FOG COMPUTING IN

HEALTHCARE

Presentation for CSQ 413 Seminar

VAMIDHA VINOD
TKR21CS090
Guide: Name, Designation
Department of Computer Science and Engineering
College of Engineering Trikaripur, Cheemeni

1
 ACKNOWLEDGEMENT

I am greatly indebted to all those who have rendered help for successful
completion of this work.
Firstly, I would like to express my sincere gratitude to our Principal,
Dr.Mahesh V V for the help extended by him in providing all the college facilities
required for this work.
I extend my genuine and heartfelt thanks to Dr.Naveena A K, Associate
Professor and Head, Department of Computer Science and Engineering for their
valuable suggestions, help and support during the work.
I am thankful and profoundly indebted to Mr.Rafeekh A P, Seminar
Coordinator and Assistant Professor, Department of Computer Science and
Engineering, for the moral support and guidance during the course of the work.
I also extend my sincere thankfulness to Mr.Anoop P V, Assistant Professor,
Department of Computer Science and Engineering and Mrs.Priyamvada R,
Assistant Professor, Department of Computer Science and Engineering for
evaluating the seminar and guiding me through the course of work.
I also like to thank Mrs.Shabna Salam, my Class Tutor for the
exemplary guidance, criticism, monitoring and constant encouragement
throughout the course and I am happy to thank other faculty members, technical
and administrative staff of the Department of Computer Science and Engineering
for their valuable support and heartfelt cooperation. I thank my family and
friends for giving me mental support and enabling me to work efficiently. I also
remember with regard to my parents, for all the help and moral support that
they provided me for my seminar. I express my voice of gratitude to all other staff
members and classmates for their help and motivation.

Vamidha Vinod

2
 VISION
(Institution)

To be a premier institution in education and research for moulding technically

competent and socially committed professionals.

MISSION
(Institution)

● Promote interdisciplinary research and innovation so as to meet the current

needs of industry and society.
● Attract, nurture and retain the best faculty and technical manpower.
● Provide state of art facility for quality technical education.
● Develop personality and professional skills of the students through interaction
with alumni academia and industry

3
 VISION
(Department)
To mould technically competent and socially committed professionals in the
field of computer science.

MISSION
(Department)
● To provide a strong foundation in theoretical and practical aspects of
computer science.
● To impart technical skills necessary to generate quality professional according
to industry needs.
● To develop human resource with the ability to apply the knowledge for the
benefit of the society.

4
 ABSTRACT

Fog computing is an architectural style in which network components between

devices and the cloud execute application specific logic, holds significant promise
for healthcare informatics. This review explores classify, and discusses various
healthcare application use cases leveraging fog computing. For that, we
categorize these applications into use case classes and list an inventory of
application-specific tasks that can be handled by fog computing. The Internet of
things (IoT) connects multiple devices worldwide. It is a growing field in the
healthcare system such as health monitoring and tracking, fitness program, and
remote medical assistance. With the advent of IoT based technologies in
healthcare, it can alleviate the pressure on healthcare systems and can reduce
the healthcare cost, and increase the computing and processing speed. Cloud
computing was introduced to manage larger and complex healthcare data in the
IoT environment. Cloud computing uses centralized cloud data centers. The
central server manages the data for all the IoT devices. The integration of IoT
with the cloud has some major issues such as latency, bandwidth overuse, real-
time response delays, protection, and privacy. So the concept of edge computing
and fog computing came into existence to overcome these issues. This paper
review the IoT-Fog-based system model architectures, similar paradigm, issues,
and difficulties in the area of cloud computing.

