healthcare
Article
Towards a Secure Technology-Driven Architecture for Smart
Health Insurance Systems: An Empirical Study
Fatima Al-Quayed 1, * , Mamoona Humayun 2, *
Citation: Al-Quayed, F.; Humayun,
M.; Tahir, S. Towards a Secure
Technology-Driven Architecture for
Smart Health Insurance Systems: An
Empirical Study. Healthcare 2023, 11,
2257. https://doi.org/10.3390/
healthcare11162257
Academic Editor: Liaquat Hossain
Received: 18 July 2023
Revised: 6 August 2023
Accepted: 8 August 2023
Published: 10 August 2023
Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Healthcare 2023, 11, 2257. https://doi.org/10.3390/healthcare11162257
and Sidra Tahir 3
1 Department of Computer Science, College of Computer and Information Sciences, Jouf University,
Sakakah 72341, Saudi Arabia
2 Department of Information Systems, College of Computer and Information Sciences, Jouf University,
Sakakah 72311, Saudi Arabia
3 University Institute of Information Technology, PMAS Arid Agriculture University, Rawalpindi 43600, Pakistan;
stahir@uaar.edu.pk
* Correspondence: ffalquayed@ju.edu.sa (F.A.-Q.); mahumayun@ju.edu.sa (M.H.)
Abstract: Health insurance has become a crucial component of people’s lives as the occurrence of
health problems rises. Unaffordable healthcare problems for individuals with little income might
be a problem. In the case of a medical emergency, health insurance assists individuals in affording
the costs of healthcare services and protects them financially against the possibility of debt. Security,
privacy, and fraud risks may impact the numerous benefits of health insurance. In recent years, health
insurance fraud has been a contentious topic due to the substantial losses it causes for individuals,
commercial enterprises, and governments. Therefore, there is a need to develop mechanisms for
identifying health insurance fraud incidents. Furthermore, a large quantity of highly sensitive
electronic health insurance data are generated on a daily basis, which attracts fraudulent users.
Motivated by these facts, we propose a smart healthcare insurance framework for fraud detection
and prevention (SHINFDP) that leverages the capabilities of cutting-edge technologies including
blockchain, 5G, cloud, and machine learning (ML) to enhance the health insurance process. The
proposed framework is evaluated using mathematical modeling and an industrial focus group. In
addition, a case study was demonstrated to illustrate the SHINFDP’s applicability in enhancing the
security and effectiveness of health insurance. The findings indicate that the SHINFDP aids in the
detection of healthcare fraud at early stages. Furthermore, the results of the focus group show that
SHINFDP is adaptable and simple to comprehend. The case study further strengthens the findings
and also describes the implications of the proposed solution in a real setting.
Keywords: smart health insurance system; technology-driven; 5G; cloud; blockchain; machine learning
1. Introduction
In recent years, the number of individuals with health insurance (HIN) has expanded
considerably around the globe. In industrialized nations such as the US, more than ninety
percent of inhabitants have health insurance. HIN promotes access to care and is connected
with decreased mortality rates, improved health outcomes, and enhanced productivity [1,2].
Those who do not use HIN endanger their physical, emotional, and financial wellness.
Meaningful healthcare coverage is essential to live a productive, secure, and healthy
life [3,4]. HIN is accessible via a range of private and public sources. It is a contract between
the health insurance provider (HIP) and the health insurance subscriber (HINS) in which
HIP bears the healthcare expenses of HINS. HIN covers ordinary healthcare concerns
and losses due to accidents, healthcare expenditures, incapacity, unintentional injury, and
damage. However, in order to obtain this benefit, HINS must constantly pay the fee for this
compensation [1,4,5]. A high uninsured rate strains the whole healthcare system. People
without HIN delay necessary treatment and depend more heavily on hospital emergency
rooms, diverting limited resources to address diseases that might often have been averted
https://www.mdpi.com/journal/healthcare
 Healthcare 2023, 11, x FOR PEER REVIEW 2 of 27
Healthcare 2023, 11, 2257 2 of 27
HINS must constantly pay the fee for this compensation [1,4,5]. A high uninsured rate
strains the whole healthcare system. People without HIN delay necessary treatment and
depend more heavily on hospital emergency rooms, diverting limited resources to
or treateddiseases
address in a lower-cost
that mightenvironment.
often have The beenabsence
avertedof or
HIN has severe
treated in a cost consequences
lower-cost
for people, communities,
environment. The absence ofand HIN the
hashealth
severecare
cost system [6–8]. for people, communities,
consequences
Previously,
and the health carethe health
system insurance claim (HIC) procedure was semi-automated, with
[6–8].
several drawbacks,
Previously, such asinsurance
the health wasted timeclaimand money
(HIC) and the
procedure was HINSs’ need to frequently
semi-automated, with visit
several
the HINdrawbacks, such asthe
office to initiate wasted time request
premium and money andand the HINSs’
to know aboutneed
HICto frequently
status. Furthermore,
visit the of
auditing HINHIC office to initiate
requires more the premium
human request
resources andand to know
is more proneabout HIC status.
to mistakes as handling
Furthermore, auditing of HIC requires more human resources
paper-based HIC data is arduous. In such a system, HIC documents are modifiable and is more prone to and
mistakes as handling paper-based HIC data is arduous. In such a system,
freely available. Consequently, the HIP, healthcare service provider (HSP), and HINS are HIC documents
are modifiable and freely available. Consequently, the HIP, healthcare service provider
vulnerable to fraud owing to a lack of privacy and transparency. Nowadays, information is
(HSP), and HINS are vulnerable to fraud owing to a lack of privacy and transparency.
collected in a digital format, revolutionizing HIC globally. The advantages of digitization
Nowadays, information is collected in a digital format, revolutionizing HIC globally. The
include convenience
advantages to the
of digitization parties
include engaged in
convenience HIC,
to the no need
parties for manual
engaged in HIC, no visits,
needfast
for commu-
nication
manual visits, fast communication between HINS and HIP, audit and checking, lack oflow cost,
between HINS and HIP, audit and checking, lack of fraudulent activities,
easy claim activities,
fraudulent verification,lowandcost,many
easy more
claim [9–11]. The digitization
verification, and many more and benefits
[9–11]. Theassociated
with HIN have
digitization motivated
and benefits individuals
associated with HIN worldwide to purchase
have motivated appropriate
individuals worldwide HINto policies.
The global
purchase HIN market
appropriate HINispolicies.
increasing rapidly
The global HINand is predicted
market to increase
is increasing rapidly further,
and is as can
predicted
be to increase
seen from further,in
the statistics as Figure
can be seen
1. The from the statistics
statistics usedininFigure
Figure 1. 1The
arestatistics
taken from the
used in Figure
“Research and1Market”
are takenreport
from the “Research and Market” report [12].
[12].

Global health insurance market in trillion USD

2.76

2022
2027
4.46

Figure 1.
Figure 1. Global
GlobalHIN
HINmarket
marketprediction in USD
prediction trillions.
in USD trillions.

Despite
Despite several
severaladvantages,
advantages,the thedigitized
digitized HINHIN market
markethashas
to to
face
facenumerous
numerous chal-
lenges, and one of the critical challenges faced globally is HIN fraud. Vastsums
challenges, and one of the critical challenges faced globally is HIN fraud. Vast sumsofof money
money are lost worldwide due to HIN fraud [13], raising HIP firms’ administrative costs
are lost worldwide due to HIN fraud [13], raising HIP firms’ administrative costs and
and the premiums HINS must pay [14]. A total of around GBP 303.8 million was lost to
the premiums HINS must pay [14]. A total of around GBP 303.8 million was lost to
HIN fraud, according to a 2015 study on the financial impact of HIN fraud [13]. This sum
HIN
may befraud, according
divided into GBPto237 a 2015 study
million on the financial
for fraudulent impact charges,
pharmaceutical of HIN GBP fraud [13]. This
43.9
sum may be divided into GBP 237 million for fraudulent pharmaceutical
million for fraudulent dental expenses, and GBP 22.9 million for fraudulent optical charges, GBP
43.9 million
charges. for fraudulent
Similarly, Thaifur et dental
al. [15] expenses,
estimate thatandhealthcare
GBP 22.9fraudmillion
costsfor fraudulent
Korea and opti-
cal charges.
Europe, Similarly,
respectively, EURThaifur et al.and
798.2 billion [15]
EURestimate that
56 billion healthcare
annually. fraud costs
The monthly Korea and
medical
Europe, respectively,
assistance payments for EUR
each798.2
memberbillion andnational
of the EUR 56HIN billion annually.
in South AfricaThe monthly
increased by medical
ZAR 192 (USD
assistance 14) to ZAR
payments for 410
each(USD 30) due
member oftothe
fraud. These HIN
national minorinhealthcare overheads
South Africa increased by
are projected
ZAR 192 (USD to be
14)about USD410
to ZAR 882(USD
million.30) due to fraud. These minor healthcare overheads
Since the to
are projected lastbedecade,
aboutcutting-edge
USD 882 million.technologies, especially AI and blockchain, have
been Since
widelythe
used
lastindecade,
the healthcare sector fortechnologies,
cutting-edge fraud detectionespecially
and prevention
AI and[3,16,17]. By
blockchain, have
analyzing previous data, AI can fuel prediction models that evaluate the likelihood of an
been widely used in the healthcare sector for fraud detection and prevention [3,16,17]. By
occurrence happening. Not only that, but they may also estimate the worth of the HIC by
analyzing previous data, AI can fuel prediction models that evaluate the likelihood of an
occurrence happening. Not only that, but they may also estimate the worth of the HIC
by analyzing past claims made by the policyholder. A claim value above the predicted
value can be forwarded for further investigations [18–21]. In the same way, end-to-end
traceability and a smart contract in the BLC help to detect and prevent HIN fraud [5,22–25].
On the other hand, 5G facilitates the fast transmission of insurance data between
different entities. At the same time, the cloud overcomes the limited capacity of the
 analyzing past claims made by the policyholder. A claim value above the predicted value
can be forwarded for further investigations [18–21]. In the same way, end-to-end
traceability and a smart contract in the BLC help to detect and prevent HIN fraud [5,22–
25].
Healthcare 2023, 11, 2257 On the other hand, 5G facilitates the fast transmission of insurance data between 3 of 27
different entities. At the same time, the cloud overcomes the limited capacity of the
blockchain digital ledger (DTL) and provides a secure storage mechanism for healthcare
data. Considering
blockchain the importance
digital ledger (DTL) andofprovides
HIN globally, this
a secure research
storage leveragesfor
mechanism thehealthcare
potential of
the above-mentioned cutting-edge technologies to propose a smart solution for
data. Considering the importance of HIN globally, this research leverages the potential HIN fraud
of
detection and prevention. The problem statement and the explicit contribution
the above-mentioned cutting-edge technologies to propose a smart solution for HIN fraud of the
study are
detection andmentioned
prevention.in the
Thefollowing subsections
problem statement and the explicit contribution of the study
are mentioned in the following subsections
1.1. Problem Statement
1.1. Problem Statementfrequently give false or deceptive information to health insurance
Policyholders
companies in an effort
Policyholders to obtain
frequently givereimbursement
false or deceptiveforinformation
unapprovedtobenefits, which results
health insurance com-in
a significant
panies losstotoobtain
in an effort the HINP. A transparent
reimbursement end-to-end benefits,
for unapproved processing method
which forinHIN
results a
significant
claims is loss to theinHINP.
required orderAtotransparent end-to-end
help identify and stopprocessing
HIN scams.method for HIN claims is
required in order to help identify and stop HIN scams.
1.2. Paper Contribution
1.2. Paper Contribution
The explicit contributions of this study are listed below:
The explicit contributions of this study are listed below:
1. We explore the existing literature to identify possible HIN fraud and the entities
1. We exploreinthe
involved it; existing literature to identify possible HIN fraud and the entities
involved in it;
2. We provide a review of the various current usages of cutting-edge technologies for
2. We provide applications,
healthcare a review of the various in
especially current usages
the HIN of cutting-edge technologies for
process;
healthcare applications, especially in the HIN process;
3. We propose a layered framework based on cutting-edge technologies to detect and
3. We propose
mitigate a layered
HIN frauds;framework based on cutting-edge technologies to detect and
mitigate HIN frauds;
4. The proposed framework is evaluated using mathematical modeling, and an
4. The proposed framework is evaluated using mathematical modeling, and an industrial
industrial focus group;
focus group;
5. The practical implications of the proposed framework are elaborated with the help
5. The practical implications of the proposed framework are elaborated with the help of
of a case study.
a case study.
Before proceeding to the background and overview of existing studies, Acronyms
Before proceeding to the background and overview of existing studies, Acronyms part
part in Abbreviations lists the study’s acronyms and their complete forms.
in Abbreviations lists the study’s acronyms and their complete forms.
Figure 2 depicts the structure of the remaining paper.
Figure 2 depicts the structure of the remaining paper.

Figure
Figure 2. 2. Study
Study roadmap.
roadmap.

