Virtual Telemedicine System for Remote Health Monitoring using

Machine Learning
 ABSTRACT

 Healthcare technology is one of the most popular studies nowadays. With the
development of healthcare technology, the lifespan of people has successfully
extended.

 However, people in the rural area are still having a hard time to obtain professional
healthcare services due to the barrier of distance and lack of doctors.

 A remote patient monitoring system is one of the best solutions to overcome this issue.

 This paper proposes an Internet of Things (IoT) based real-time remote patient
monitoring system that is able to guarantee the integrity of the real-time.

 The sensors used are Heart Beat, Temperature Sensor, Sweat Sensor, Spo2 Sensor. The
data which is been received is send to the cloud and Anaconda software has been used.
 1. INTRODUCTION
 In today’s scenario, health care problems are increasing at a very high pace like
coronary heart diseases, obesity, lung failures causing death rate of 7.2 million people
per year hence it is the need of the hour to overcome from all such problems.

 In the present busy days constant monitoring of the patient’s body parameters such as
temperature and heart rate etc. becomes difficult.

 Hence to remove the burden of monitoring patients health from doctor’s head. This
project presents the methodology for monitoring patients remotely using IOT and
embedded technology.

 A remote patient monitoring system is one of the solutions to overcome the barriers of
health service. It provides professional health care services to the rural population
through information and communication technologies (ICT).

 Remote patient monitoring system allows the doctor to observe the patient’s health
condition remotely at any time and any place. The prevention can be done during the
early detection of the disorder.
 1.1 BACKGROUND TO THE PROJECT
TITLE
 Now-a-days, the medical electronics-sensors (E-sensors) are playing an important
role in health care centers. The patient monitoring is one of the major advancements
in research field.

 Here we use the temperature sensor, heartbeat sensor to monitor the patient’s body
temperature and heart rate respectively.

 Hence like the use of thermometer in home to check body temperature before
doctor’s consultation, this proposed model (devices) can be used to check the
patient’s health condition in home as first aid information to the concerned patient
otherwise now-a- days consulting doctors or going to diagnosis centers become
difficult in terms of pandemic and also financial aspect.

 To minimize this situation, we describe the design of a Arduino microcontroller

based advanced/high performance integrated health portable monitoring system.
Hence the name Virtual Telemedicine System for Remote Health Monitoring
using Machine Learning.
 1.2 RATIONALITY OR PURPOSE OF
CHOOSING THE TOPIC
 The worldwide virtual healthcare delivery markets are expected to achieve
significant growth as the pandemic emergency emerges as a worldwide issue. .

 Minimizing the need for individuals to go to health care facilities for evaluation
decreases the risk of contracting COVID-19 disease.

 The existing system requires many separate components to be carried by an

individual to measure their own pulse rate ,blood pressure, temperature and so on.
 
 It restricts doctors assigned to every individual patient to be present near patients all
the time. 

 All these stipulations can be met by implementing the Virtual Telemedicine System
for Remote Health Monitoring using Machine Learning.
 2. LITERATURE SURVEY
SL.NO PAPER AUTHOR ABSTRACT
1. Android application P. Szakacs-Simon ; S. A. The main scope is to
developed to extend Moraru ; L. Perniu monitor people at risk in
health monitoring Publications: IEEE 9th real time even if they are
device range and International Conference outside of their home and
real-time patient on Computational wireless home gateway is
tracking Cybernetics (ICCC) out of range.

2. Android based health Maradugu Anil The health care scheme is

care monitoring Kumar ; Y.Ravi Sekhar focus on the measurement
system Publications: and Monitoring various
International Conference biological parameters of
on Innovations in patient's body like heart
Information, Embedded rate, oxygen saturation
and Communication level in blood and
Systems (ICIIECS) temperature using a web
server and android
application.
SL.NO PAPER AUTHOR ABSTRACT
3. Mobile sensor data Won-Jae Yi ; Weidi In this paper, we present a
collector using Jia ; Jafar Saniie system using an Android
Android smartphone Publications: IEEE smartphone that collects,
55th International displays sensor data on the
Midwest screen and streams to the
Symposium on central server simultaneously.
Circuits and Bluetooth and wireless
Systems Internet connections are used
(MWSCAS) for data transmissions among
the devices.
 3. OBJECTIVE

 To design virtual remote health monitoring system.

 To process the data collected from the patient through various sensors like
temperature, heartbeat, sweat, oxygen levels etc.

 To store the values in cloud for further accessing.

 To make use of IOT platform and cloud handling data along with machine learning for
display and analysis.

 Faster response can be achieved by using this technique.