5
 CHAPTER 1
INTRODUCTION

Internet of Things (IoT) is the vision of a future connected world. IoT lets
devices become intelligent, powerful, and more efficient. IoT is making
our life easier by connecting people and objects to one another. IoT is a
technology where devices are interconnected. Devices on the Internet can
communicate and exchange data with other objects in the world of the IoT.
In an IoT environment, various forms of equipment, such as
medical sensors, vehicles, tracking cameras and household appliances can
interact, collaborate, and learn from each other. In IoT, things or devices
having sensors such as wearable, implanted, and environmental are
interconnected through a network. Connected sensors produce a large
amount of data. To execute the desired functionality, the devices or things
in the IoT are controlled remotely. IoT is improving our quality of life by
making a large number of systems and applications available.
IoT is used in many fields such as smart homes, healthcare, smart
cities, automation, smart grid, traffic management, agriculture, and so on.
Medical services and health care facilities are the most significant areas for
IoT growth. IoT development in healthcare reduces cost and increase the
quality of user’s life as they can monitor their everyday activities such as
dietary habits, sleep cycles, and exercise routines to produce specific tips
that help maintain a healthier lifestyle. Moreover, the use of IoT has
benefited many medical fields in the healthcare environment, such as
continuous real-time tracking, management of patient information, health
emergency management, management of blood information, and health
management. IoT devices and sensors produce a large amount of health
information that gathered, processed, and analyzed. Devices containing
sensors have low power, limited memory, and network and battery
limitations, so IoT data needs to be computed, stored, accessed, and
analyzed. There is a major issue in the storage and security of the enormous
volume of data that is produced by Healthcare IoT devices. A platform that
handles all this is Cloud. Cloud has unlimited capabilities of storage and
processing power.
Cloud computing shares and maximizes resources, allowing users to access
services from any location via the internet. IoT generates immense volumes
6
 of diverse data accessible through cloud computing. Integrating cloud and
IoT reduces costs and facilitates big data gathering. In healthcare, cloud
computing tracks patients, maintains records, and manages illnesses by
analyzing collected data. This minimizes hospital resources, moves
healthcare services to the home, reduces expenses, and ensures quick
access to health facilities in emergencies. It also aids in the early
identification and diagnosis of illnesses.
However, integrating IoT and cloud presents challenges. Cloud
computing isn't suitable for latency-sensitive applications. The vast amount
of IoT data increases the cloud's burden, consuming network bandwidth
and requiring time for data processing. However, the cloud is unsuitable
for critical applications. Cloud-based applications face problems such as
high bandwidth requirements, intermittent delays, and safety and security
issues. Healthcare applications need real-time monitoring, which the
cloud cannot provide. Data transfer to and from the cloud causes delays,
which are critical in healthcare where timely responses can save
lives.Cloud-based systems collect data from various sites and devices, but
sending the output back causes response delays and requires high
bandwidth for large data. Data security and user privacy are also major
concerns, making individuals hesitant to use the cloud. Fog computing has
been proposed to address these problems.
This is an architectural style for distributed systems in which
application-specific logic resides not only in data centers (the cloud) or the
devices closest to the users, but also in the infrastructure components
between them. Examples of such infrastructure components are gateways,
routers and access points. This added flexibility of computation opens new
possibilities for solving healthcare challenges. Better patient mobility and
increased integration will enable uninterrupted monitoring and also enable
entirely new applications.
The following contributions have been made through this research work
1. Reduced Latency: Fog computing minimizes delays by processing data
closer to the source, which is crucial for real-time healthcare applications.
2. Improved Bandwidth Efficiency: By handling data locally, fog computing
reduces the need for high bandwidth, easing network congestion and
improving efficiency.

7
3. Enhanced Security and Privacy: Fog computing offers better security and
privacy controls by keeping sensitive health data closer to the patient,
reducing the risk of data breaches.
4. Real-Time Data Processing: It enables immediate analysis and response,
which is vital for critical healthcare situations, ensuring timely medical
interventions.
5. Reliable and Continuous Monitoring: Fog computing supports continuous
patient monitoring with reliable data transmission and processing,
improving overall patient care.
6. Cost-Effective Solutions: By reducing the dependency on centralized
cloud resources, fog computing lowers operational costs, making
healthcare services more affordable.
7. Scalability: It allows for scalable healthcare solutions by distributing
computing resources across multiple nodes, handling large volumes of
data efficiently.

8
1.1Overview of Fog Computing
Sensor-to-cloud implementation is not achievable for many applications in the
healthcare sector. Some applications do not allow the storing of medical data
outside the hospital, and patient safety is a major concern in the case of network
and data center failure [11]. Fog computing is an approach to the mentioned
problems. The extension of cloud computing is fog computing. It is introduced
between the cloud and IoT devices. Rather than transmitting data for storing and
processing to the cloud, fog provides services near to end devices [12]. Healthcare
applications are sensitive to latency and require immediate response, so that fog
can analyze data formed by IoT devices with minimal overhead in real-time [9].
The fog architecture reduces the cloud burden. Fog and cloud computing offer
end-user capacity, processing, and network facilities. However, fog is different
from the cloud as fog nodes are closer to the network edge, so time delay can be
minimized. Fog nodes perform minimal computing so only routers, set-top boxes,
access points, mobile phones, etc. need to be used. Fog provides real-time
analysis as storage and calculation are performed near the end devices [8]. Within
the network, fog preserves sensitive data that improves user protection. These
features are mandatory in healthcare.