2.2.Preliminaries
Preliminariesand
andRelated
RelatedWork
Work
This section
This will
section provide
will anan
provide overview ofof
overview HIN, the
HIN, challenges
the faced
challenges byby
faced HIN, and
HIN, the
and the
role of cutting-edge technologies in HIN.
role of cutting-edge technologies in HIN.
2.1. Why There Is a Need for HIN
2.1. Why There Is a Need for HIN
Having HIN improves access to care, which leads to fewer deaths and improved health
Having HIN improves access to care, which leads to fewer deaths and improved
outcomes. HIN may represent the difference between disease and health, or life and death.
health outcomes. HIN may represent the difference between disease and health, or life
One of the primary causes of mortality, particularly in emerging and underdeveloped
and death. One of the primary causes of mortality, particularly in emerging and
nations, is a lack of HIN. Healthy individuals and young adults may believe they do not
need HIN since they are seldom ill. Accidents and diseases, on the other hand, may happen
to anybody at any moment. Without HIN, we are responsible for funding our own medical
expenditures, placing us in severe financial jeopardy [26–28]. Therefore, HIN is vital to
living a healthy and peaceful life. Below are some key benefits of HIN.
Healthcare 2023, 11, 2257 4 of 27

Financial benefits
The following financial benefits are associated with HIN:
• HIN aids in the payment of medical costs and prescription medication costs.
• It assists you in avoiding significant medical debt.
• It limits how much you may spend on healthcare each year.
Health benefits
People usually do not visit a healthcare center regularly without HIN. Individuals
with HIN have the following benefits:
• HIN helps you live a longer, healthier life.
• It connects you to a consistent source of care.
• It aids in the diagnosis and treatment of ailments and disorders.
Peace of mind
An essential thing in an individual’s life is peace of mind; however, the chances of
illness are always there. Most people are not only afraid of disease but also more fearful of
the cost they have to bear. HIN provides peace of mind in the following ways:
• Reducing financial worries regarding medical expenses.
• Reducing health worries regarding a lack of access to care.
• Removing the stigma associated with being uninsured.

2.2. Healthcare Insurance Claim (HIC) Processing

The HIC processing cycle needs to be understood to better understand HIN fraud.
Various stakeholders are involved in processing HIC, such as doctors, paramedical staff,
hospital administration, HIN subscribers, and HIP [29–31]. The hospital administration
initiates the claim when a patient receives treatment from the hospital; all of the HSPs
involved in the treatment enter the treatment information into the smart healthcare system
(SHS). The HSP personnel who is responsible for preparing HIC makes a claim and has
it verified by the concerned HSP authorities. Once the HSP verifies the claim, it is sent to
the HIP for further processing. When the HIP receives a claim, it verifies it and checks the
patient’s eligibility and claim validity. Once the HIC is verified and validated, it is sent
for approval in the concerned section. Once the HIC is approved, the amount of HIC is
transferred to the HSP or HINS’s account. The complete steps of HIC process are provided
inREVIEW
Healthcare 2023, 11, x FOR PEER Figure 3. 5 of 27

Figure 3. HIC processing steps.

Figure 3. HIC processing steps.
2.3. Healthcare Insurance Fraud
HIN fraud impacts people and companies equally, costing tens of billions of dollars
annually. It may result in higher HIN rates, more invasive medical treatments, and more
taxes. HSP, HINS, HIP, and other parties that purposefully mislead the healthcare system
Healthcare 2023, 11, 2257 5 of 27

2.3. Healthcare Insurance Fraud

HIN fraud impacts people and companies equally, costing tens of billions of dollars
annually. It may result in higher HIN rates, more invasive medical treatments, and more
taxes. HSP, HINS, HIP, and other parties that purposefully mislead the healthcare system
to obtain illegal benefits or payments may engage in HIN fraud [16,20,22–24]. Frequently
reported forms of HIN fraud are depicted in the taxonomy of Figure 4. Below, we discuss
Healthcare 2023, 11, x FOR PEER REVIEW 7 of 27
these forms of fraud in detail.

Figure4.
Figure 4. Taxonomy
Taxonomy of
of HIN
HIN frauds.
frauds.

2.3.1. The
Healthcare Service Provider
above explanation (HSP) Fraud
demonstrates that HIN fraud may be conducted by HSPs,
HIPs, HINSs,
HSPs are and
the others who purposefully
professionals involved indefraud
providingthe HIN system to earn
and managing illegal benefits
healthcare services.
or payments.
These Hence,
professionals there is
include a needhospital
doctors, to investigate HIN fraud
administration, and provide
ambulances, and a solution by
laboratories.
using cutting-edge technologies. In the subsequent sections, we explore
HSP involvement is a prime factor that facilitates most HIN frauds, even those that HINSs the role of cutting-
edge technologies
commit. Some common in healthcare
HIN fraudsfraudcommitted
detection. by HSPs include: (1) Double-billing—In a
double-billing scam, an HSP submits numerous claims for the same set of services. It is
2.4. Role of
common forAIHSPs
in Healthcare
to double-bill the same entity (such as the government) for the same set
of services. They mayhave
AI approaches obfuscate
showntheir
to beactivities by changing
an effective the service’s
tool for detecting HIN date or description,
fraud. The HIN
or the patient’s or provider’s identity. (2) Phantom billing—Phantom
fraud detection system may be automated using AI. According to recent research, billing is invoicing
AI has
the
primarily been utilized to identify HIN fraud using various techniques and modelsis
HIN or the government for services that were never provided. (3) Unbundling—This
a[32,33].
kind ofAnomalies
fraudulentand activity
fraudinare
which doctors
detected usingcharge for each
behavioral stage of
profiling a multi-part
approaches process
based on
instead of lumping
ML algorithms. To them together
this end, under one code.
each individual’s By separately
behavior is modeled billing eachdeviations
to detect phase, the
HSP
frommaynormsearn more HIN
[34,35]. money while
fraud still providing
detection the sameare
ML approaches level of care.into
classified Unbundling,
supervisedor
“fragmentation,” is breaking up
learning (SL), semi-supervised a single(SSL),
learning medical
and operation
unsupervisedintolearning
many codes (USL). to increase
individual payments.
A dataset The medical
of previously known coding industryand
fraudulent often engages
valid in unbundling.
records is used in the Instead
SL
of utilizing the normal billing code, a surgeon who participates
approach in HIN fraud detection. These data are used to capture fraud patterns and in unbundling could input
different
constructcodes for incision
the model. and suturing
The primary during
advantage of athe
regular surgicalisoperation.
SL approach As a result, a
that the classification
doctor
results produced by this technique are simple to understand. Many classifications code
will receive more money. (4) Upcoding—Using a more expensive billing and SL for
the same medical
regression analysis service is called
methods “upcoding”
are applied in HINor up-charging.
fraud detection. ThisSLmeans that the
techniques HSPtois
used
charging
detect HINpatients
fraudfor a greater
include quality
neural of service
networks, than they
decision are statistical
trees, providing.analysis,
The HSPBayesian
falsely in-
creases their remuneration
approaches, SVM, graphbyanalysis,upcodingand the severity
rule-basedor complexity of the patient’s
systems [36,37]. Apart from condition
its
or treatment. For
advantages, financial
SL has gain,
several an HSP may sometimes
disadvantages. misdiagnose
Data collection and data theirproduction
patients’ condi-
are
complex in SL approaches. When the HIN dataset is relatively large, data labeling
becomes more complicated. It is difficult to distinguish between labels in fraud data when
they are imprecise and ambiguous. Because of these limitations, SL implementation might
be difficult in some instances. These SL limits may be overcome with the aid of USL.
USL algorithms detect fraudulent HIN activities in unlabeled HIN datasets. The
Healthcare 2023, 11, 2257 6 of 27

tions, such as diagnosing with chronic bronchitis when they have acute bronchitis or with a
minor skin lesion when they actually had a bigger, more complex excision. (5) Unnecessary
treatment/drug repetitions—In this fraud, the doctor continues the treatment/drugs for a
long time without needing them. (6) Unnecessary procedures/diagnostics—Sometimes,
doctors recommend costly procedures/diagnostics that are not required (e.g., recommend-
ing a biopsy treatment for the patient with a temporary disorder). (7) Creating forged
prescriptions—HSPs counterfeit or fraudulently modify or create a prescription or other
document in this scam. Two examples are adding a medicine or changing the date, amount,
or dispensing instructions. (8) Kickback—This is a term used to describe a misuse of
money that benefits a person of authority to use his position or influence to benefit another
individual, organization, or enterprise. It is a prevalent deception in the HIN area, where
HSPs such as pharmaceutical corporations incentivize other HSPs to suggest certain drugs.
(9) Billing for non-entitled services as entitled—This fraud usually happens when an HINS
is not entitled to a particular service by the HIP and the doctor provides him this service,
putting this treatment in another category entitled to him.

2.3.2. Healthcare Insurance Subscriber (HINS) Fraud

The HINS is the one who purchases HIN policy from the private- or public-sector
organization. Most HIN fraud comes from HINSs seeking to obtain financial benefits.
HINSs perform HIN fraud in different ways: (1) The HINS may change prescriptions by
himself or with the help of an HSP. (2) Sometimes, HINSs make false claims and try to
receive money for a treatment they never received. (3) Another method of HIN fraud is
claim fixation, in which both the HINS and HSP are involved, modifying the claim amount
to obtain mutual financial benefits. (4) One of the most commonly occurring types of
fraud that significantly affect HIPs is identity swapping, in which an HINS’s insurance
card/policy is used by someone else to access healthcare facilities. (5) HINSs sometimes
receive unnecessary services due to the availability of HIN (e.g., the HINS may visit the
HSP and demand unnecessary diagnostics). (6) Another fraud committed by HINSs is
medical equipment fraud, in which the HINS tries to obtain something more than what is
required (e.g., the HINS needs a manual wheelchair, and he applies for a fully-automated
wheelchair). (7) Age/gender inconsistency is another form of fraud, in which an HINS
tries to hide his age/gender to obtain additional benefits for him or someone else. These
are the possible methods of fraud initiated by the HINS. (8) Last but not least is the misuse
of a healthcare card, in which a person uses his healthcare card for his family members or
friends with the consent of the HSP and obtains the treatment cost from the HIP.

2.3.3. Healthcare Insurance Provider (HIP) Fraud

HIPs conduct HIN fraud by misleading, hiding, or distorting information, although
not to the level that HSPs and HINSs are engaged in healthcare fraud. The reason for
not being involved in HIN fraud is to maintain the reputation of the HIP’s company and
customer attraction. Still, some HIP fraud occurs when an impostor HIN agent attempts
to sell you a false HIN plan or when a fraudster contacts an HINS to verify personal
information for his plan and then steals the HINS’s identity. HIN firms may sometimes
perpetrate fraud by fabricating medical equipment claims, particularly for older people or
persons with disabilities, who often need a large amount of durable medical equipment.
Sometimes, an HIP deliberately delays the claim processing and expects some incentive for
the quick processing of HIC.
The above explanation demonstrates that HIN fraud may be conducted by HSPs, HIPs,
HINSs, and others who purposefully defraud the HIN system to earn illegal benefits or
payments. Hence, there is a need to investigate HIN fraud and provide a solution by using
cutting-edge technologies. In the subsequent sections, we explore the role of cutting-edge
technologies in healthcare fraud detection.
Healthcare 2023, 11, 2257 7 of 27

2.4. Role of AI in Healthcare

AI approaches have shown to be an effective tool for detecting HIN fraud. The HIN
fraud detection system may be automated using AI. According to recent research, AI has
primarily been utilized to identify HIN fraud using various techniques and models [32,33].
Anomalies and fraud are detected using behavioral profiling approaches based on ML
algorithms. To this end, each individual’s behavior is modeled to detect deviations from
norms [34,35]. HIN fraud detection ML approaches are classified into supervised learning
(SL), semi-supervised learning (SSL), and unsupervised learning (USL).
A dataset of previously known fraudulent and valid records is used in the SL approach
in HIN fraud detection. These data are used to capture fraud patterns and construct the
model. The primary advantage of the SL approach is that the classification results produced
by this technique are simple to understand. Many classifications and SL regression analysis
methods are applied in HIN fraud detection. SL techniques used to detect HIN fraud
include neural networks, decision trees, statistical analysis, Bayesian approaches, SVM,
graph analysis, and rule-based systems [36,37]. Apart from its advantages, SL has several
disadvantages. Data collection and data production are complex in SL approaches. When
the HIN dataset is relatively large, data labeling becomes more complicated. It is difficult to
distinguish between labels in fraud data when they are imprecise and ambiguous. Because
of these limitations, SL implementation might be difficult in some instances. These SL
limits may be overcome with the aid of USL.
USL algorithms detect fraudulent HIN activities in unlabeled HIN datasets. The ab-
sence of tagged data is an advantage of HIN fraud detection via USL [38,39]. When labeled
data is lacking, USL is utilized to identify HIN fraud. HIN fraud detection employs various
USL approaches, including association rules mining, k-means clustering, data mining,
and k-nearest neighbor. SSL combines the advantages of both SL and USL. Compared to
unlabeled data, hybrid SSL is employed when a limited amount of labeled data is available.
In SSL, a predictive model is developed utilizing both labeled and unlabeled data [40].
In HIC fraud detection, combined clustering and classification as an SSL approach is
often utilized.

