See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/322542647

Iot Patient Health Monitoring System

Article  in  Indian Journal of Public Health Research and Development · October 2017

DOI: 10.5958/0976-5506.2017.00519.8

CITATION READS
1 7,098

4 authors, including:

Shola usha rani

VIT University
7 PUBLICATIONS   6 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Shola usha rani on 05 January 2019.

The user has requested enhancement of the downloaded file.

 Iot Patient Health Monitoring System

Shola Usha Rani1, Antony Ignatious2, Bhava Vyasa Hari2, Balavishnu V J2

1
[AP (Sr)], 2Student, SCSE, Vellore Institute of Technology, Chennai Campus, Chennai, India.

Abstract
The increased use of mobile technologies and smart devices in the health zone has brought on extraordinary
effect on the world’s critical care. Health specialists and doctors are using these technologies to create
critical change in medicinal services during clinical settings. Likewise, many users are being served from
the upsides of the M-Health (Mobile Health) applications and E-Health (social insurance upheld by ICT)
to enhance, help and assist their well-being. The Internet of things is progressively permitting to coordinate
gadgets fit for associating with the Internet and give data on the condition of health of patients and give
data continuously to specialists who help. The main aim of this ‘Patient Monitoring System’ is to build up
a system fit for observing vital body signs, for example, body temperature, heart rate, pulse oximetry. The
System is additionally equipped for fall detection and sleep pattern analysis. To accomplish this, the system
involves many sensors to screen fundamental signs that can be interfaced to the doctor’s mobile or the web.
The gadget will exchange the readings from the sensor to cloud remotely and the information gathered will
be accessible for analysis progressively. It has the capacity of reading and transmitting emergency signs
to the cloud and then to doctor’s web portal or to Doctor’s Smartphone. These readings can be utilized to
recognize the health state of the patient and as an alert system against the emergency health condition.

Keywords: IOT, Raspberry pi, AWT cloud, Patient Monitoring.

Introduction Patient monitoring is not another new framework in

medicinal services as it was first begun in the year 1625
Patient Monitoring System can be characterized as for checking the body temperature and pulse of patients.
the system utilized for observing physiological signs that Subsequently, this framework has started to discover
incorporate the parameters like the electrocardiogram its utilization and acknowledgement for checking
(ECG), respiratory signs, intrusive and noninvasive blood diverse sorts of physiological parameters and health-
pressure body temperature, gases related parameters, related angles that are being performed [1] as of not long
and so forth. Understanding and checking monitoring ago. These days’ patient monitoring frameworks are
system is a piece of M-health innovation. It can be accessible in two structures:
named as m-health or mobile health. These systems are
utilized for the practice of medicinal and general health zz Single-parameter monitoring system: This
with the assistance of cell phones. These frameworks system is utilized for measuring the blood pressure
observation can be utilized nearby or remotely. Patient of a human body, observing ECG, checking SPO2
monitoring is relevant in various circumstances when a (oxygen level in the blood), etc.
patient is in the accompanying conditions: zz Multi parameter monitoring system: This
zz In unstable physiological regulatory systems – for system is utilized for checking different crucial
instance, in the case of overdose of anesthesia. physiological indications of patients by transmitting
the fundamental data like ECG, breath rate and
zz In a life-threatening condition – for instance, when blood pressure, and so on. Because of these reasons,
there is an indication of heart attack in a patient. multi-parameter observing system holds a huge
zz In a situation leading to the developing of a risky part in the field of medicinal devices.
life-threatening condition.
These days, the health care sensors are playing a
zz In a critical physiological state. fundamental part in hospitals. The patient checking
 1330 Indian Journal of Public Health Research & Development, October-December 2017, Vol.8, No. 4