1.2An Architecture for a Healthcare System

The fog computing architecture offers the ideal distribution of

networking, processing, and storage device capabilities.

9
a) Device layer: Patients have sensors and monitors attached to take care of their
health. In real-time, these devices can sense and transmit data. These devices are
located on the device layer, have responsibility for healthcare selection and
transmission of healthcare data to the fog layer for accessing via Wi-Fi or mobile
network.
b) Fog computing layer: The layer of fog computing extracts medical details from
different IoT health tracking devices. This layer is used for IoT health information
collection and analysis in real-time.
c) Cloud computing layer: This layer is liable for the storage, preparing, and
executing activities that the fog layer is unable to handle and execute. For future
actions, patient condition and reports are moved to the cloud layer from the fog
layer.

10
 CHAPTER 2
LITERATURE SURVEY

The idea of fog computing and its impact on IoT was published in 2012. Fog
computing is the fastest growing interest area over time. Starting with four papers
in 2013 and three papers in 2014, the number of publications rose to thirty-three
in 2015 and twenty-two in 2016. Nearly half of these papers are directed to fog-
enabled infrastructure. These architectures have mostly been designed in the
context of healthcare and smart environments. The healthcare system strives to
make it possible for sufferers to regularly interact with their consultants and their
consultants to provide responsible patients with effective and value-added care.
The key method for translating IoT healthcare from ideas to implementations and
services is Fog computing. To help IoT healthcare systems, fog computing can be
used. Propose the work done so on fog systems and issues such as health
monitoring, data security, confidentiality, and data protection are best discussed
in the study. Fan et al. [19], propose a rehabilitation system using an ontology-
based automating design methodology (ADM). This rehabilitation system helps
to minimize problems related to elderly people. In fog-enabled facilities that
handle sensitive data, such as health tracking, security is a key issue. Ahmad et
al. [14], introduce the cloud access security broker (CASB) which strengthens
privacy and confidentiality for data in the health fog framework. CASB provides
the solution for improving health fog security issues, which are placed between
various cloud service providers and end-users. It combines various kinds of
security policy enforcement. Quwaider et al. [20], presented a paper on the global
health awareness system, which is using a multitier cloud system architecture.
This system of multi-tier health awareness would minimize delay and allow for
better exchange of health data. This system can be used for early detection of
epidemic disease before it spread over a wide region. The health awareness
system provides important health data on time and is used for the identification
and avoidance of diseases. Wireless Body Area Network (WBAN) is the basic
block of the global health awareness system. WBAN is useful in data collection
and communication. The proposed system is integrated with the MapReduce
technique to handle data processing and sharing issues. Regular exercise
decreased mental stress and anxiety level is beneficial to one's health. Physical
health however can be seriously impaired during workout sessions. Thus, Bhatia
et al. [21] focused on evaluating a person's real-time health monitoring during his
warm up session by using IoT technologies and testing them for health severity

