the
can be created and all data can be sent and displayed there. microcontroller sends an alert command to the GSM module
Login with verification can be implemented also and using the which triggers it to start sending messages to the relevant
HTTP protocols, it enables the transmission to mobile phones person in charge/ doctor or guardian.
easier. The system implemented by the author consists of The implemented design is shown below in Fig. 4. The
Wemos D1 Mini and ESP8266 Wifi Module, DHT11 black box contains the central processing unit which is the
Temperature and Humidity sensor and an android smartphone. Arduino Uno along with the temperature sensor, pulse sensor
In summary, most of authors have proposed of health and the accelerometer.
monitoring system for the patient in a coma. But, no one
 International Conference on Smart Computing and Electronic Enterprise. (ICSCEE2018) ©2018 IEEE

The exact length is sent and the value also. Once received on
the cloud, it will be plotted at that moment. The next sensed
data will be then plotted with respect to time. The frequency of
transmitting the data is set to be one minute. Two conditions
were set for the temperature sensor. Firstly, if the temperature
is more than 37°C, it means that the coma patient is either
experiencing fever or is in a critical condition thus an alert will
be sent via message to the doctor. The doctor will be able to
see the temperature and the pulse also via the online
monitoring system. If the temperature of the person is below
the set value, it means that the patient is also in a critical
condition and might be dying. Alerts will be sent to the
responsible party and the doctor also.
For the pulse sensor data to be able to upload on the
Thingspeak’s cloud, the API key was changed accordingly to
the respective channel. The establishment of connection was
similarly done as for the temperature one. The output of the
Fig. 3. Flow chart of the health monitoring system sensor will be converted into heart beat per minute (BPM). The
BPM will be sent via the TCP IP protocol along with AT
The GSM module and the WiFi module are also placed commands. The length of the data will be sent along with its
inside. Initially the accelerometer is placed at a specific values that was sensed previously. If the transmission of the
position as it is needed to be horizontal for calibration. This is data has not been successfully transmitted, it will do this step
because when the box will be tilt in any direction, it will trigger again due to the introduced loop. In case any abnormal BPM
the buzzer for demonstration purpose. All sensors and modules detected i.e. if it is more than the set value. It will trigger a
were connected together on PCB board with the Arduino. For message to the responsible party to alert them that the patient
demonstration purpose, 2 holes were made so that the sensors has woke up from the coma or the patient is in a critical
were made available for testing. At the top compartment, a condition. Two condition were set, firstly if it is more than the
LCD screen was placed. While it is running, it will show the normal rate, it will send an alert to the doctor alerting him of
BPM, pulse rate and temperature of the sensors. In case any the issue and if the BPM is lower than the normal value, it will
abnormal parameters were detected, it will trigger the SMS send the message to the responsible party and the doctor also.
alert. Finally, all the sensed value was sent to the cloud of Both doctor and responsible party will be able to see the
Thingspeak for IoT and can be monitored anywhere using the temperature and the BPM of the patient on the online
private channel authorization. monitoring system and in case of any issue, they will know
when exactly it occurred.
The implemented health monitoring system is made for
coma patient and for critically ill patient also consisting of
paralysis issues. These patients need to be rested on the bed
where they will be monitored 24/7. It can be used either in a
clinical environment or in home based. The three vital signs
which are essential for monitoring are pulse, temperature and
the eye blink. Thus, the temperature sensor will be placed on
the index finger and the pulse on the middle finger. These will
be placed and kept secured and stable by using Velcro or a
glove which will stop light going through it affecting the pulse
LED and will keep the temperature reading more accurate for
the LM35. The gyroscope will be placed on the wrist. In case
Fig. 4. Final hardware design the critically ill or the coma patient moves, a buzzer will be
triggered and a message will be sent accordingly to the doctor
After the temperature has been sensed by the LM35 and and patient’s members. The entire system is autonomous hence
processed by the Arduino, the ESP8266 will send the data the reduction of clinical staff is decreased and if using at home,
online. For this to happen, the WiFi module need to be the need of physical presence for monitoring is lowered also.
programmed. Firstly, it is connected to the power supply which The cost of staff is decreased exponentially. In case of any
is 3.3 V. Then it was connected to a WiFi network to be able to abnormal parameters are detected, such as if the temperature or
access the internet by using AT commands. The operation pulse sensor are above the threshold value, a SMS will be sent
mode was set to “3” i.e. standalone + access point and multiple to the doctor notifying him of the issue and if it is below the
data will be sent at the same time. The TCP IP command was threshold value, the patient lost his/her life, SMS will be sent to
entered to be able to do a handshake with the cloud service. the doctor and the patient’s members. A live monitoring system
Then the “GET” command was initiated with the API key and is also implemented using IoT. Data parameters such as pulse
with the field on which it will be sent and plotted on the cloud. and temperature are exported to Thingspeak’s cloud via the
 International Conference on Smart Computing and Electronic Enterprise. (ICSCEE2018) ©2018 IEEE