monitoring is one of the significant improvements turned out to be very shabby. The system can likewise
as a result of its creative innovation. A programmed be made IoT (Internet of Things) empowered and M2M
remote health observing system is utilized to quantify (Machine To Machine) is good. This system, usage
patient’s body temperature, pulse by utilizing implanted of such a healthcare checking system is displayed.
innovation. The proposed system utilizes sensors Thus, this will possibly profit an extensive population.
like pulse sensor, oximeter, temperature sensor, For the healthcare checking system to be solid, every
accelerometer and gyroscope. These sensors mostly sensor should timely measure the information taking the
include in observing the health condition, fall detection recommended examining rate of the parameter, and the
and sleep pattern of the patient. information should be sent to the data processor with
no overlap. Every sensor has fluctuating necessities
Background regarding information length or size and examining
rate the sensor information gathered without overlap by
A large portion of the developing nations have information processor can replace notepad at patient’s
extremely poor healthcare foundation there are not very bed with smart gadget and patient’s information can be
many clinics [2] in contrast with blasting population. accessed to from specialist’s Smartphone or web.
Few of doctor’s facilities are deficiently prepared
where very less number of specialists is available. The Existing System:
basic diagnostic equipment for the diagnosis of life- zz In a hospital, either the nurse or the doctor has to
threatening diseases is absent. In the event that this paper move physically from one person to another for
could fabricate an ease compact health detecting gadget, health check, which may not be possible to monitor
involving a few sensors, equipped for measuring the their conditions continuously. Thus, any critical
vital attributes of a human body, and can speak with the situations cannot be found easily unless the nurse
doctor’s facility database, it could furnish with quality or doctor checks the person’s health at that moment.
therapeutic guidance. The restorative administration is This may be a strain for the doctors who have to
given after one of the authority specialists from a group take care of a lot number of people in the hospital.
of particular specialists display everywhere throughout Also, when medical emergencies happen to the
the globe assesses those health parameters on the clinic’s patient, they are often unconscious and unable to
database. press an Emergency Alert Button.
In today’s social protection system for patients who zz One of the application protocols that are being used
remains in home amid post operational days checking to transfer data is Hyper Text Transfer Protocol
is done either by means of administrator/medicinal (HTTP) for general communication over Internet.
guardian. Endless watching may not be expected by this However, when HTTP is applied to communication
system, in light of the fact that anything can change in in IOT, protocol overhead and resulting performance
prosperity parameter within some fraction of seconds degradation are a serious problem. Moreover, IP
and in the midst of that time if the specialist is not in addressing depends on physical location, which
the premises causes more important damage. So with causes the problem of complexity of network control.
this advancement made period where the web directs the
world gives an idea to add to doctors from a group of Proposed System
specialized doctors present all over the globe [3] where zz Our system continuously monitors patient’s vital
time to time consistent checking of the patient is refined. signs and sense abnormalities. The monitored data
is delivered to medical staff. Upon encountering
Also, if the health detecting gadget is made to speak
abnormalities, the system alerts the medical staff
with a compact system like a tab or a cell phone which
about the abnormal parameter. Thus, reduces the need
has the default capacity of speaking with Cloud (hospital
for manual monitoring done by the medical staff.
or clinic database), then the entire system would be
considerably more financially effective. This is on the zz Our proposed system uses MQTT client to send
grounds that these days a great many people have entry data from sensors to cloud platform. It is a publish/
to versatile specialized devices and these devices have subscribe, extremely simple and lightweight
 Indian Journal of Public Health Research & Development, October-December 2017, Vol.8, No. 4 1331