11
evaluation. The Bayesian Belief Networks (BBN) approach is used in this paper
to assess the overall intensity of health in the form of a probabilistic metric. Gia
et al. [22], proposed a cost-saving health monitoring system that consists of a fog
layer and energy-efficient sensor nodes. This method lowers the expense of health
care and increases healthcare service quality. The nRF protocol is the basis of
these sensor nodes which are energy-efficient. The system conducts efficient
decisions and offers services for immediate attention. Today’s IoT devices are
vulnerable and unable to protect themselves. Winnie et al. [23], represent that fog
nodes function on the edge side and improve data protection, consistency,
specificity, and lowers delays. For an application such as medical data, these are
the key factors. The security of healthcare data is also improved by the use of fog
computing. To discover and regulate hypertension attacks at an initial point, Sood
et al. [24], suggested an IoTFog-based health tracking method. The artificial
neural network is used to predict the overall threat of hypertension attack in
elderly communities or individuals that are functionally impaired and live
isolated. Rahmani et al. [25], proposed a fog computing approach used by
healthcare applications to boost various IoT architecture features such as energy
consumption, interoperability, performance, trustworthiness, etc. To explain the
effectiveness and usefulness of the system in resolving a medical case study an
Early Warning Score (EWS) health monitoring has been introduced. Depending
on the high, intermediate, and low-level IoT layers Khan et al. [26], survey and
review primary IoT security challenges. This paper outlined security
requirements, existing attacks, threats, and these solutions. Khan et al. discuss
future transparent open research problems and challenges and characteristics of
blockchain-based security solutions. A massive volume of data is created by IoT
and the transmission and processing of the whole data will enhance the cloud
computing responding period. For the end-user, the increased system response
will result in high service latency. A hybrid approach is introduced by Shukla et
al. [15], which incorporates fuzzy and reinforcement learning in healthcare IoT
and cloud to improve network latency and services. This hybrid strategy uses an
algorithm to reduce latency by performing batch workloads on IoT data. By
minimizing service capacity and high network, the reinforcement learning and
fuzzy inference framework can help to get input from past patient health records
and to make real-time decisions. Mobility support is a crucial necessity for several
real-time IoT systems, as lost or postponed data throughout mobility can have
significant implications. Several approaches have recently been suggested for
awareness of mobility in the remote health tracking system. Gia et al. [27],
propose a mobility support strategy through an effective handover process for a
WiFi-based real-time IoT health monitoring system. This approach focuses on
IoT systems that collect a lot of data in real-time remote tracking of health. The
12
handover mechanism enables the link to remain between the sensor nodes and the
low bandwidth device. Abdel et al. [28], present a fog-based IoT framework for
real-time tracking of type-2 diabetes patients. To diagnose the infected
individuals a hybrid technique based on type-2 neutrosophic with VIKOR method
and evolving alerts were used. For monitoring elderly health, Hasse et al. [29]
proposed an e-health system. Physiological and general health parameters are
regularly obtained from the elderly using Mysignals HW V2 technology and a
mobile app that proposed the idea of the fog server. The problem of arthritis is
facing by many people. Regular joint health monitoring and consultation by a
physician will assist patients with this chronic disease. A WBAN-based
framework is proposed by Tanwar et al. [30], to evaluate real-time health care for
patients problem related to arthritis. To minimize false detections in the proposed
architecture, the Bayesian network classifier is used. Using deep learning and
machine learning techniques Sarabia et al. [31], present a highly efficient
intelligent fog-cloud computing framework for timely detection of falls. By the
DL inference method, the system detects falls on the ground. The use of the DL
model to detect falls enhances and reliability as it achieves a high precision with
fewer parameters than the ML model. The fog approach is responsible for
achieving a timely response as the observational data does not require to be
transferred to the cloud. For real-time surveillance of an affected person, a fog-
assisted remote monitoring framework by Kumar et al. [32] is used. To decrease
the delay between the IoT and a cloud server, an algorithm and computational
fog-assisted wearable sensor framework is used. An emerging field of study is
delay-sensitive and energy-efficient mobile health monitoring. Mukherjee et al.
[33], propose a system for indoor/outdoor users that is delay-conscious and
energy efficient fog-based Internet of Health Thing (FogIoHT). The average
delay, average signal attenuation, and consumption of energy are minimized by
this method. In this system, the Small Cell cloud enhanced e-node B (SCceNB),
cloudlet, roadway modules, gateways, and routers are used as fog equipment. The
weighted majority game theory is used when many fog devices are available to
choose a fog system for data processing. Qualnet network simulator is used to
model a fog-based health care system network scenario. MATLAB conducts a
theoretical analysis that shows that the proposed architecture is a time-sensitive
and energy-efficient system.