2.5. Role of Blockchain in Healthcare

A BLC is a network of interconnected blocks that serves as a digital ledger (DTL) of
transactions [41]. Each block contains a timestamp, block number, block header, version,
nonce, transaction data, and a cryptographic hash of the preceding block. A Merkle tree is
used to preserve transaction data [42]. In a BLC, if any of the transactions in a block are
changed or amended, the hash value of that specific block will be drastically affected. It
also disrupts the BLC, making it easier to spot modified transactions. As a result, once a
transaction has been put to the BLC, it cannot be changed or amended. Data on the BLC
is thus unchangeable. HIN applications benefit from BLC as a decentralized architecture
since it enables distributed HIN application activities independent of a centralized server.
Transparency among all parties engaged in the HIN process is created via the replication
of DTL data across all nodes [36]. Due to the fact that the data in the DTL is duplicated
across all nodes in the network, an attack that takes place on any one node in the BLC
network has no effect on the state of the DTL. The preservation of unalterable HIN data is
one of the prerequisites for an HIN fraud detection system [43,44]. The BLC’s immutability
characteristic satisfies the aforementioned need. It is essential to maintain the validity and
integrity of the HIC records. Cryptographic methods are used to encrypt the HINS’s data
on the BLC, ensuring that only an HINS with legitimate authority may access and decode
the data. It significantly improves the HIN’s data security and privacy. Since the identities
of HINSs on a BLC are preserved using cryptographic keys, HINS data may be shared
among all parties engaged in the detection of HIN fraud without disclosing the names of
the HINSs [45]. BLC technology enables smart contracts, which are used to control activity
after an agreement and are written as executable code, developed in Solidity language
and implemented on Virtual Machines [46]. The whole HIN business process is governed
Healthcare 2023, 11, 2257 8 of 27

by the transaction logic included in it. Once the transaction saves the operation, smart
contracts are put into effect. In order to govern how their HIN data are shared or utilized
in the network, HINSs may specify rules for the HIN process using a smart contract.
The working and roles of cutting-edge technologies in HIC processing and fraud
detection are discussed in the above sub-sections; now, we explore the existing stud-
ies on using cutting-edge technologies for HIC fraud detection and prevention in the
following subsection.

2.6. Role of 5G and Cloud in Healthcare

Fifth generation (5G) is a term referring to wireless network technology, and its use in
the healthcare industry has created new opportunities for medical innovation and increased
access to treatment. It is an air interface that is unified and powerful and was developed
with an improved capacity to enable the user experiences and services of the future genera-
tion. The technology of 5G networks is one of the important technologies for the digital
transformation of society [47–49]. It is also a requirement for the interconnectedness of
everything involved in smart healthcare. It is possible to decrease discrepancies in the
allocation of medical resources and hasten medical developments via the promotion and
implementation of 5G smart healthcare [50]. The use of cloud computing in the healthcare
industry makes the exchange of medical records simpler and more secure, automates back-
end processes, and even makes it easier to develop and maintain telehealth applications.
The healthcare business can reduce expenses while improving operational efficiencies
by using cloud computing [51,52]. The performance of the healthcare business will be
improved by 5G when combined with the cloud since cloud-sharing of large files will
become quicker and more efficient with 5G.

2.7. Related Work

This section will provide an overview of recent studies that have used AI or BLC or
both to identify and mitigate HIN fraud to provide a picture of the state of the art.
A rigorous assessment of AI- and BLC-enabled secure HIN fraud detection is presented
by Kapadiya et al. [17]. This study offers a taxonomy of numerous HIN security challenges
and a BLC- and AI-based safe and intelligent system for detecting HIN frauds. A case
study involving HIN is also included to provide an overview of the proposed approach.
Finally, it discusses outstanding difficulties, research hurdles in BLC implementation, and
an AI-powered HIN fraud detection system.
According to Saldamli et al. [5], HIN firms are utilizing an antiquated approach to
keep track of their client’s health care data and insurance-authorized information. As
a result, several scams may occur in the HIN domain. The consumer attempts to claim
the HIN from two distinct parties by defrauding the HIN businesses. As a result, there
is a need to develop a secure system to monitor all of a user’s HICs to identify fraud.
This article proposes a solution prototype that eliminates intermediaries and makes use
of BLC technology to store and track data. Once data is written on a BLC, it cannot be
changed or erased; this aspect of BLC ensures data integrity while identifying fraud. If any
modifications are made to the BLC node, all further changes must be performed on the
other nodes. BLC also maintains a record of log actions, which administrators may examine
at any time. Despite its drawbacks, according to research results, BLC technology is one of
the most popular strategies utilized today in the healthcare field. As a result, BLC enables
us to solve many of the issues faced by the healthcare business, such as data integrity, data
security, and fraud detection.
According to Ismail and Zeadally, HIN fraud detection is a big challenge for healthcare [24].
HIN fraud costs billions of dollars each year. Given the significance of the issue, this
paper proposes Block-HI, a BLC-based framework for detecting HIN fraud based on a
proposed taxonomy of 12 different fraud scenarios. When the number of HIN branches and
HICs transactions increased, the performance of Block-HI was tested in terms of execution
time and total quantity of data sent for ledger updating. The results show that when the
Healthcare 2023, 11, 2257 9 of 27

number of HIN branches rises, the execution time of Block-HI decreases by just 0.69% on
average, although the number of claims increases by 33.51%. This is because the consensus
procedure was executed. Despite the fact that performance degrades as the number of
branches grows, the suggested framework, which automates fraud detection, can uncover
distinct fraud situations compared to a human procedure.
Amponsah et al. suggest a method for detecting and preventing fraud in HIC
processing utilizing ML methods and BLC technology [53]. The original HIC dataset
is classified using a decision tree in this work. The retrieved information is encoded
in an Ethereum BLC smart contract to identify and prevent healthcare fraud. Accord-
ing to the trial data, the top-performing tool has 97.96% classification accuracy and
98.09% sensitivity. The study’s results demonstrate that the suggested solution improves
the capacity of BLC smart contracts to identify fraud with an accuracy of 97.96%.
According to Gera et al., several concerns are linked to the HIN sector, such as
fake HICs and claims unlawfully influenced by competent authorities [54]. This article
addresses these concerns by building a BLC-based HIN application. The consensus
guarantees that the HIN company’s claim procedure is carried out securely. In particular,
every transaction in the proposed approach is cryptographically signed and kept on
the BLC as a collection of blocks. This method protects HIC transactions and prevents
any fraudulent efforts. The IBM BLC technology and its underlying components are
used to create a prototype application. The experimental findings demonstrated that the
suggested implementation reduces HIC fraud.
According to Goyal et al., having health insurance is vital for everyone, especially
with rising medical expenditures, since medical crises may have a significant financial
and emotional effect [55]. However, the existing insurance system is too costly, and
the claim settlement procedure is overly protracted and cumbersome. Consequently,
policyholders cannot file an HIC with their insurance carrier. The authors of this study
provide a rapid and cost-effective framework for an HIN business using BLC technology
and ML. By constructing a smart contract, BLC may eliminate intermediary organiza-
tions, making the whole process easier, safer, and more efficient. The smart contract
pays the HIC based on the HINS’s documentation. A ridge regression model is used to
best compute premiums based on the total amount claimed during the current policy
duration and various other criteria. A random forest classifier is used to anticipate risk,
which aids in the calculation of risk-rated premium rebates. The suggested model’s
findings demonstrate that it is dependable, cost effective, and rapid. The whole HIC
processing method with fewer intermediaries and smart contracts makes the procedure
faster, and risk-rated premium rebates assist in incentivizing those with lower claims to
keep their insurance. It also prevented the additional cash created in the present HIN
system by dispersing it in risk-based rebates. Results also reveal that the random forest
classifier delivered higher accurate results out of the numerous multi-class classification
ML approaches.
Amponsah et al. suggest a BLC-based method to improve the financial sustainability
of HIN systems [56]. Poor data management, lengthy lines of communication, lack of
transparency among stakeholders in HIC processing, fraud, and corruption are the issues
confronting the HIN and jeopardizing its viability. This paper suggests a cloud-based
BLC approach to overcome these issues. According to the findings of this work, there
is a clear demonstration of communication line reduction, efficient management and
storage of data using the cloud, efficient HIC processing lifecycle, restricted data access,
hypothetical reduction in HIC processing, security, and high transparency in the system.
After evaluating the system’s read and write operations, it can be concluded that it
accomplishes its goals since all activities were completed in the world state and BLC
systems, respectively. The study also emphasizes the significance of information, service
quality, and user happiness in effectively implementing the cloud BLC-based solution.
The above discussion shows that cutting-edge technologies play vital roles in
e-health and HIC processing. However, existing studies mainly focus on improving
Healthcare 2023, 11, 2257 10 of 27

HIN acceptability and the security of HIN systems. Furthermore, existing studies
mainly focus on a particular technology instead of leveraging the potentials of com-
bination of technologies. We did not find any study that explicitly addresses fraud
detection and mitigation in the end-to-end HIC processing cycle. To address this gap,
we propose an end-to-end security model for HIN fraud detection using cutting-edge
technologies in the next section.

3. Proposed Methodology SHINFDP

To address fraud detection and mitigation for HICs, we proposed an end-to-end
layered security framework named SHINFDP, as shown in Figure 5. The proposed
SHINFDP consists of six layers: user layer, communication layer, storage layer, predic-
tion layer, security layer, and HIC processing layer. According to SHINFDP, a patient
approaches the healthcare center (HC) online or physically. The patient’s query is con-
sidered by the relevant staff in the HC, and the particulars of the patient are stored in
SHS. The patient is referred to the relevant HSP based on the stored particulars. The
HSP provides the treatment to the patient and adds the details of treatment in the SHS,
from which another HSP accesses the patient’s details for preparing the HIC. Once the
HIC is prepared, it is stored in the cloud, from which the prediction layer accesses it. The
prediction layer ensures the claim’s validity and stores it in the BLC. The HIP accesses the
HIC through the prediction layer and performs the necessary actions. The transactions’
details are stored in the smart contract of the BLC; each layer of SHINFDP communi-
cates with the BLC using the smart contract. Below, we describe the functionality of
each layer.

3.1. User Layer

Three main users are involved in the proposed SHINFDP, including the HIP, HSP,
and HINS. The HINS is the insured patient who initiated the system by approaching
the HC for treatment. The insured patient can visit the HC in person or may contact
the HSP using a smart healthcare system (SHS). The HSP concerned with handling
patients assesses the HINS, checks his insurance policy, and sends him to the relevant
HSP for receiving treatment. The HSP includes various healthcare professionals (e.g.,
doctors, paramedic staff, diagnostics/lab staff, pharmacist, etc.). The HSP performs the
treatment according to the patient’s situation and enters the treatment details in the
SHS, from which the HSP concerned with HIC processing fetches the data and prepares
the HIC. The prepared HIC is approved by the relevant authorities and is stored in the
healthcare cloud using a proper encryption mechanism. The prediction layer responsible
for detecting legitimate HICs and fraudulent HICs provides classification of the HIC and
stores the legitimate HIC in the BLC after the execution of the BLC consensus mechanism.
A smart contract is created for all the legitimate HICs. The HIP accesses the legitimate
HIC and processes it according to the patient’s HIN policy. The claim record is also
stored in the BLC’s digital ledger (DTL), from which each party can access it. The user
layer uses the services of various other layers; this layer uses 5G services for fast data
transmission between multiple entities. Furthermore, this layer is connected to a storage
layer, where the HIC is stored for further processing. This layer is also associated with
the security layer, as all user records are stored in the BLC. The details of the tasks
performed by all three categories of users are outlined in Algorithm 1.
Healthcare 2023, 11, 2257 11 of 27

Algorithm 1: Working of user layer

Let ρ = H IS appraoches HC physically; HC = Healthcare Cloud
E = Encryption; CA = Claim amount; LH IC = Legitimare health insurance claim
Begin
1. H IS → HC ⇑ (5G, ρ)
2. HSP → EnterRecord( H IS) → SHS
3. HSP → Re f er ( HSP) → T
4. H IS → Recieve( T ) → HSP
5. HSP → EnterIn f o ( T ) → SHS
6. HSP → Prepare( H IC )
7. HSP → Validate( H IC ) → HSP
8. HSP → Forward( H IC ) ⇑ (5G, E) → HC
9. Add HIC →Blockchain
10. H IP → Recieve( LH IC )
11. H IP → Process( LH IC )
12. H IP → Dispatch(CA) → ( H IS, HC )
End

3.2. Communication Layer

This layer leverages the potential of modern 5G technologies to provide fast data trans-
mission between multiple layers. As proposed, SHINFDP provides end-to-end security
for HIC processing; therefore, 5G services are vital as they will provide massive capacity,
low latency, and high speeds. The HINS will use 5G to communicate with the HSP and
HIP and to track the HIC. The HSP will use 5G to transmit and store patients’ data and
the corresponding HIN policy. Furthermore, 5G will be involved in data transmission
between the SHS and the cloud. The prediction layer will also use 5G services to fetch and
analyze data. The security layer will use 5G’s potential to execute consensus mechanisms,
generate smart contracts, and manage the BLC network. The HIC processing layer will use
5G potentials to fetch legitimate HIC for processing and dispatching the HIC amount to
the concerned authority. Algorithm 2 elaborate the working of communication layer.