messaging protocol, designed for constrained can be interpreted by aligning the sensors. These
devices and low-bandwidth, high-latency or readings are transmitted to the AWS IOT through the
unreliable networks. The design principles are to MQTT which can be utilized by both Smartphone and
minimize network bandwidth and device resource analytics module.
requirements whilst also attempting to ensure Hardware Description: The hardware part of the project
reliability and some degree of assurance of delivery. involves the Raspberry Pi 3 Model B, MAX30100,
MPU-6050. The two sensors are connected to the Pi via
Monitoring System Description the I2C interface. The sensor values are read by the Pi,
This paper proposes a model of Patient Health processed, and then sent to the AWS IOT server using
Monitoring System, with different components like fall the Pi’s Wi-Fi module. The MQTT protocol is used for
detection and sleep pattern analysis. The sensors utilized the transmission.
as a part of this project are Accelerometer and Gyroscope
(MPU6050), Heart beat sensor, Body temperature sensor,
and blood oxygen level (MAX30100), and Proximity
sensor (KY032). These sensors work autonomously
of each other. The measured reading from the sensor
is broke down for the patient and is made accessible
to the specialist or to any concerned individual in the
type of the web or smart phones. This web interface
and additionally versatile application serves as the user
interface for this model. The other element added to this Fig. 2: PMS with MQTT and AWS communication
application is examination of the information in past to
caution visualizing the latest and the current reading MQTT: It is a Publish/subscribe, amazingly basic
of the exposure of the patients monitored, along with and lightweight messaging convention, intended for
the display of graph. Another element added to this constrained gadgets and low-transfer speed, high-
application is investigation of the information in past idleness or unstable and deceptive systems. The design
to caution the specialist and patient about huge changes principles are to limit network bandwidth and gadget
event, or make an alarm to specialist or any concerned resource necessities while additionally endeavoring to
individual related with the patient when it sees any guarantee reliability and some level of confirmation of
probability of therapeutic crisis. The interfacing between delivery. These principles end up making the convention
the equipment and the product part is done on the stage perfect of the developing “machine-to-machine” (M2M)
of AWS IoT. The readings are sent to AWS IoT through. or “Web of Things” world of associated devices, and for
The Modular diagram of the patient monitoring system portable applications where transmission capacity and
is described in Fig 1. battery power are at a premium.

Data analysis

Fig. 1. Module design of Patient Monitoring System Fig. 3: State diagram of sleep pattern of a patient.

Design approach: The sensor modules fused in the In this model, there are three stages of sleep awake
implanted device yields the computerized value which state, Light Sleep state and Deep Sleep state. The
 1332 Indian Journal of Public Health Research & Development, October-December 2017, Vol.8, No. 4

transition diagram for the sleep pattern analysis is shown User Interface
in the Fig 3, where the state of sleep is defined using the
Android App Interface: The Smartphone application
previous state of sleep, data from accelerometer sensor
provides the user with an interface to interact with the
and the orientation of the patient. Also, there can be
device. The application provides the user real time
change in states if the patient continues in a state for
reading of the sensors thereby getting the patient’s status
a longer duration for e.g. if the patient is in light sleep to the subscribed clients. The application queries data
and stays for a duration of more than 5 minutes then the from Dynamo DB and displays it. Heartbeat, SPO2,
probability of patient to go to deep state is high. Thus, last fall detected and current sleep state are shown. The
sleep state can be predicted using the accelerometer application also has a separate screen for showing the
data and orientation posture of the patient. From our sleep history chart. The android application shows the
observations, the accuracy of the state prediction is measured parameters heart rate, SPO2, last fall detected,
higher than threshold based predictions. and current detected sleep state. The android application
shows the measured parameters heart rate, SPO2, last fall
The figure 4 above shows the flow of data from detected, and current detected sleep state as in Fig 5.
sensors to the end device (mobile phone) through various
AWS: AWS IoT is a managed cloud platform that lets
stages. The data from various sensors are filtered in the connected devices easily and securely interact with cloud
raspberry pi and is given for the data analysis stage. Thus, applications. AWS IoT provides authentication and end-
the data from accelerometer [4] and gyroscope is used for to-end encryption throughout all points of connection,
the fall detection [5] [6] and for the sleep pattern analysis. so that data is never exchanged between devices and
The result is then passed to the AWS IoT platform as AWS IoT without proven identity. Thus, data is securely
being transmitted to the AWS IoT platform through
different topics. These different topics are used to know
MQTT protocol. AWS helps the data to be stored in the
the current state of the patient and heartbeat rate. Also, DynamoDB database and the data can then be used for
different rules are being set to notify the relatives and sleep pattern analysis. In case of emergency it helps in
doctors about the state of the patient. sending mobile notifications to the relatives and doctors
of the patient (fig 7).

IOT Healthcare Platforms: Since the engineering of

IOT-based health care equipment is more refined than
that of common IOT gadgets and requires an ongoing
working system with more stringent necessities, there
is a requirement for a modified processing platform
with run-time libraries. To figure a relevant platform, a
service-oriented architecture (SOA) [7] can be taken to
such an extent that administrations can be overburdened
and used by utilizing distinctive application program
interfaces (APIs).