13
 CHAPTER 3
PROPOSED METHODOLOGY

Fog computing in healthcare is designed to enhance the efficiency and

responsiveness of healthcare services by decentralizing data processing and
storage to the network's edge. The proposed methodology involves deploying fog
nodes, such as smart gateways or edge servers, close to medical devices and
sensors. These nodes can perform real-time data analytics, filtering, and
preliminary processing, reducing latency and bandwidth usage by minimizing the
amount of data sent to the cloud. The architecture typically includes layers of
computing resources, from edge devices to central cloud servers, enabling
seamless integration and interoperability. This approach improves patient
monitoring, supports timely medical interventions, and enhances the reliability
and security of sensitive health data by leveraging localized processing and
contextual intelligence.
3.1 Architecture Design:
• Fog Nodes: These are intermediate devices that process and store
data locally before sending it to the cloud. They are strategically
placed near the edge of the network, such as in hospital premises,
to handle data from medical devices and sensors.
• Hierarchical Structure: The architecture often follows a
hierarchical model where data flows from edge devices to fog
nodes and then to the cloud. This hierarchy ensures efficient data
processing and transmission.
• Communication Protocols: Establishing robust communication
protocols between devices, fog nodes, and the cloud is crucial.
Protocols like MQTT, CoAP, and HTTP are commonly used.
3.2 Data Management:
• Real-time Processing: Fog nodes perform real-time data analysis
and processing, enabling quick decision-making for critical
healthcare applications.

14
 • Data Aggregation: Data from multiple sources are aggregated at
fog nodes to reduce redundancy and enhance data quality before
being sent to the cloud for further analysis.
• Data Storage: Temporary storage at fog nodes ensures that
essential data is available for immediate use, while long-term
storage is managed in the cloud.
3.3 Security and Privacy:
• Data Encryption: Ensuring data is encrypted during
transmission and storage to protect sensitive healthcare
information.
• Access Control: Implementing strict access control mechanisms
to ensure that only authorized personnel can access the data.
• Anomaly Detection: Utilizing machine learning algorithms at
fog nodes to detect and mitigate security threats in real-time.
3.4 Resource Management:
• Load Balancing: Distributing computational tasks across
multiple fog nodes to optimize resource utilization and avoid
bottlenecks.
• Energy Efficiency: Designing energy-efficient fog nodes to
reduce power consumption, which is critical for battery-operated
medical devices.
• Scalability: Ensuring the system can scale to accommodate an
increasing number of devices and data volume.
3.5 Application Development:
• Healthcare Applications: Developing applications for various
healthcare scenarios, such as remote patient monitoring,
telemedicine, and emergency response systems, leveraging the
fog computing infrastructure.
• Interoperability: Ensuring that applications can seamlessly
interact with existing healthcare systems and protocols.
15
3.6 Implementation and Deployment:
• Pilot Projects: Conducting pilot projects in controlled
environments to test the fog computing infrastructure and
applications before large-scale deployment.
• Training and Support: Providing training for healthcare
professionals and IT staff on how to use and maintain the fog
computing system.
• Continuous Monitoring: Implementing monitoring systems to
continuously assess the performance, security, and reliability of
the fog computing infrastructure.

16
 CHAPTER 4
RESULTS AND DISCUSSIONS

This paper discusses the effect of the internet of things (IoT) and fog computing
in healthcare, using a systematic literature review. This review reveals that with
the expansion of the IoT and smart devices, data nodes are growing at an
extraordinary speed allowing massive data that is to be kept and analyze in the
cloud for a variety of reasons. These reasons specially discuss the problems of
central storage, challenging computation, and sharing of data. Fog computing acts
to refine the whole process as a bridge between both the end-user and the cloud.
Fog computing successfully minimizes excessive interaction from node-to-cloud
data generation. Moreover, to ensure the protection and security of data, certain
rules and regulations can be integrated into the Fog. In traditional applications
that are strictly cloud-dependent, the increases in delay and the standard of
services required deterioration. For this reason, it improves data security,
precision, and reliability, decreases the level of delay, and improves the service
quality by adding fog as a middle layer and operating on the edge side. In ehealth
related systems and applications, cloud and fog computing has attracted profound
interest. The IoT-Fog-cloud architecture commonly uses soon as more IoT
devices are being built and fast computing demand is on the rise