Algorithm 2: Working of communication layer

1: Input : Data D, Source address S, destination address T

2: Output : Find best routing path with low latency LL and High speed HS
3: Procedure
4: Bu f f er = 0
a. If (bu f f er <= bu f f er.Length) then
b. Receive ( D, S, T )
c. Find best routing path(↓ LL, ↑ HS)
d. transmission( D, S, T )
e. else
f. empty(bu f f er )
g. Go to Step c
h. End if
End Procedure

3.3. Storage Layer

This layer of SHINFDP leverages the potential of cloud computing to store HIC data
on the cloud securely. As the proposed SHINFDP is smart, the data collection mode
will be online only. All the data generated by the HSP will be stored on the SHS. The
HSP responsible for HIC generation will fetch data from SHS, prepare the HIC, and store
it on the cloud. The data generated by this layer include payment-related sensitive in-
formation of the HINS and HIP so that it will be shared and controlled through smart
contract conditions for security considerations. The prediction layer will fetch data from
Healthcare 2023, 11, 2257 12 of 27

this layer for classifying legitimate and fraudulent HICs. The HIP will fetch the data of le-
gitimate HICs from this layer for further processing. Working of storage layer is elaborated
in Algorithm 3.

Algorithm 3: Working of storage layer

1: Input : H IC Data
2: Output : Secure storage o f data in the cloud and Blockchain
3: Procedure
HSP → FetchData ← SHS
Generate ( H IC ) ← SHS
Validate( H IC )
SmartContract( H IC )
H IC → BlockChain
End Procedure
Healthcare 2023, 11, x FOR PEER REVIEW 11 of 27

Figure 5.
Figure 5. Proposed
Proposed methodology.
methodology.

3.4.
3.1. Prediction
User LayerLayer
This
Threelayer
mainof users
the proposed SHINFDP
are involved in thewill leverageSHINFDP,
proposed AI’s potential to classify
including HICsHSP,
the HIP, into
legitimate
and HINS.and Thefraudulent
HINS is theHICs. This
insured layer who
patient will fetch HICthe
initiated data fromby
system theapproaching
cloud and will
the
clean the data for redundancy, missing values, and incorrect information.
HC for treatment. The insured patient can visit the HC in person or may contact The process of
the HSP
using a smart healthcare system (SHS). The HSP concerned with handling patients
assesses the HINS, checks his insurance policy, and sends him to the relevant HSP for
receiving treatment. The HSP includes various healthcare professionals (e.g., doctors,
paramedic staff, diagnostics/lab staff, pharmacist, etc.). The HSP performs the treatment
Healthcare 2023, 11, 2257 13 of 27

preparing data for analysis by deleting or changing data that is erroneous, incomplete,
irrelevant, redundant, or badly formatted is known as data cleaning. In order to clean
the data, suitable data cleaning technique of ML will be used. Data cleaning is one of
the important steps in ML; this process will detect missing or null values, outliers, and
erroneous data. The claim amount range for each treatment will be determined and
saved in the healthcare cloud repository. To identify outliers, the data’s lower and upper
bounds will be computed. Outliers are values that are both below and above the lower and
upper bounds. This will be accomplished by employing the appropriate ML algorithms
to find outliers in the data. The cleaning process will reduce the data size; after that,
data from different sources will be integrated and transformed into a uniform format for
preprocessing. Once the data is transformed, ML algorithms will be applied to this data
to classify it into legitimate and fraudulent data. The techniques used at this layer can
be SL, USL, or SSL, depending on the nature of the data. AI data analytics is the key to
minimizing the frequency of HIN fraud detection and the expense of false claims filed
against HIN policies. The working of prediction layer is elaborated in Algorithm 4.

Algorithm 4: Working of prediction layer

1: Input : H IC Data
2: Output : Classi f ication o f claim
3: Procedure
FetchData ( H IC ) ← Cloud
CleanData( H IC )
PreProcessData( H IC )
Trans f ormData( H IC )
ApplyMLalgo
Classsi f y( H IC ) → ( Legitimate, malicious)
End Procedure

3.5. Security Layer

The security layer is one of the critical layers of SHINFDP, which leverages the potential
of BLC to prevent improper manipulation of the HIC. BLC is a series of blocks, each of
which is nothing more than an encrypted real-world transaction. It is a transactional DTL
that is open-source and community based. It is a peer-to-peer mechanism that guarantees
openness and accountability. Each transaction is stored as a block in the DTL and may
be programmatically validated by numerous users connected with the application. Every
node has access to a copy of the BLC. It is built such that no one entity or node may conduct
fraud. Every node utilizes the same BLC, and no conflicting versions are possible. There is a
continual verification process with mining, and the consensus intends to provide evidence
for every transaction. This technology is unusual in that it can keep data immutable
and does not rely on any central authority. As a result, it maintains transaction integrity
and may effectively prevent malpractice or hacking. It can also replace conventional
organizations while providing a greater degree of security. The HIC procedure involves
many parties, including HIP, HINS, and HSP. Each participant will have a peer node in
the BLC’s peer-to-peer network. The HIP utilizes the HSP peer, which is responsible for
HIC processing. The prediction layer requires a peer to complete the verification procedure.
An HSP peer may help with claim development and verification. All of these peers will
be allowed to participate in HIC processing, and the transactions will be stored in a DTL.
Each transaction is digitally signed and recorded on the BLC. In a distributed context, an
irreversible collection of transactions is therefore preserved. Furthermore, smart contracts
are generated for each transaction and are stored on the BLC. A smart contract allows each
layer of SHINFDP to interface with the BLC layer. Smart contracts are programs for the
authentication and payment of HIN coverage and are automatically executed when all the
conditions are met. Consensus algorithms are used in smart contracts to authenticate every
transaction and input. A smart contract is connected with the IPFS to reduce total data
storage and quick data retrieval costs. The benefits of using a smart contract for HIN fraud
Healthcare 2023, 11, 2257 14 of 27

detection include transparency, task automation, time saving, protection of HIN policy
documents, and risk assessment. The detailed working of the security layer is outlined
in Algorithm 5.

Algorithm 5: Working of security layer

Let ST = Smart Contract; HCl = Healthcare Cloud; E = Encryption;
CA = Claim amount; LH IC = Legitimare health insurance claim; T = Treatment;
NB = New block; BCT = BlockChain
Begin
Step 1. H I N Commitment
H IP → ProvideDetails( H I N ) ← H IS
H IS → Purschase( H I N ) ← H IP
H IP → CommitTransaction( H I N ) ← H IS
Generate(ST ) ← CommitTransaction
Add( NB) → BCT
Step 2. H IC Generation
H IS → Request( T ) ← HSP
HSP → Handle( T )
HSP → Input( T ) → SHS
HSP → Prepare( H IC )
HSP → Validate( H IC ) ← HSP
HSP → Forward( H IC ) ⇑ (5G, E) → HCl
Generate(ST ) → T
Add( NB) → BCT
Step 3. H IC Processing
H IP → Recieve( H IC )
H IP → Query( H IC, HCl )
H IP → Apply(ST ) ← H IC
I f ( H IC ∈ LH IC )
H IP → Process( LH IC )
H IP → Re f und(CA) → ( HSP, H IS)
else
H IP → Re f use( H IC )
H IP → SignTransaction()
Add( NB) → BCT
End

3.6. HIC Processing Layer

The key stakeholders of this layer are HIPs, each of which can be a private or a
governmental firm responsible for paying health services directly to the HINS or the HC.
The HIN processing layer is connected with all the other layers of the proposed framework.
It is associated with a storage layer to fetch HIC information from the healthcare cloud. It
communicates with the prediction layer to obtain information about claim authenticity. It
uses security layer features to ensure HIC authenticity through smart contracts. This layer
uses 5G services of the communication layer for data fetching and transmission. This layer
is also connected with the HC and HINS to inform them about the results of HICs.

4. Framework Evaluation
In this section, we will show the details of framework evaluation using various
methods. We first evaluated the proposed framework using mathematical modeling. The
feasibility of the framework was assessed with the help of a focus group by evaluating
various parameters using concept mapping and Likert scale evaluation, and finally, a
case study of wearable devices was used to further elaborate and evaluate the proposed
framework. Below, we discuss these evaluation results in detail.
 Healthcare 2023, 11, 2257 15 of 27

4.1. Mathematical Modeling

Below, we
Healthcare 2023, 11, x FOR PEER REVIEW explain the evaluation and step-wise working of SHINFDP. Notations part 16 of 27
in Abbreviations provides the list of mathematical notations used in the model.
Let the smart For efficient HIN process ℘ , the ℘ take transmission patient data of 𝚰 ρatasmaximum input, transmit speedit 𝑺(𝒙) to theand smart minimum
care 2023, 11, x FOR PEER REVIEW healthcare latency system 𝓛(𝒙) S 𝒽 from
is required, where it which is stored is on
only the healthcare
possible if 𝓖
cloud is H
implemented
𝒸
16 of 27 , and the hashand ofworking
, x FOR PEER REVIEW the data is
properly,added in
as to
shown the blockchain
in Equation B 𝒸 .
(5). The machine learning 16 of 27 algorithm M 𝓁 is applied
For efficient to this℘data , the transmission
to extract legitimate of 𝚰 Cat . The maximum HIN provider speed H𝑺(𝒙) 𝓅 tries and to maximizeminimumthe ℘ by only
𝓗𝓬 ⇔ 𝚰𝝆 ⋯ 𝓛𝓲
latency 𝓛(𝒙)
For efficient ℘ , processing
the transmission
is required, legitimate
which of is𝚰 Cat . Thus,
only maximum our objective
possible speed if 𝓖 isfunction 𝑺(𝒙) implemented and𝕳 will 𝓖 minimum be as and follows: working
𝓜𝓵 = 𝕺𝖃 ⇔ ⋮ ⋱ ⋮ (6)
latency 𝓛(𝒙) is required,
properly, as shown Objective
which in Equation is only Function possible
(5). OF =if max 𝓖 is ℘( x)where
implemented and working 𝓓 ⋯ 𝓕
Healthcare 2023, 11, x FOR PEER REVIEW 0 ≤ x ≤ n 𝓬 𝓬
properly, as shown in Equation (5). 𝚰𝝆 ⋯ 𝓛𝓲
𝓗𝓬 ⇔
1, x FOR PEER Healthcare
REVIEW2023, 11, x FOR PEER REVIEW The stored claim 𝕳data 𝓖 𝕺 𝖃 will be fetched from 𝓗𝓬 via 𝓖 by ensuring its security
16 of 27 16 of 27
℘  𝚰 ρat= maximum true (6) speed 𝑺(𝒙) and m

𝓜 𝓗𝓵𝓬 ⇔ = For 𝕺 𝖃 𝚰
efficient
𝝆 ⋯ ⇔ 𝓛 ⋮
𝓲 , the ⋱ ⋮
transmission n of
using 𝕳. This𝕳 input 𝓖 data will be decrypted and using suitable ML techniques it will be
For efficient ℘ , the transmission 𝓜For𝓵 = 𝕺𝖃℘latency
efficient =of℘⇔ 𝚰 ,ρat
into legitimate 𝓛𝓲 and
,𝓛(𝒙)
S⋮maximum
the H is⋱𝒸 ,required,
𝒽 ,transmission G𝓓 ⋮, M 𝓬 speed 𝓁⋯ ,B 𝒸𝓕
ofwhich ,𝑺(𝒙)
𝚰𝓬 Cat =⇒ isand
maximum only ∑ minimum Lspeed
possible
𝓬 , as(6) 𝒾 = true if𝑺(𝒙) 𝓖 isand minimum
implemented (1) and
latency 𝓛(𝒙) is required, latency which𝓛(𝒙)
classified
is only isproperly, possible
required, 𝓓as𝓬 which if⋯𝓖 𝓕
shown isisin implemented
only
fraudulent
Equation possible (5).
claim
and if 1𝓕is
c=𝓖working F
shown in Equation (6).
=
implemented
𝒸 false and working
𝓬
The stored claim data 𝕺𝖃 will be fetched from 𝓗 via 𝓖 by ensuring its security
properly, as shown in Equation properly, (5). as shown in Equation (5).𝒚𝒊𝒆𝒍𝒅𝒔 𝓬 𝓓𝓵 = 𝓓𝓵 + 𝕳, 𝒙 = 𝟎 𝓒𝓶
𝚰it𝝆 will ⋯ be𝓛𝓲
The stored using 𝕳. claim Thisdata input 𝕺𝖃data will will beconstraints
fetched
be decrypted from of ∝ and 𝓗S𝓑𝓬 ,= viausing ⎯⎯𝓖 by 𝒇(𝒙)
suitable = ML its
ensuring techniquessecurity
𝓗 ⇔ (7)
Under the 𝒽 M𝓁 , B𝒸 , G , H𝒸 𝓓𝓵 = 𝟎 ∩ 𝓬+𝕳 𝕳 𝓖, 𝒙 ≥ 𝟎 𝓢
using 𝕳. This input into
classified datalegitimate
will be decrypted 𝓛𝓲 and𝓗fraudulent 𝓬 and
⇔ using𝚰 𝝆 ⋯
suitable
claim 𝓛 𝓕 𝓲 𝓬 , ML
as 𝓗 𝓜
techniques
shown⇔ 𝓵 =
in 𝚰 𝝆 𝕺
Equation𝖃 ⋯ it will𝓛 𝓲 ⇔ be
(6). ⋮ ⋱ 𝓬 ⋮
𝕳 𝓖 𝓬 𝕳 𝓖 16 of 27𝓓
1, x FOR PEER REVIEW
classified into legitimate 𝓛𝓲 and 𝓜fraudulent
𝓵 =According𝕺𝖃 ⇔ to Equation 𝓕⋮ 𝓬V 𝓜 ⋱shown 𝓵· ·= ·⋮ 𝕺the
(7), blockchain
S𝖃in𝒽 Equation ⇔H ⋮ (6).digital ·⋱· · ⋮ B ledger (6) 𝓬 𝓓⋯ 𝓵 will 𝓕𝓬add the hash (6) of the

claim , as
𝒚𝒊𝒆𝒍𝒅𝒔 𝓓𝓵 = 𝓓𝓵 + 𝕳, 𝒙 = 𝟎 𝓒𝒸𝓶
i 𝒸
data 𝓑 by = ⎯⎯
executing 𝒇(𝒙) = 𝓓𝓒 . and⋯ 𝓕
will .
generate a 𝓓. 𝓢 ⋯
that 𝓕
will . be populated by all the
(2) entities
Healthcare 2023, 11, x FOR PEER REVIEW 𝓬 . . (7) 16 of 27
For efficient ℘ , the transmission 𝒚𝒊𝒆𝒍𝒅𝒔 𝓬 of𝓓𝓵The 𝚰 ρ=at= 𝓶
𝓓stored
maximum