4. BLOCK DIAGRAM
Hardware requirements:

 Esp32
 Heartbeat sensor
 Temperature sensor
 Spo2 sensor
 Sweat sensor
 LAPTOP

Software Requirements:

 Arduino IDE
 Embedded C
 Anaconda Jupyter Notebook
Block diagram specifications:

 ESP32: ESP32 is a series of small and fast microcontroller boards like Arduino but
comes with inbuilt Wi-Fi and Bluetooth features. The ESP32 boards can be
programmed using many different programming languages. For example, we can
program the ESP32 board in C++ language(like the Arduino) or MicroPython
- Single or Dual-Core 32-bit LX6 Microprocessor with clock frequency upto 240MHz
- 520 KB of SRAM, 448 KB of ROM and 16 KB of RTC SRAM.
- Supports 802.11 b/g/n Wi-Fi connectivity with speeds up to 150 Mbps.
- Support for both Classic Bluetooth v4.
- 34 Programmable GPIOs.

 Heartbeat Sensor: Electrocardiogram sensor is used to measure the heart rate of a

person. Electrocardiography (ECG) is used to get information of cardiovascular system
of the person. The device consists of MAX86150 Integrated ECG and PPG Bio-Sensor.
It operates at 1.8 volts. It monitors heart rate and also acts as pulse oximeter. Generally,
if a person is in trouble, his/her hands shake. The device consists of SW-420 Module of
vibration sensor. It operates at 3.3 volts to 5 volts DC voltage.
 Spo2 sensor: The MAX30100 is an integrated pulse oximetry and heart rate monitor
sensor solution. It combines two LEDs, a photo detector, optimized optics, and low-
noise analog signal processing to detect pulse oximetry and heart-rate signals. The
MAX30100 operates from 1.8V and 3.3V power supplies and can be powered down
through software with negligible standby current, permitting the power supply to
remain connected at all times. Some of its features are:
-Complete Pulse Oximeter and Heart-Rate Sensor Solution Simplifies Design
-Integrated LEDs, Photo Sensor, and High-Performance Analog Front -End
-Tiny 5.6mm x 2.8mm x 1.2mm 14-Pin Optically Enhanced System-in-
Package
-Ultra-Low-Power Operation Increases Battery Life for Wearable Devices

 Sweat sensor: Sweat detection can be used as an alternative to blood detection and
allows continuous monitoring. Sweat can be more easily sampled than say tears,
saliva or intestinal fluids. The sensors are embedded onto the plastic in a manner
similar to the way a printing press imprints text on a newspaper. Inside the sensors
are spiralling microfluidic tubes, that can derive information on the rate at which the
wearer is sweating.
 Temperature sensor: They are devices to measure temperature readings through
electrical signals. The LM35 series are precision integrated-circuit temperature
sensors, whose output voltage is linearly proportional to the Celsius (Centigrade)
temperature. The LM35 does not require any external calibration or trimming to
provide typical accuracy. Some of its features are:
-0.5°C accuracy guarantee able (at +25°C)
-Rated for full −55° to +150°C range
-Suitable for remote applications
-Low cost due to wafer-level trimming
-Less than 60μA current drain
-Operates from 4 to 30 volts
-Low self-heating, 0.08°C in still air
-Low impedance output, 0.1 W for 1 mA load
 Arduino IDE: The Arduino Integrated Development Environment (IDE) is a cross-
platform application that is written in functions from C and C++. It is used to write
and upload programs to Arduino compatible boards, but also, with the help of third-
party cores, other vendor development boards. The Arduino IDE supplies a software
library from the wiring project, which provides many common input and output
procedures.

 Embedded C: Embedded C is most popular programming language in software field

for developing electronic gadgets. It plays a key role in performing specific function
by the processor.

 Anaconda Jupyter Notebook: The Jupyter Notebook application allows to create

and edit documents that display the input and output of a Python or R language script.
 5. METHODOLOGY
 Figure shows the block diagram of Virtual Telemedicine System.

 Sensor devices such as temperature sensor, heartbeat sensor, spo2 sensor and sweat
sensor are used to sense the body parameters.

 The devices are implemented as a wrist band and the patient is monitored.

 ESP32 is the powerful tool for IoT applications having inbuilt Bluetooth and Wi-Fi
which is programmed using the Arduino IDE.

 Data is collected from different sensor networks and processed by the Esp32
microcontroller for analysis.

 ThingSpeak cloud is a free web service for displaying the data online and one can
access and monitor the data from ThingSpeak from anywhere.
 The analyzed data is sent over the internet and stored in the cloud for access by the
doctors.

 By using ThingSpeak server, we can monitor our data over the internet using the API
and channels provided by ThingSpeak.

 The patient's medical history is stored in the cloud. From the cloud the processed data
can be retrieved for treatment.

 Machine learning is the concept of programming the machine in such a way that it
learns from its experiences and different examples, without being programmed
explicitly.

 It is an application of AI that allows machines to learn on their own. Machine

learning algorithms are a combination of math and logic that adjust themselves to
perform more progressively once the input data varies.

 Being a general-purpose, easy to learn and understand language, Python can be used
for a large variety of development tasks. It is capable of doing a number of machine
learning tasks, which is why most algorithms are written in Python.
 Machine learning uses algorithms such as linear regression and decision tree for
training and updating the data.

 Linear Regression is one of the most popular Supervised Machine Learning

algorithms in Python that maintains an observation of continuous features and
based on it, predicts an outcome.