17
 CHAPTER 5
ADVANTAGES AND DISADVANTAGES

5.1 ADVANTAGES
1. Reduced Latency: Fog computing processes data closer to the source,
ensuring real-time analysis and quicker response times, which is critical
in healthcare for timely medical interventions.
2. Bandwidth Efficiency: By handling data locally, fog computing reduces
the need for constant data transfer to the cloud, saving bandwidth and
reducing network congestion.
3. Enhanced Security and Privacy: Keeping sensitive health data at the edge
rather than transferring it to the cloud minimizes the risk of data breaches
and enhances patient privacy.
4. Continuous and Reliable Monitoring: Fog computing supports continuous
patient monitoring, ensuring reliable data transmission and improving
overall patient care.
5. Cost Reduction: Reducing dependency on centralized cloud resources
lowers operational costs, making healthcare services more affordable.
6. Scalability: Fog computing allows for scalable healthcare solutions by
distributing computing resources across multiple nodes, efficiently
handling large volumes of data.
7. Interoperability: Facilitates seamless data integration and communication
between various healthcare devices and systems, improving the efficiency
of healthcare delivery.
8. Local Decision Making: Enables immediate analysis and local decision-
making, crucial for critical healthcare situations requiring fast responses.
9. Resource Optimization: Dynamic load balancing and resource
management ensure optimal utilization of computing resources,
enhancing system performance.

18
5.2DISADVANTAGES
1. Complexity: Implementing fog computing requires sophisticated
infrastructure and expertise, which can be complex and costly to set up
and maintain.
2. Security Risks: While fog computing enhances security by keeping data
local, edge devices may still be vulnerable to cyber-attacks, requiring
robust security measures.
3. Resource Constraints: Edge devices may have limited processing power
and storage capacity compared to centralized cloud servers, potentially
limiting their capabilities.
4. Maintenance and Management: Managing a distributed network of fog
nodes can be challenging, requiring regular maintenance and monitoring
to ensure reliability and performance.
5. Standardization Issues: Lack of standardized protocols and frameworks
can hinder interoperability and integration between different fog
computing systems and healthcare devices.
6. Scalability Challenges: While fog computing is scalable, managing and
scaling up a large number of edge devices can be difficult and resource-
intensive.
7. Data Consistency: Ensuring data consistency and synchronization across
multiple edge nodes and the cloud can be challenging, particularly in real-
time applications.
8. Initial Costs: The initial investment in fog computing infrastructure can
be high, potentially posing a barrier for smaller healthcare providers.

9. Regulatory Compliance: Ensuring compliance with healthcare regulations

and standards (e.g., HIPAA) can be complex when dealing with
distributed data processing and storage.

19
 CHAPTER 6
FUTURE SCOPE

The future scope of fog computing in healthcare is vast and promising, poised to
revolutionize the way healthcare services are delivered. As medical technologies
advance and the Internet of Things (IoT) continues to grow, fog computing will
play a crucial role in enabling real-time, data-driven decision-making at the edge
of the network. This will facilitate immediate responses to critical health
conditions, improve patient monitoring, and enhance the management of chronic
diseases through continuous and localized data processing. Additionally, fog
computing will support the integration of artificial intelligence and machine
learning algorithms at the edge, offering personalized and predictive healthcare
solutions. With ongoing advancements in security protocols and interoperability
standards, fog computing will ensure secure and seamless data exchange between
various healthcare devices and systems. This will lead to more efficient healthcare
delivery, reduced costs, and improved patient outcomes. As healthcare providers
increasingly adopt digital health solutions, fog computing is set to become a
cornerstone of next-generation healthcare infrastructures, driving innovation and
improving the quality of care across the globe.

20
 CHAPTER 7
CONCLUSION

The Internet of Things (IoT) connects devices to the internet, making them
smarter and more powerful. It is used in various industries, including healthcare,
smart homes, smart cities, and agriculture. In hospitals and homes, IoT can alert
health professionals to emergencies in real-time. Cloud computing helps track
patients, store records, and manage diseases, but it faces issues like connectivity
demands, delays, and security challenges. Fog computing addresses these
problems by processing data closer to the source, reducing latency and network
load. This paper surveys IoT-Fog-cloud architectures, reviewing methods and
standards. Future work may involve applying iFogSim to smart city scenarios,
smart parking, waste management, and coal mining, and improving data
placement management to reduce service latency, network traffic, and energy
consumption. Fog computing is ideal for healthcare as it filters data to preserve
privacy, adapts to regulations, and improves system reliability. For maximum
benefit, comprehensive architectures are needed beyond individual use cases.

21