+𝓓.𝓵𝕳,=𝒙 claim𝟎= . .∩𝓬𝟎speed +𝕳 𝓒.. 𝓶, 
data

𝒙𝕺 ≥𝑺(𝒙)
n 𝖃𝟎 will
 ..𝓢and 𝓬𝓬beminimum fetched 𝓬 . from 𝓗 via 𝓖 by ensuring

its
𝓵 𝓬 M𝓁 .  𝓬
𝓑 = 𝕺 ⎯⎯ involved 𝒇(𝒙) = to secure 𝚰 and
if𝕺input 𝓗 𝕺 . 𝓖𝚰𝓢fetched (7)
The𝓛(𝒙) 𝕳. 𝓖 will ·𝓗 L𝓖
𝓬 𝝆 𝖃
latency stored claim
is required,
PEER For
data which The
efficient𝖃 will stored
is only beusing
℘ claim
, fetched
the 𝓵 =
𝓓possible 𝟎This
data
transmission∩ +𝕳
from T o𝖃, 𝒙digital
𝓬·is≥
·via 𝟎be
data
·ofimplemented Cat
bywillensuring
𝓬 maximum
from
be decrypted
Gwill andits ·𝓬security
·working
speed via and 𝒾𝑺(𝒙)
by and
using ensuring
suitable
minimum itsML security
techniques i
Healthcare 2023, 11, x FOR
using 𝕳. This According REVIEWto Equation
𝕳. (7), the blockchain ledger
𝓛 𝓓fraudulent
𝓵using add the hash 𝓕 of the 16 of 27
properly, as shown input indata using
Equation
latency 𝓛(𝒙) will be (5).
isandThis
decrypted
required, input
classified and
data
which using
will
into is𝓢 be suitable
legitimate
only decrypted possible ML 𝓲 𝓣
techniques
and
and ⋯
𝓸if 𝓖 is implemented 𝓗 suitable
it
𝓹
will claim be ML 𝓬 ,techniques
as shown
and working init will
Equation be (6).
Accordingby to executing
Equation 𝓒and
(7), 𝓶the blockchain digital ledger 𝓓 will add the hash of allthe
classifieddata will generate 𝓕𝓲a𝓬 and that will be populated by the entities
𝓵 in 𝒚𝒊𝒆𝒍𝒅𝒔
into legitimate 𝓛𝓲According
classified For fraudulent
into legitimate
efficient
to Equation ℘ , 𝚰𝓛
claim the
(2), ,patient
as 𝓬 shown
transmissionfraudulent data 𝚰 = Equation
input of ⋮
claim 𝚰 ρat ⋱ 𝓕maximum
will 𝓬(6).
, asbe ⋮ shown ⎯⎯
validated 𝓛in𝓲 Equation
speed by first 𝑺(𝒙) (6).
and minimum
checking that
data by executing involved 𝓒 properly, and will
to𝓶secure as shown
𝚰𝝆 and generate in Equation
𝕺𝖃 . 𝓗aexists 𝓬⇔
(5).
𝓢𝓬𝓖 that𝝆 will be populated ⋯ 𝓛 𝓲 ℂ 𝒚𝒊𝒆𝒍𝒅𝒔
by 𝒇(𝒙) all the 𝓓
entities 𝓵 = 𝓓 𝓵 + 𝕳, 𝒙 = 𝟎 𝓒𝓶 (8)
the patient’s
latency 𝓛(𝒙)
record is 𝕳
required, in S and
which that is 𝓑
he/she
only𝓬 = 𝚰 ⎯⎯ has
possible ⋯ a valid 𝓗 = if 𝓖
insurance
is implemented policy; after and that, he
working
1, x FOR PEER REVIEW involved to secure 𝚰𝝆 and 𝓜
𝒚𝒊𝒆𝒍𝒅𝒔 = 𝕺 𝓓 ⇔ = 𝓓 ⋮ +
𝒽𝒚𝒊𝒆𝒍𝒅𝒔 ⋱𝕳, 𝒙 ⋮
= 16 𝟎 of 𝓒 27 𝓓 𝝆 = 𝓓 + 𝕳, 𝓬 𝒙 𝓓
16 = of=
𝟎 (6) 𝟎
27 𝓒
𝓕 𝓬∩ +𝕳 , 𝒙 ≥ 𝟎 𝓢
𝓑𝓬𝕺 .⎯⎯𝓵 𝒇(𝒙)𝖃the 𝓵 𝓵 𝓶𝚰𝓵 ⋯ 𝓵 𝓛 𝓵 𝓶
will be =𝖃provided = required 𝓣𝓑 𝓬 = ⋯⋯ ⎯⎯ 𝓗𝒇(𝒙) 𝓗𝓬 ⇔ =𝓖 after 𝝆 𝓲 (7) T ℴ , insurance𝓬 claim(7)
properly, as shown 𝓓𝓵 = in
𝓜𝟎𝓓 𝓸 ∩𝓬 treatment,
Equation +𝕳
= 𝕺 , 𝒙𝓕(5). ≥𝓬be
𝓹 𝕳and
𝟎𝒚𝒊𝒆𝒍𝒅𝒔 ⇔ 𝓢 𝓓⋮𝓵 =obtaining 𝟎
⋱ ∩ +𝕳 𝓗⋮ 𝒙treatment
, with ≥ 𝟎the𝓢help
or efficient ℘ , the For transmission
efficient of ℘ , 𝚰the Catwill maximum
transmission
be made. The
𝓣speed
insurance
𝓸This of𝚰ℂ⋯= 𝚰 C𝑺(𝒙)
According
𝓗
claim
𝓵 and
⋮𝓹maximum
atneeds ⋱ totominimum
will
𝖃Equation
be ⋮ stored speed validated
⎯⎯ on
𝓬
𝓛𝓗 𝑺(𝒙)
(7), the the by and blockchain
healthcare 𝓹
minimum cloud digital 𝓬
H𝒸ledger of patient
(8) G
using 𝓓,𝓵and (6)
input
will add
the data the 𝚰ha𝝆,
𝓓 𝓲 ⇔ ⋯ 𝚰𝓕 ⋯ 𝓛
The stored claim datawhich 𝕺 ifthe 𝓖obtained treatment 𝒚𝒊𝒆𝒍𝒅𝒔 𝓣
𝓗 , working
and𝓒blockchain 𝓖𝓓through 𝕳 the data 𝝆 stored 𝓢𝓬 the 𝓓 𝓲 on 𝓗 , that will yield the
by result
𝚰will be fetched ⋮from via by ensuring its security
y 𝓛(𝒙) is required, 𝓛(𝒙) 𝓖𝓸added 𝓬𝓬add 𝓬ledger
latency According
which isisonly to Equation
required, corresponding
possible (7),
According
𝖃 the
ℂ =
isisblockchain
hash
only to ⋱
⋮ data
implemented Equation
value by digital
possible 𝚰executing
𝝆willif ⋯
(7),
⎯⎯be
and ledger 𝓗
𝓬the
𝓛is 𝓲𝓬 𝓶
and
implementedto 𝓵 willthe 𝓕will
𝓬 new block
digital
𝓖 the
generate and hash of athe
working of 𝓵 will
that
blockchain will𝓬 add beBthe hash
populated
𝒸 where of
the the all the
𝕳. 𝓛 𝓜 𝓵 =identifying
𝕺 𝖃 ⇔and⋮ reporting ⋱will (8) ⋮be 𝓕 , as shown (6)(8).
using
data by executing
ly, as shown in properly,
Equation as shown 𝓒in𝓶Equation
This
(5). input data machine
The will
data
and will
stored of
be
bylearninglegitimate
decrypted
executing
generate
(5). claim 𝚰𝝆involved a⋯𝓒𝓢𝓶𝓬𝕺
algorithm
data HIN
and 𝓗 claims
using
that
and
𝖃to will
𝓬 will
willbe𝓕
secure suitablegenerate
be
𝓲applied by
𝚰𝝆populated
𝓬fetched
and to ML a𝕺
from 𝓢
techniques
.𝓬 by
extract
𝖃 𝓗all
that 𝓬𝓓
will
legitimate
via it
the𝓖be entities
bypopulated HIN𝓬 claim
𝓕ensuring
by
itsLall in
𝒾 . the entities
security Equation
𝓬 ⋯ input
classified
involved into to secureThe using
legitimate insurance
𝚰𝝆 and 𝕳. 𝓛𝓲This
involved 𝕺and claim
𝖃 . input towill
fraudulent secure be validated
data 𝚰claim
𝝆 and
will be 𝓕decrypted
𝕺 by
𝓬, 𝖃 as . 𝓗 shown 𝓹 with and in the
Equation
using helpsuitable of(6). patient ML 𝓓
𝓬
techniques 𝕳
data 𝚰𝝆 ,it will be
The insurance 𝓗 𝓬 ⇔treatment
claim
𝕳 𝓖 will
𝚰𝝆 be⋯ validated 𝓛and
𝓣𝓸 ,stored 𝓲 𝓬by⇔ 𝓗 𝚰with
𝓗through 𝝆 ⋯
𝕺the 𝓛help 𝓲 ofH 𝒸𝓗
patient , x,⎧𝓬as input𝚰𝝆 𝓣will ∩⋯
𝓸 data 𝚰𝓵𝝆𝓗 𝓹 result
, the
𝓗𝓬 𝓗 𝓖
the obtained the data stored on , that yield