ESP8266 module. For example, if the doctor has received an Fig. 7 shows the BPM obtained through experiments and
alert and if he is not around, he will be able to see the health it’s clearly shown that the raw signal is varying however as the
status of the patient via his mobile phone. In case the patient average IBI is calculated, the BPM value calculated is said to
passed away, it will be able to know when exactly it happened. be fairly accurate.
The data exported to the cloud is encrypted and secured by the
domain of the cloud website hence providing security of data
and also for privacy concerns, these data will be able to be
viewed to people with the login access of the channel.
Fig. 5 shows the main Graphical User Interface (GUI) that
was developed using MATLAB program to show all the results
as well as allow the user to easily monitor the patient’s health
parameters in real time. The GUI has two graphs which
represents the two most critical health parameters mainly;
temperature and heart rate. It also shows the current exact
values of temperate and BPM below the graph as well as the
condition of the patient.

Fig. 7. BPM & IBI output result

Fig. 5. Main Matlab GUI

IV. EXPERIMENTAL RESULT AND DISCUSSION

This section explains the result analysis for each health
parameters and the overall result of the systems in MATLAB
GUI. Fig. 6 shows the heart rate signal that is processed after
its being acquired from the sensor and then plotted through
Ardunio serial monitor.
Fig. 8. Heart rate graph

Fig. 9 shows the roll and pitch results obtained by the

MPU-6050 gyroscope for monitoring physical
movements. As the gyroscope is placed directly on the
patient’s hand, pitch and roll values vary according to the
physical hand movement of the patient. A physical
movement is detected once those values exceed certain
preset threshold which is usually set after several testing
that takes into consideration the uncertainty of the sensor
itself to avoid errors. In this case, once the sensor’s roll or
pitch value exceeds the threshold of 5 degree/second, a
movement is detected. The sensor also triggers a buzzer
once a movement is detected to alert the doctor or the
person in charge. The preset threshold can also be
Fig. 6. BPM & IBI graph
decreased in order to detect slight physical movements of
 International Conference on Smart Computing and Electronic Enterprise. (ICSCEE2018) ©2018 IEEE

the patient however, that would result in higher chances

of errors.

Fig. 11. GUI result using MATLAB

Fig. 9. Movement detection

Body temperature vary from one part of the body to

another. Most suitable type of temperature measurement at
mouth is at 37.0 °C. 37.0 °C is the normal core body
temperature. Axillary temperature degrees is an external
measurement taken in the armpit or between two sides of the
skin in the body. This is the longest and most accurate
measurement of body temperature, and decreases the normal
temperature to 97.6 degrees Fahrenheit or 36.4 degrees Celsius.
Straight temperature degrees are an internal measurement taken
Fig. 12. Response when eye opened
in the rectum, which is at 99.6 degrees Fahrenheit or 37.6
degrees Celsius. This is less time consuming and more accurate Fig. 12 shows the result of the eye blink sensor when the
than the body temperature measurement, being the internal eye is open which the graph is stable. The x-axis of the graph
measurement. But certainly, to a large extent, it is not the most represent the number of the frames used in the live video and
comfortable way to measure body temperature. y-axis represent the eye blink, from the live video the system
In this project, the LM35 sensor was placed on the arm beat capture image of the eye area only. Therefore, the graph will be
in order to get accurate result and it’s difficult to put the sensor plot if the is any detection for the open eyes. Once the sensor
in the mouth to get the actual temperature of the body. So, the detects any blink the graph will change immediately. Then, the
result obtain for the body temperature is in range of 32.20 and system will send SMS message to the doctor, nurses or person
33.69. The temperature result via serial monitor and via GUI in in charge that the person in a comatose blink his/her eyes.
MATLAB given by Fig. 10 and Fig. 11 respectively.