Fig. 4: End to End communication for PMS Fig. 5: User Interface to End User
 Indian Journal of Public Health Research & Development, October-December 2017, Vol.8, No. 4 1333

of a change in the state, the new state is sent to AWS

IoT using MQTT. The accelerometer and gyroscope
readings and sends a notification to AWS IoT in case a
fall is detected.

The figure (Fig 8 & 9) below shows the input

readings of IR and red led for the SPO2 and heartbeat
sensor. Figure 8 represents the data for SPO2 from both
IR led and Red led. Figure 9 represents the input data for
heartbeat from IR led. This data is then filtered to get the
SPO2 and heartbeat value.
Fig. 6: Sleep pattern graph

Results and Discussions

The ambition of the project was to plan a system

which could gather reading of various important
indications of the patient and after that evaluate at cloud
then caution the doctor or concerned individuals about
the health condition. This was accomplished by building
implanted system which depends on sensors to transmit
the reading of important signs to cloud administrations
given by AWS IOT stage. These reading are chronicled
and can be obtained by either the web interface to
give a pictorial representation of information or by the Fig. 8: SPO2 graph
information analysis module to decide the seriousness
of the patient.

The figure above shows the output of the data

analysis done on the accelerometer and gyroscope data
for sleep state detection. The sensor readings are used for
monitoring any change in state and in case of a change in
the state, the new state is sent to AWS IOT using MQTT.

Fig. 9: Heart Beat graph

Conclusion

Thus, the proposed system could gather, reading of

various important indications of the patient and after that
evaluate at cloud then caution the doctor or concerned
individuals about the health condition. It monitors the
vital signs and sense abnormalities. These abnormalities
Fig. 7: SMS notification when fall notification alert the medical staff, it reduces the manual monitoring.
The system uses MQTT communication to send the data
The figure (Fig 6 & 7) above shows the output of the to cloud platform. This message protocol transmits the
data analysis done on the accelerometer and gyroscope readings of important patient’s vital sense and helps
data for sleep state detection. The sensor readings are a web interface to give a pictorial representation of
used for monitoring any change in state and in case information.
 1334 Indian Journal of Public Health Research & Development, October-December 2017, Vol.8, No. 4

Ethical clearance: Not applicable 4. A.K. Bourke, J.V. O’Brien, G.M. Lyons A.K.
Bourke et al.Gait & Posture ,”Evaluation of
Source of funding: Nil
a threshold-based tri-axial accelerometer fall
Conflict of Interest: Nil detection algorithm”, 26 (2007) 194–199.

5. Qiang Li, John A. Stankovic, Mark Hanson,

References
Adam Barth, John Lach,” Accurate, Fast Fall
1. S. M. Riazul Islam, Daehan Kwak, MD. Humaun Detection Using Gyroscopes and Accelerometer-
Kabir , “The Internet of Things for Health Care: A Derived Posture Information”, DOI: 10.1109/
Comprehensive Survey”,Date of publication June BSN.2009.46, Sixth International Workshop on
1, 2015, DOI 10.1109/ACCESS.2015.2437951.
Wearable and Implantable Body Sensor Networks,
2. Vandana Milind Rohokale, Neeli Rashmi Prasad, BSN 2009, Berkeley, CA, USA, 3-5 June 2009.
Ramjee Prasad, “A Cooperative Internet of
Things (IoT) for Rural Healthcare Monitoring and 6. J. Chen, K. Kwong, D. Chang ,”Wearable Sensors
Control”,978-1-4577-0787-2/11/ ©2011 IEEE. for Reliable Fall Detection”, Publisher: IEEE,
DOI: 10.1109/IEMBS.2005.1617246.
3. Alexandros Pantelopoulos, Nikolaos G. Bourbakis
,”A Survey on Wearable Sensor-Based Systems 7. Li Da Xu,” A Survey Internet of Things in
for Health Monitoring and Prognosis”,Publisher: Industries”, IEEE Transactions on Industrial
IEEE DOI: 10.1109/TSMCC.2009.2032660. Informatics, Vol. 10, No.4, November 2014.

View publication stats