classified into The legitimate claim
𝓛
𝕳 𝓖and 𝓹data fraudulent 𝖃= will 𝓒
claimbe fetched
𝓕 is true
from
shown in 𝓬𝓒 via
Equation 𝓢𝓬by 𝒚𝒊𝒆𝒍𝒅𝒔 ensuring its security
(6).
𝓜of 𝓵 = 𝕺𝖃 ⇔ ⋮ 𝒚𝒊𝒆𝒍𝒅𝒔
𝓜 ⋱ = 𝕺
⋮ 𝓣 𝓓 ⋯ 𝓵𝓲 ⇔ = 𝓗 𝓓 ⋮
𝓵 + 𝕳,⋱ S 𝒙 ( x⋮ ) 𝟎 = 𝓣 (6) 𝓶 ⋯ 𝓬 𝓗
⎪𝚰 = ⋮ ⋱ 𝓶 (6) ⋮ ⎯⎯ 𝓛𝓲 (3)
the obtained treatment
legitimate 𝓣
HIN𝓑𝓸𝓬 using
, = ⎯⎯ 𝕳.
and
claims 𝓵
through 𝓛
𝒇(𝒙) 𝓲 = input data will
This 𝖃
by 𝓸 the data
identifying 𝓹
stored and on𝒽
𝒚𝒊𝒆𝒍𝒅𝒔be decrypted reporting 𝓗 𝓬 , 𝓸that G 𝓕 will ,
, 𝓬x and as ℂ 𝓹
yield
is true shownusing the
𝒚𝒊𝒆𝒍𝒅𝒔 suitable
∆ in result
Equation
(7) ML (8).
techniques it will be
𝓓by ⋯ 𝚰ℂ =𝓕𝓬 ⋮ and 𝓓 =
⋱ 𝒚𝒊𝒆𝒍𝒅𝒔 𝟎 ∩
⋮𝓓𝓬 ⎯⎯ +𝕳 ,
⋯ 𝚰ℂ𝐦𝐚𝐱 𝒙 ≥
𝓛𝓕=𝓬,𝓓 𝟎 ℘ 𝓢 (𝒙)
=𝓬 𝓓⋱= 𝚰 ⋯
𝓒𝓲𝓶 𝓗
𝓵⋮ shown 𝓵+ 𝕳,⋮ Equation
𝓢 𝓱𝒙→ 𝓕𝓗
= ⎯⎯ 𝝆𝟎 𝓬 {𝓗 𝓛 𝓹 ,(8)𝒔𝓬 , 𝓗𝒓 } 𝓕𝓬 (8) (9)
𝓵
of legitimate HIN claims 𝓛𝓲classified 𝓬 identifying
𝓑⋯ ⎯⎯ 𝓛
reporting =𝓕𝒙𝓲𝓬fraudulent 𝒏as in (8).
𝓬 = 𝓗 𝒇(𝒙)
into legitimate 𝓲 and 𝟎 𝓓=𝓵 𝟎⋯ 𝕳+𝕳 claim
⎨ 𝓬 , as shown in Equation (6). (7)
According 𝚰
to 𝝆 Equation ⎧ 𝓓ledger
𝓬(2), 𝚰
in 𝓕𝓬∩
order 𝓓 𝚰
𝓓𝓖𝓒𝓵 by
to 𝝆 make ∩ 𝓗
℘ a ,
𝓬 𝒙 ≥
successful, 𝟎 ∆ 𝓕𝓢 𝓬 𝓓
smart 𝕳healthcare system,
he stored claim dataAccording 𝕺𝖃 stored toclaim
Equation 𝕺𝖃 the
(7), 𝓗 be 𝓖fetched
𝓬 blockchain digital 𝓗 𝓵 will 𝓢add ⎪𝐦𝐚𝐱 the hash of𝓬 the 𝓵
𝒙𝓬 =𝓗
The will be fetched datafrom will via byThe ensuring from 𝝆
its 𝓬𝓗via security willensuring its→security
S ⎧ 𝚰
insurance
∩⎪ 𝓵𝒚𝒊𝒆𝒍𝒅𝒔 𝓬𝕳 claim 𝓶 be
𝓓 ⎩ =𝚰𝝆𝓓𝓵𝓖+
𝓬 𝓵 validated 𝕳, 𝓗 by 𝓹 with
𝓒𝟎𝓶 16 𝓒𝓶 the help of patient input
𝓢of𝓬to27H𝒸 using G .
𝕳. data by𝕳. executing 𝓒 and need will to be
generate functional, a 𝓢 that and
𝓗𝓬𝓹will 𝓑 this
will = ⎯⎯
beis only 𝒇(𝒙)
populated possible
= by𝓓 𝓗 by when all the it is connected
entities
𝓓 ≥𝚰𝝆𝟎 𝚰𝝆 ,(7)
are This
2023, 11, x FOR
input PEER
data
using REVIEW
will be Thisdecrypted
input data
and willusing be suitable
decrypted ML and 𝝆techniques using suitableit
The insurance claim 𝓶 𝒽 willThe validated
According insurance
to Equation the byclaim𝓬
obtained (7),
𝓗 with
the
= 𝓢𝓬treatment 𝓒blockchain
∆
bethe 𝓢 help𝓣of
validated will ML
be digital
, and techniques
patient
𝓵 =
ledger
through
𝓹𝟎 with
input ∩ +𝕳
it
the
the will
data 𝒙help
𝓵, will data be ,of
add stored patient
𝓢the on input
hash 𝓗of 𝓬 , the datawill
that yield t
𝚰 𝕺 𝐦𝐚𝐱 ⎪℘ (𝒙) 𝓱(6).
𝓶 →shown 𝓗𝓬𝓬{𝓗𝓹in,𝓸𝓗 𝒔 ,𝓗 𝒓 } (6). 𝓬 (9)
ed into legitimate 𝓛
involved
classified
the obtained
𝓲 and to secure
fraudulent
into legitimate
treatment data𝝆 and
claim
by𝓣the𝓛 𝓕 and . ,
obtained
𝓲, and as
𝖃𝓬 through
executing fraudulent
shown𝟎 𝓒 𝒙 𝒏
treatment
of in
and
the Equation
claim
∆
data
legitimate will 𝓣 𝓕
⎨
stored
, ,
generate
𝓬and
HIN as through
onclaims a 𝓗 𝓢 , 𝓛
thatEquation
that
the by datawill
will be
yield
identifyingstored populated
the on 𝓗
result
and , by
that
reporting all
will the 𝓕 entities
yield, as theshown result in Equat
For efficient 𝐦𝐚𝐱 𝓸 ℘ , the ℘ (𝒙) = 𝓢𝓱 → 𝓗of
𝓶 𝓸{𝓗𝚰𝓹Cat , 𝓗maximum 𝒔·the ∆·, ·𝓗𝒓 }E 𝒸𝓓 𝓬 𝓬 𝓲 speed 𝑺(𝒙) · ledger
· ·andB 𝓬 𝓬
   
According transmission to Equation 𝓬⎪ (7), blockchain 𝓵compromised𝕳O digital X (9) 𝒸minimum
𝓓 will add the hash of the
of legitimate HIN involved
claims of 𝓛
𝓓 = 𝓓𝓲 𝓵𝒚𝒊𝒆𝒍𝒅𝒔
𝟎legitimate
to𝒙 by 𝒏Corollary.
secure identifying
+ 𝕳, 𝒙which 𝚰
=𝓣𝝆𝟎𝓸 𝓒is
HIN ⎨
andThe ⋯
𝓓 and
claims 𝕺
Security
𝓗 . 𝓛
reporting 𝐦𝐚𝐱 byof the𝓖 →
identifying𝓕 ℘ 𝓗, can
as
𝟎𝕳 𝓬𝓖𝓒..𝓶 be
shown and in
reporting Equation if 𝓕𝓕 (8).
, is
as generated
𝓵
shown in and
Equationnot detected
(8). by ℘.
H𝓒=
𝒚𝒊𝒆𝒍𝒅𝒔 𝖃𝓓 𝓲+ .𝕳, 𝒙 .= 𝓬if 𝓬𝓬 . working 𝓓𝓵 𝕳
𝒇(𝒙)𝓛(𝒙)
latency 𝓶 𝓵only 𝓹⎩ 𝓓𝓵 .generate 𝓒 𝓢 .. . . and
= is𝓵 𝓑required, possible is𝓶
∆ 𝓵 𝚰𝝆.. will implemented
O
..populated
𝓑𝓬 = ⎯⎯ = ⎯⎯ by𝒇(𝒙) =⎪𝐦𝐚𝐱 = 𝒚𝒊𝒆𝒍𝒅𝒔 a 𝓢 𝓬 that 𝚰 ∩ (4)detected,

𝓬 data Furthermore, executing
⋮ (5). 𝓗 𝒸𝓖 ,𝓶 𝓗 → and𝓗𝒐𝒓 𝓗 may . be(7) . involved 𝓬 in . HIN will⎧ . be
fraud. (7) The 𝓕 isby all
immediately the entities
properly, as𝓓 𝓵 =𝟎∩ in+𝕳 𝚰ℂ, 𝒙 =≥ 𝟎 ⎩ 𝓢𝓓⋱𝚰𝓬𝝆𝓵 𝚰= ⋮∩
𝓹 𝟎 𝒔 +𝕳 𝓬𝓓,𝓣
⎯⎯ 𝒙𝒓𝓛≥ 𝟎⋯ 𝓢𝓗 𝝆
𝓒𝓶𝓬 𝓢𝓬
𝓸·𝓲·𝕳 𝓓𝓵⎪ 𝕳 (8)
 
𝓒
· 𝓢𝓬O⎧
shown Equation 𝓵𝓶 𝓬 𝓹 𝓗𝓬
involved and any ⎧ ⋯ 𝚰𝝆𝝆𝓗
to𝚰manipulation
secure and ∩inT𝕺
𝓗𝓬 𝚰 = 𝓶 ⋮ 𝓒𝜤 ℴℂ𝖃 .is 𝓕 prohibited
𝓢 X using 𝚰𝒚𝒊𝒆𝒍𝒅𝒔
𝝆 D ∩𝓁 SHINFDP.
the 𝓒
· · · ∆H
𝓢
𝝆 𝓬 𝓬
𝚰𝝆 if𝓬of ⋱𝓓 ⋮ add
⎪is𝐦𝐚𝐱 ⎯⎯
℘𝓗𝓬 𝓛 𝓶
(𝒙) = 𝓬→ 𝓗𝒔 , 𝓗𝒓 }(8)
ccording to Equation According
Healthcare(7),2023,
the blockchain
11, to
Corollary. x FOR Equation
ThePEER digital
Security
REVIEW(7), the ledger
of the ℘𝓓⎪
blockchain 𝓵 will
can be∆digital
add 𝓗𝓬ℂ⇔
compromised theledger
𝕳 𝓖 𝚰
hash ⋯
𝓕the 𝓵𝓬 will
𝓣
𝓛
𝟎 generated
𝓲𝒙 ⋯ 𝒏∆ 𝓗
the𝓲and hash
⎨
𝓢of
not the
𝓱 detected 𝓗𝓬 {𝓗by 𝓹 , ℘. 16 of 27
y executing 𝓒𝓶Corollary. 𝓒 𝓢 𝐦𝐚𝐱 ℘ (𝒙) = 𝓢 𝓢 → 𝓗 𝓗 𝐦𝐚𝐱
{𝓗 , ℘
𝓗 𝝆 (𝒙) , ⋯
𝓗 = } 𝓗𝓸
𝓢 𝓬 → 𝓗 ℘ {𝓗𝓹
𝓕
, 𝓗 , 𝓗 𝚰 } H
andThe
data by
will insurance
executing
generate
The Security
Furthermore, claim
a𝓶𝓗 and
of𝓬 𝓹thewill
that
,𝓗 According
𝟎℘
will be
will
𝒐𝒓𝒏can
validated
generate
𝓗bebe 𝓜 to
populated
𝒓 compromised
may =
Equation
a by 𝓱
𝓵 be involved𝕺 that by (3),
with
𝓹 if𝟎 𝓕
𝓬 all
will
a𝒙in⇔
for
𝓬𝓗𝓹 the
𝒏isHIN
smart
be ⋮ ,𝚰𝓗=
𝓹generated
𝒔 help
entities
populated ⋱and of
𝒐𝒓⎨⋮𝓗and
𝒔fraud.𝒓 secure
⋮
patient
𝓱
The by
not all
𝓕𝓬detected
𝓬 ,
input healthcare
𝓬 the 𝒚𝒊𝒆𝒍𝒅𝒔
𝓹 data
⎪𝐦𝐚𝐱entities
by𝓛 ℘.
𝒔 (9) cloud
𝒓 , ∆ 𝓓 needs 𝕳 to
𝚰ℂ .𝓵 The value of(8)𝕳 is
(6)𝒸 work (9)
𝒙𝒔 Proof. Assume 𝓬 𝖃that
⎨ be accomplished ℘
𝒓 ⋱intends ⋮is immediately to⎯⎯ manipulate 𝓖 →detected,
𝝆
𝓗
ed to secure 𝚰𝝆 Furthermore, 𝕺𝖃 . and 𝚰 and 𝓣𝓸𝒓, 𝕺 𝓓 𝓵𝓬 𝓗
ℂ if and
⋯ 𝓕𝓹only ∆ 𝚰 Catand T 𝓲 are 𝓬
𝓒𝓶block𝑺(𝒙) 𝓢
involved
andobtained
the to secure
𝓗any 𝓹, 𝓗
treatment properly,
𝒔 𝝆𝒐𝒓The
manipulation 𝓗 and
may 𝖃 in
insurance 𝜤ℂ For
which
.immutable,
through
be involved
is claim efficient
will the
⎪so
prohibited 𝐦𝐚𝐱 indata
willit HIN be
using
does stored
𝓖 →security
, the
∆ fraud.
validated
not the
𝓗𝓬 which
𝓓
on transmission
The
SHINFDP.
allow by 𝕳
𝓬𝓕
this , 𝓬that
𝓗 ⎪ is
𝚰𝝆𝐦𝐚𝐱 𝓬
change will ofifyield
immediately
with
⋯𝓖 → the
𝓗𝓗
and help themaximum
⎩ 𝓓 result
detected,
declares
𝓬 𝓬by adding
of
𝓵 𝚰𝝆ℴpatient
𝓕𝓬
𝕳 the
speed
encrypted
alteredinput data and𝓬 as
and ,𝓕𝓬minimum
𝚰𝝆claim . This will
of legitimate HIN claims data
𝜤 𝓛 are by stored
latency
identifying O 𝓛(𝒙) ⎩.
𝓣
Further
is
and𝚰𝝆, and required,
reporting 𝓒 𝓕 will , 𝓢
as be is⎩ implemented
shown only
𝚰𝝆 in 𝚰Equationpossible 𝓗 𝓒 if (8).𝓖 𝓢 is the hash
implemented H of the stored
and working
and any manipulation The𝓣𝓸stored the in
⋯ ℂclaim
obtained is
𝓗𝓹 data 𝕺𝓣
𝓲 prohibitedtreatment
prevent using𝓗X
𝖃 𝓸will 𝓹⋯ , 𝓗
the
𝓸 be 𝒔𝓗 𝒐𝒓
SHINFDP. 𝓗through
fetched
𝓹 ledger 𝒓 from 𝓶
from 𝓗𝓬 via 𝓖 byℂ ensuring
𝓬
the 𝓬
data
manipulating stored on . The ,𝓶𝓢that between
𝓬 𝓬 its security 𝒔
𝓬
will yield 𝓗 the and result 𝓗 𝓹 also keeps
data on the
properly,
𝒚𝒊𝒆𝒍𝒅𝒔
The blockchain
𝓗 , 𝓗
insurance as 𝒐𝒓showndigital
𝓗 claim in Equation
𝒚𝒊𝒆𝒍𝒅𝒔
will be D and
validated(5). creating by 𝓗 a 𝚰 new with block the in help the blockchain
of 𝕳patient B
input . data 𝚰 , we
Proof.
using Assume
𝚰ℂ =𝕳. ⋮This⋱inputof legitimate that
⋮ data a
HIN
⎯⎯𝚰ℂwill
both = claims
𝓲 ⋮Corollary.
𝓛parties
𝓹 be ⋱𝓛
on𝓲 the
𝒔decrypted by
𝒓 ⋮ The intends
identifying
same and 𝓓Security
⎯⎯ 𝓵 pace using to 𝓁
𝓛𝓲𝕳 ofsuitable manipulate
and (8) the does ℘ can
reporting notℂbeallow
ML 𝓹 𝓕
. The
techniques, as
compromised
𝓬 value
shown
anyone (8)it will of
ifto𝓕violate
in Equation
𝓬be
is
is generated (8).𝒸
the 𝓢𝓬 .and Below,not𝝆 detec
𝓗 the , 𝓗 𝒐𝒓
obtained 𝓗 ⎧
treatment 𝚰 ∩𝓣 , and through 𝚰 the data stored 𝕳
on 𝓕HIN 𝓗 , that will yield the result
Proof.
Corollary. Assume
immutable,
classified that
𝚰𝝆into⋯
The Security a 𝓗
solegitimate
ofitCorollary.
does
𝓹𝓬 ℘
the 𝒔not
discuss
can 𝓛𝓲allow
The
𝓕be𝓬and the intends
𝒓 𝚰compromised
this
Security ⋯ change
possible
⎪fraudulent
Furthermore,
𝝆
𝓗
𝓗 to
of𝓬 𝓬the 𝓸𝓗
ifways 𝓒
manipulate
and
𝓕 ℘𝓶
claim 𝓬 ,can 𝓗
is 𝓕of 𝓢
declares
𝓕 𝓬𝒔be 𝓬𝒐𝒓,disrupting
generated as 𝓗 ℂ the
shown
compromised . The
and
may altered
in
𝓓
value
the
is𝓗maximum
𝓬 ⇔Equation
not
be 𝓕 𝚰𝝆is generated
block
security
ifinvolved
𝕳 𝓖 𝕳𝓬
detected of as
by⋯and
(6).
in is
℘.𝓬 .𝓛 𝓬This
𝓲integrity
and
fraud. willnotThe of 𝓕
detected℘𝓬 and is by ℘. the
how
immediately