Fig. 13. Pulse Sensor IoT Result

Fig. 10. Temperature result via Ardunio
 International Conference on Smart Computing and Electronic Enterprise. (ICSCEE2018) ©2018 IEEE

Fig. 13 and Fig. 14 shows the result of the IoT data for the V. CONCLUSION
pulse sensor and the temperature respectively. The sensed data In this paper, we proposed a health monitoring system to
was successfully sent to the cloud of Things-peak. As seen in monitor the coma patients using GSM and Internet of Things
the Figures, the sensed data is plotted with respect to time. (IoT). In this proposed system, the vital parameters such as
When the system is not in use, it will not plot anything pulse rate, eye blink, accelerometer, and temperature are
therefore when the next value is plotted, it will join the last monitored and recorded in cloud storage. The proposed system
value with the new sensed one with respect to time. This analyzes the parameters values. In case these values are
applies for the temperature sensor also as seen below. abnormal, GSM module will be triggered to send SMS
Whenever an abnormal condition is detected such as if the message to predefined phone number. The overall accuracy of
temperature or the pulse sensor is above or below the referred the system is above 90% in different conditions. However,
value, it will trigger the alert message thus respective messages some factors affected the overall accuracy of the project.
will be sent to the doctor and the patient’s members. Different lighting conditions resulted in different output
readings of the eye blink sensor as it mainly depends on image
processing. However, we are seeking to enhance the overall
accuracy of the project in the future.
Firstly, better algorithms can be implemented such that it
allows the eye blink sensor to maintain its accuracy throughout
various lighting conditions. Secondly, multiple accelerometers
can be attached on different parts of the body to better monitor
the physical body movements. Last but not least, a more
accurate temperature sensor can be used to obtain accurate
body temperature readings.

REFERENCES
Fig. 14. Temperature Result for IoT [1] Kansal, N. and Dhillon, H.S. Advanced Coma Patient Monitoring
System. International Journal of Scientific & Engineering
Research, 2(6). 2011.
[2] Sneha Chowdary Kogant, Dr. H N Suma, Appaji M. Abhishek. Analysis
and Monitoring of Coma Patients using Wearable Motion Sensor
System. International Journal of Science and Research (IJSR). 4(6).
2015.
[3] Jay Patel., Ramsinh Chavda., Megha Christian., Shweta Patel and
Ramanuj Gupta. Image Processing Based Coma Patient Monitoring
System With Feedback. International Journal of Recent Scientific
Research.7 (2), pp. 8885-8888. 2016.
[4] Adivarekar, J.S., Chordia, A.D., Baviskar, H.H., Aher, P.V. And Gupta,
S. Patient Monitoring System Using GSM Technology. International
Journal of Mathematics and Research, 1(2). 2013.
[5] Pimplaskar, D., Nagmode, M.S. And Borkar, A., Real Time Eye
Blinking Detection and Tracking Using Opencv. International Journal of
Engineering Research and Applications. 13(14), p.15. 2015
[6] Chandra, R. et al. Gsm based health monitoring system 1’, pp. 62–
65.2014.
[7] Purnima, P. S. Zigbee and GSM based patient health monitoring
Fig. 15. Final GUI for live monitoring system’, International Conference on Electronics and Communication
Systems (ICECS), pp. 1–5. 2014.
[8] Ajay, H. A Novel Cardiac Arrest Alerting System using IOT’, 3(10), pp.
78–83. 2017.
[9] Govinda. K, Shaik Shama Zabeen, Alapati Saijagathi, R. International
Journal of Pharma and Bio Sciences ISSN.7(3), pp. 626–632. 2017.
[10] Pasha, S.Thingspeak Based Sensing and Monitoring System for IoT with
Matlab Analysis’, (6), pp. 19–23. 2016.
[11] Gan, G. IoT real time data acquisition using MQTT protocol. 2016.

Fig. 16. Final GUI for live monitoring