 A decision tree is built by repeatedly asking questions to the partition data. The aim
of the decision tree algorithm is to increase the predictiveness at each level of
partitioning so that the model is always updated with information about the dataset.

 Training data is the data used to train a machine learning algorithm or model to
accurately predict a particular outcome, or answer, that we want our model to
predict.

 The algorithm is trained with hospital dataset to determine the doctors experience,
type of diagnosis and other required information.
 
 Training datasets can include text (words and numbers), images, video, or audio.
They can be available in many formats, such as a spreadsheet, PDF or HTML.

 The machine learning algorithms along with the trained data and the sensor
parameters are used to analyze the patient’s health condition and is continuously
updated to the cloud.

 A Threshold range of the parameters is fixed to monitor the current status of patient.

 If any parameters like heartbeat, temperature and oxygen level deviates from the
standard range, then patient must be alerted and notification is sent to the family
and doctor.

 The analyzed data and the treatment given by doctors are again stored in the cloud
which can be retrieved by the doctors to know the history of the patient and for
further treatment. The values obtained and the state of the person is made available.
 6. APPLICATIONS
 Virtual Telemedicine system enables patients to receive medical attention at the
convenience of both doctor and him, and at the same time, he is safe.

 Telemedicine systems with digital health monitoring capabilities, combined with

artificial intelligence and machine learning, will provide more accurate diagnoses
and treatment recommendations.

 Telemedicine facilitates access to services for a patient having disabilities. It also

increases access for other groups, such as the elderly, the culturally isolated, and
the incarcerated.

 By continuous evaluation of patients’ the system provides access to send

notifications and alerts to family and doctor in case of emergencies.

 Doctors can now track their patients' welfare over long distances using high-tech
medical equipment. This technology enables information such as pulse rhythm,
blood pressure, temperature, and more by transmitting data from sensors to cloud.
 7. ADVANTAGES

 The virtual telemedicine system is more accurate when compared to other systems
since it is a real time monitoring.

 The system is more efficient, reliable.

 The system is useful for the patients residing in inaccessible areas or isolated regions.

 The virtual system has reduced manual operation to a major extent as the sensors are
always connected and continuous monitoring is enabled.

 When compared to conventional ways of healthcare administration, telemedicine is a

more cost-effective approach. This is valid since doctors and patients do not need to
travel each time; they need to see a doctor or patient and also saves time.
 8. DISADVANTAGES

 Virtual clinical treatment decreases human interaction among the healthcare

professionals and patients that increases the risk of error in clinical services.

 Hackers and other criminals may be able to access a patient’s medical data,
especially if the patient accesses telemedicine on a public network or via an
unencrypted channel.

 Delay in alarms or false alarms may be generated due to battery issues in sensors.
 9. CONCLUSION
 The virtual telemedicine system provides low complexity, low power
consumptions and highly portable for remote health monitoring of patient’s and it
can eliminate the need of utilization of expensive facilities.

 In this study a real-time health monitoring system of patients located in remote

areas has been proposed.

 The developed system is comprised of wearable sensors, cloud platform using

machine learning algorithm. The system is adaptable and has the ability to extract
several parameters such as heart rate, oxygen levels, and temperature of patients.

 The extracted data is being transmitted to cloud using Bluetooth for further
processing. Machine learning algorithm is used to determine the status of the
patient through continuous monitoring.

 An alarming system based on threshold values has also been designed which sends
alert message to the doctor and family in case of abnormalities.
 10. REFERENCES
1. “Android Application Developed to Extend Health Monitoring Device Range and
Real-time Patient Tracking”, ICCC 2013,IEEE 9th International Conference on
Computational Cybernetics, P. Szakacs Simon * , S. A. Moraru * and L. Perniu.
2. “Android based health care monitoring system”, International Conference on
Innovations in Information, Embedded and Communication Systems (ICIIECS),
Maradugu Anil Kumar ; Y.Ravi Sekhar
3. “Wearable Sensor Data Fusion for Remote Health Assessment and Fall Detection
Won-Jae Yi, Oishee Sarkar, Sivisa Mathavan and Jafar Saniie
4. “Android based patient monitoring system” International Journal For Technological
Research In Engineering,May-2014, Deep Modi 1 , Jignesh Vyas 2 , Priyank Shah 3
ME Scholar, 2 Associate Professor (H.O.D.), 3 Assistant Professor
5. “A Mobile Software for Health Professionals to Monitor Remote Patients” IEEE
2012 Maneesha V Ramesh, Sruthy Anand.
6. Y. Hao and J. Foster, “Wireless sensor networks for health monitoring applications,”
Physiological Meas., vol. 29, no. 11, pp. R27–R56, 2008.
7. Aart Van Halteren, Richard Bults, KatarzynaWac, Dimitri Konstantas, Ing Widya,
Nicolay Dokovsky, George Koprinkov, Val Jones, Rainer Herzog, “Mobile Patient
Monitoring: TeMobiHealth System”, Te Journal on Information Technology in
Healthcare 2004; 2(5): 365–373