𝝆 𝓹 S(x), 𝒓 x 𝓵
immutable, so𝓗
prevent it 𝓹does
, 𝓗𝒔𝓹not
𝓗 , 𝓗Furthermore,
𝒐𝒓 𝓗 𝒐𝒓ofmay
𝒔allow 𝓗
legitimate
this from change
𝓗 HIN and
manipulating
, 𝓗 claims
∆ G
𝒐𝒓 declares
𝓗 = 𝓛
f (𝚰 𝓲x )
. by =
the
The ⎧
identifying
𝓜 𝓢
altered 𝚰
𝓕 = between𝕺 ∩
℘
block and as 𝓗 𝓕
reporting
⇔ .
and ⋮
This 𝓗 ⋱ 𝓕
will 𝓬 𝓕, as
also ⋮ shown
keeps in Equation (5) (8). (6)
Furthermore, proposed
𝒓𝐦𝐚𝐱 ℘ (𝒙)𝒚𝒊𝒆𝒍𝒅𝒔
𝒓 be involved
=and SHINFDP
𝓹 𝓢any in HIN
𝒔 → manipulation
𝓗𝓬with helps
𝒓{𝓗
𝓓
fraud.
may ℂ
= ,the
𝓗 𝓓
to
be Thesave
involved
in
,+ 𝓗L 𝓬
𝓵
𝕳,𝜤}(x),
𝝆
ℂ𝒙
the
𝓬𝓗 is
is=𝓬x
in
𝖃 < x𝓒
prohibited
𝟎
from
immediately
HIN
𝓒 is
𝓶
𝒔
minimum security
𝓢using
𝓬
fraud.𝓬 data
detected,
The 𝓹
the(9)breaches.
SHINFDP.𝓬 is immediately detected,
Healthcare 2023,
he insurance claim 11, xwill
FOR
The bePEER
insuranceREVIEW
validated claim
by 𝓗 will with be the
validated help of 𝓱
by 𝓗
patient input
𝓵 𝓹 𝒔
𝓵 ⎪
data
help 𝒓 𝚰 of , patient 𝓶 input 𝓓 𝓢⋯𝓬𝕳 𝚰 ,𝓕𝓬 (7)
16 of 27
and any 𝓗
prevent 𝓹 , 𝓗parties
both
manipulation 𝒔 𝒐𝒓 𝓗in 𝒓on from
𝜤and the𝟎
is any 𝒏𝓑 =
𝓹 𝒙manipulating
same
prohibited 𝓬 pace
manipulation ⎯⎯using ⎨and𝒇(𝒙)𝚰ℂthe .in
does =
The 𝜤ℂ𝓹∆𝓓
SHINFDP. 𝓢𝓬prohibited
not
is =𝟎
between
allow anyone
∩ using𝓗
∆𝝆 𝒔 and
≥⎧𝓬the
to SHINFDP.𝓗𝓹 also
violate 𝓓𝓬 𝓵keeps
the . 𝝆Below, we
ained treatment the 𝓣𝓸obtained
,parties
anddiscuss treatment
through the data𝓣ℂ𝓸pace , stored
and throughon 𝓗Breaches:
⎪the
, 𝐦𝐚𝐱
that 𝐦𝐚𝐱
data will stored 𝓵(𝒙)
℘anyone
yield 𝓓
= onthe𝓵 +𝕳 𝓗 𝕳
𝓢result , , that
→ 𝒙𝓗 𝟎{𝓗
will 𝚰𝓢 𝓬,𝓢 𝓗.𝒔∩Below,
yield , of
𝓗the }result (9)
both on the the samepossible
For efficient 1.and
ways
For efficient 𝓗 ℘𝓹 ,Proof.
doesof 𝓬
℘thenot
disrupting
, the allow 𝓖
transmission
𝟎 transmission
𝒙
Assume 𝒏→ 𝓗
the
Whenever 𝓬 that security of
𝕺 to
⎨of𝖃𝓢 𝓗 𝓱 𝓬
a𝚰will
violate
ρ𝓹at𝓗and tries
at𝓹maximum
maximum the
, 𝓗fetched
integrity
to 𝝆deliberately
𝓹
𝒔 𝒐𝒓𝓗𝓗 𝓬𝓬 𝒓speed
𝒓
𝓒𝓶 S ℘
intends 𝓗𝓢we
and𝑺(𝒙)
delay
(x𝓬𝓬)via to how
and or
and the
manipulate
𝓖minimum
manipulate minimum 𝚰latency
theThe
ℂ . its
𝚰ℂsecurity
, thevalue 𝓢𝓬
imate HIN claims 𝓛𝓲 by
of legitimate identifying
HIN claimsand 𝓛reporting by
The
identifying 𝓕 ,
stored
as⎩ 𝚰
shown
and𝝆
claim
reporting
in
data
Equation 𝓒 𝓶 𝓕 , 𝓬
(8).
as shown ⎪
be∆ in 𝓓
Equation
from
𝕳 (8).
by ensuring
discuss
Proof. Assume the possible
According
proposed that ways
SHINFDP
latency to 𝓗
aL(Proof.𝓛(𝒙)
x)𝓹is 𝓲
of ,𝓗
Equation helps 𝒐𝒓
disrupting
required,
is𝒔Assume 𝓗(7),
to 𝓬
𝕳.𝒓save
between
required, the the
which
that
intends the 𝓗a𝓹is
which ℘𝓗
security
blockchain to from
and
only ,𝓗
ismanipulate 𝓗
anddigital𝒐𝒓
security
possible
itonly ⎪helps
𝓬
𝓗
integrity
𝐦𝐚𝐱
possible ledger 𝚰allow
ifbreaches.
𝓖ℂto
G.→is
intends of 𝓓
solve
𝓗
The ℘ ∆
𝓵𝓖towill
implemented and
value this𝓵 add
manipulate how the
problem
of 𝓗 𝕳and
thehash is𝚰by of
working
. and the
providing
The properly,
value aofcopy 𝕳asit is as 𝓕
of the
using immutable,
This input so
data
𝓹 𝐦𝐚𝐱 does
𝒔will 𝒔 ℘
⎩
be(𝒙)
not =
𝒓 decrypted
𝚰𝝆 be populated 𝓢if this
𝓱 𝓬→ 𝓒 𝓗ischange
and
𝓶𝓬 by
implemented
{𝓗 𝓢𝓹𝓬 ,all
using and
𝒔 , 𝓗𝒓 }
suitabledeclaresℂ ML working
the altered
techniques block will
(9) be𝓬.
proposed SHINFDP
data xby it executinghelpsshown toas 𝓒save𝓶in 𝓓
and the
Equation 𝕳
will℘
transparent from
generate
(5). security
𝓗
agreementa , 𝓢
𝓗 𝟎𝓓 𝒙that
breaches.
𝒐𝒓
𝒔𝓛𝓲 and
𝒏 between 𝕳
𝓗 will ⎨
both parties. 𝓕 𝚰 the entities
𝓢 𝓕 𝓗 𝓗
immutable, 11, so properly,
does not immutable,
allow shown this in
change
so Equation
it does
prevent and not (5).
declares
allow thisthe change
altered from and
block declares
manipulating as ∆ 𝓬 𝓕𝓬 ,𝓓. the
This altered
will
. The block between as . This will
and 𝓹 al
𝓵
classified into 𝓬 𝓵
⎧ legitimate 𝒓 fraudulent claim as
ℂ shown 𝓬 in Equation (6).
Healthcare 2023, FOR PEER REVIEW 16 of 27
𝓵 ℘.𝚰𝕳, the 𝓢
𝚰of𝝆 the ∩Whenever 𝚰𝝆 to∩ 𝓹 𝓬 𝒔
Corollary. The 𝓗 Security ⎧ 𝚰 ℘ 𝓒can 𝕺be 𝓢 𝓗
compromised if 𝓕 𝓒 is 𝓢
generated ⎪and 𝐦𝐚𝐱 not 𝓖 →
detected 𝓗 by
1.
involved
prevent 𝓗𝓹 , 𝓗𝒔 𝒐𝒓⎪𝓗𝒓 from 𝓹 toBreaches:
secure 𝓗
prevent 𝓬 𝝆 and
2.
manipulating
𝓶 𝓗
𝓗𝓹 , 𝒔𝓗both𝖃 .
𝓬 𝒔⎪ 𝒐𝒓 𝚰𝓗
Breaches:𝓹 tries
ℂ .𝒓 The
parties 𝓗 from𝓢
𝓬 deliberately
Whenever on 𝓬
𝓬manipulating
𝓶between
the𝓗same 𝓬 𝓗 𝒔 𝓗
delay
tries
pace 𝚰and or
to
.𝚰and
The manipulate
hide𝓗does 𝓢𝓬also its
between
𝓬not identity
keepsthe
allow 𝓗𝒔anyone
ℂ or and 𝓗to
misuse 𝓬 his
also keeps
𝓹 violate 𝓥 𝒊 the 𝓢𝓬 . Be
to obtain
𝓗 ,the
𝓗 𝒐𝒓 ∆𝓗𝒓𝓹pace ℘allow 𝒚𝒊𝒆𝒍𝒅𝒔⇔
𝕳 𝓕
𝚰𝒔⎩ ℂ
𝝆⋯𝓓 𝓛=
𝝆 immediately
𝚰hinders
𝓹
𝓲· ·𝚰𝓓,𝓵 +𝓒way
𝒾a𝕳,
𝓶 𝒙 𝓢= 𝓬 𝟎 𝓒𝑺(𝒙)
both𝓗 𝓗 𝓗 𝓢
Furthermore, may For efficient ∆HIN ,solvethe transmission 𝓬or 𝓖 𝓬 isof atis ·anyone
𝓵 maximum L speed 𝓶 and minimum
1.
𝐦𝐚𝐱 𝓹 Breaches:
℘ (𝒙) between
𝓹= Whenever
𝒔 same
Corollary. and
The 𝓹be involved
tries
illegal
(𝒙)
Security to
helps of in
deliberately
benefits,
the to ℘ canthefraud. 𝓑
be 𝕺
delay
this = ,The
compromised 𝓒
problem
⎯⎯ manipulate
and 𝒇(𝒙) 𝓢by
if= 𝓕 ρproviding the
𝓢 generated ℂhisdetected,
the copy
and 𝓬 and not ofthus the
detected 𝓢
preventsby ℘. any ℘ and
kind of
parties on 𝓢𝓱 →both 𝓗𝐦𝐚𝐱 {𝓗parties ℘
and , 𝓗 𝒔does
on,=𝓗 thenot
discuss}𝓢
𝒓 𝓱 𝓜
same→ 𝓣 𝓗
the
𝓸 𝓵𝓬 = 𝕺
pace
{𝓗
⋯ anyone
possible 𝓗,
𝓬and𝓗 𝖃 ,to
does
𝓗
ways
H𝓶 ⇔𝒸violate
⇔}
(9)
𝒓 H G⋮  ⋱
notof 𝓬 allow
the
disrupting
𝓬
𝓓a𝓵 = 𝓬 . Below,
⋮ 𝟎if ∩𝓖+𝕳 the to we
security
(9) violate
, implemented
𝒙 ≥ 𝟎 𝓢𝓬
the
and .
integrity
𝓬 Below,
(6) working ofwe (7)
𝟎 𝒙 manipulation
𝒏 𝓗 ⎨ 𝜤 𝓗 𝟎 𝓬prohibited
latency 𝒙 𝒏 𝓹 𝓛(𝒙) 𝒔
is ⎨ required, which𝓹 𝖃 𝓹𝒔
is only possible. . . is and
and any
between andin is helps 𝓗the to using
,fraudulent
𝓗 𝒔 𝒐𝒓
solve 𝓗 thethis 𝚰⋮M
SHINFDP.
problem by 𝒚𝒊𝒆𝒍𝒅𝒔 providing copy 𝓕.𝓬 integrity
of
transparent agreement between both parties. the
discuss the possible 𝓹 Furthermore,
ways ℂ
⎪𝐦𝐚𝐱 𝓖properly,
𝒔 ∆of disrupting
discuss 𝓹𝓓 possible
𝓵 The
proposed
𝕳𝚰⎪ the
=𝒓 ways
ℂSecurity
may
security execution.
∆ 𝓁be
ℂ SHINFDP ⋱of=involved O
disrupting
and ⋮X
𝓓 𝓵can⎯⎯
integrity
helps 𝕳⇒in HIN 𝓛⇔the
to
𝓓 of
save . ℘theand
.
fraud.
security
𝓲 𝓬 ⋯ 𝓕𝓬if 𝓕 is generated (8)
.
The℘. and how
from . is
the immediately
security of ℘ detected,
breaches. and how(6) the
transparent2.SHINFDP 𝓗agreement
𝒔 Breaches:
and any → 𝓗
Corollary.
between
Whenever
manipulation 3. 𝓬 the 𝓗 asboth
According ℘in 𝓗
shown 𝜤 𝐦𝐚𝐱
parties.
Breaches: tries
is in 𝓖 →
prohibited
to Ifofto
Equation
to
Equation 𝓗𝓗the
hide 𝓬 ℘using
tries (5).
its
(7), to be
identity
℘ compromised
manipulate
the
the SHINFDP.
blockchain or misuse 𝚰 , the
digital 𝓗
his
𝓬 𝓥
ledger data to 𝓓
obtain
that andis
will not
stored
add detectedusing
the hashby ℘. of the
𝓬 ℂ 𝚰 𝝆 𝚰𝝆 ⋯ 𝓗
proposed helps
⎩ 𝚰aThe proposed to save SHINFDP
𝓒𝓶 𝓢⎩ 𝒓 from
𝒔 helps security save
𝒓 breaches.
𝓒𝓶 𝓬
the from
𝓢𝓬 𝓕𝓬 D𝒸 · · · F 𝒸 security ℂ breaches. 𝓬 𝒊 𝓵
Proof.
2. 𝓗𝒔Assume Breaches: that
illegal
𝝆
Whenever 𝓗𝓹
benefits, , 𝓗𝓗
stored
Furthermore,
the𝒔 𝒔𝒐𝒓 𝕺 𝓗
claim
tries , 𝓒 𝓗data
tointends
𝒓 executing ,𝓗
hide
and 𝓗 𝕺𝒔𝓢𝖃𝒐𝒓 to
its 𝓒 𝓗
will identity
hinders be
manipulate may fetched
his be
or 𝚰ℂfrom
involved
way misuse .and The 𝓗 𝓗
in
his
thus value
𝓬HIN
𝓢 𝓥
via 𝓖 of
fraud.
to
prevents by 𝕳
obtain ensuring
The anyis 𝓕 kind is its security
immediately
of detected, the 𝚰ℂ
data 𝓔𝓬 and is connected
by 𝖃 1.𝓶 𝓹 𝓹 𝓬 Breaches:
𝓶 𝒓 and to the 𝓑𝓬 using
willWhenever generate 𝓗𝓬 ⇔ 𝕳 a𝓹 becomes 𝓬𝚰𝝆
tries 𝒊 that to⋯ disconnected
will 𝓛𝓲
deliberately be 𝓬
populated delay or by
from existing all
manipulate the entities
immutable,𝓗𝓹 Breaches:
1. illegal soThe it does
benefits, using not
insurance
the 𝕳. 𝕺 𝚰𝖃This
1.allowand, 𝓒
claim𝓗 input
this
any
𝓗 and will 𝓢 data
change be
manipulation will
𝓗 and
validated
hinders , 𝓗 be
declares
in his 𝜤decrypted
𝒐𝒓by wayis 𝓗 𝓗𝓗 the
prohibited
and with and
alteredthus the using
using block
help
prevents 𝕳 𝓖
the suitable
of as 𝓕
patient
SHINFDP. any . 𝚰 ML
This,
kind 𝚰input techniques
will of 𝓢 data 𝚰 , it will 𝕳 be
𝚰 , 𝓢
fraudulentProof.
Whenever Assume
ℂ execution.
involved
𝓶 that
Breaches:
tries
𝓹 𝓹 𝓑 , which 𝓬 a
to
to securedeliberately
Whenever
between
𝓹 𝒔𝚰𝝆 and𝓜
ℂ delay
𝒓𝓹𝓹 𝓹𝕺𝓵highlights
𝖃=
intends
and . 𝕺𝓗
tries or manipulate
to to
deliberately
𝖃 𝒔 helps ⇔fraudulent
manipulate to⋮ solve
the 𝓬 ℂ⋱ ℂactivities.
delay the .⋮ 𝓬 problem by providing
or
this The
manipulate value 𝝆 ofthe ℂ is the 𝓬(6)a cop
ary. The Securityprevent
Corollary.
of the ℘the
fraudulent𝓗 can
3. The
𝓹 ,𝓗be𝓗
obtained
𝓗𝒔𝚰𝒓𝒐𝒓
Security 𝓗
classified
compromised 𝒓 of
execution.
Breaches:
ℂ𝓹immutable, from
treatment 𝓗the into if℘
If so𝓕𝓗 𝓣
legitimate
can
manipulating
𝓬it , be
isdoes
𝒓𝓸triesand𝓬 compromised
generated
𝓗 to 𝓛𝚰and
through
not manipulate .immediately
and
and The
𝓲transparent
ℂthis not
the
𝓗 if𝓢𝓕
fraudulent
𝓬detected
data 𝚰change,is
𝓬ℂbetween the generated
stored to𝓗 claim
by ℘.
𝓗ondata𝒔𝓕 𝓗
andandthe
,that
𝓬this asnot 𝓗
𝓬 , that
shown𝓓
ais detected
𝓹both also
will
stored ⋯ in Equation
by
keeps
yield 𝓕the
using ℘. the as(6).
result 𝓕𝓬 . This
between and 𝒔 between
helps to solve 𝓹 allow this
problem
𝒔 helps agreement by and solve
providing
𝓬 declares
between
𝓣immediately
the
problem
⋯𝓢to. 𝓗
copy
𝓬 altered of by
parties. block
𝓬 providing a copy willof the
more, 𝓗𝓹 , 𝓗𝒔 𝒐𝒓 𝓗𝒓𝓗
Furthermore,may beon 𝓗 ,𝓗
involved 𝒐𝒓
𝓗 in𝓗 HIN 𝓗may
𝒓Proof. fraud. be The
𝓛Assume
involved
The 𝓕not
above 𝚰is
𝓬 that in discussion
immediately
HIN a 𝕳 𝓗 fraud. ,data
𝓗 shows The 𝒐𝒓
detected, 𝚰𝓗 𝓕.that 𝓕
is properly
intends,𝓸𝓗 implementing
𝓹in detected, 𝚰identity
SHINFDP makes of 𝕳 theisHIN
both
3. parties
𝒓 of
transparent 𝓔𝓹𝓬the
legitimate
Breaches: If𝒔same
HIN
prevent
agreement
and pace
is𝒓connected claims
tries to, and
between 𝓗
transparent
𝓹 𝒔 to𝓲does
𝒐𝒓
manipulate by 𝓗2.
both
the identifying
𝓑from𝓬𝓗
𝒓 agreement
parties.ℂ allow
,𝒚𝒊𝒆𝒍𝒅𝒔
using the
𝒔 manipulating
Breaches: anyone
and 𝓬𝓹
between
becomes
𝕺
reporting
𝒔Wheneverto
that
𝓓 both
𝓵
violate
=𝒓𝓬isThe𝓓
disconnected
ℂ
stored
𝓬+
parties.
𝓵 ⋮
the
𝓢as𝕳, shownusing
⋱𝒙tries
𝓬𝒔 between 𝓬= from𝟎⋮manipulate
Below, to𝓗 𝓗𝒔we
Equation
𝓒existing
𝒚𝒊𝒆𝒍𝒅𝒔
hide
𝓶 andits
𝓖𝓲 by
𝓗 . The
ℂ(8).𝓹 alsoor
value
keepsmisuse his 𝓥𝒊 t
y manipulation indiscuss
and𝜤 any
is manipulation
prohibited
the possible using inthe
ways 𝜤 SHINFDP.
is of The
process
prohibited
immutable, disrupting stored fair
using
so 𝓑it
theclaim
and
𝓬 =
the
does ⎯⎯ data
transparent.
SHINFDP.
securitynot 𝒇(𝒙)
allow and =𝖃 this will integrity 𝚰
Furthermore, ℂbe
change = fetched of and ℘ the from
declares
and proposed how ⎯⎯
𝓬thevia the 𝓛
SHINFDP
altered ensuring
block provides
as 𝓕 its. This
(7) security will (8)
flexibility to
2. ℂ 𝓔𝓗 𝓬 𝒔 and is
Breaches: connected both
Whenever
𝓑𝓬 , which immediately
ℂto the
parties
2. 𝓗 𝓗𝓑
on
𝒔𝕳. the
using
Breaches:
tries
𝒔𝓬 This same 𝕳
to
highlights pace
becomes
hide
Whenever
illegal its
the and identity
benefits,
fraudulent 𝓗
does
disconnected 𝓓not
tries
𝒔 decrypted the or = 𝕺
𝓵𝓓activities. 𝟎
𝓵 𝖃𝕳𝚰𝝆 𝓶
allow
misuse
to ∩
from
,hide𝓒+𝕳 anyone ,
his
its
and⋯ 𝒙
existing ≥ 𝓥 𝓢 𝟎to
identity
𝓗 to 𝓢
violate
hinders
𝒊𝓬 suitable obtain
𝓬 or the
misuse
his 𝓢
way . Below,
hisand
𝓬techniques 𝓥 thus
𝒊 we
to 𝓬 obtain
prevents any
using input data will 𝚰 be ∩ and using 𝓬 𝓕𝓬 ML according it will be
𝓗 ℘, ways
𝓗 𝒔 𝒐𝒓 𝒓⎧
𝓗disrupting 𝚰ℂ . and 𝓢𝓬 kind 𝓗 𝓗
prevent the stakeholders in selecting
from cloud services and ML algorithms to their choice.
proposed illegal SHINFDP
benefits, helps to
the 𝕺classified
discuss the , save
𝓒possible
illegal
𝓶 and
the 𝓹𝓢
benefits, from
hinders of
thesecurity
fraudulent 𝕺 his 𝖃𝓛, way 𝓒 𝚰manipulating
𝝆breaches.
ℂand
𝓬 and the
execution. 𝓒𝓢thus 𝓬 security 𝓢prevents
hinders Theway
his 𝓕any
integrity between
and of ℘
thus
of 𝒔and
prevents andhow any 𝓹the also
kindkeeps of
𝓗𝓑 , Assume
which immediately 𝖃highlights 𝓬 legitimate
the fraudulent 𝓶𝓗
activities. 𝓶 𝓬 claim
Assume that aProof. 𝓬, 𝓗
𝒔 𝒐𝒓 𝓗 The 𝒓 𝚰thatabove
intends a 𝓗
Accordingdiscussion
to , 𝓗 The 𝒐𝒓
manipulate into
𝓗
tolimited
Equation
shows 𝚰
and
intends that. (7),The ⎪
the
properly
expensiveto value𝓲 and
blockchain
manipulate fraudulent
implementing
storage
of 𝕳 𝚰
digital
of
is .the The ledger
SHINFDP
blockchain value 𝓓
, as
𝓬anyone of
𝓵 shown
will 𝕳
makes
DTL add
can
is inbe
the the Equation
HINhash
overcome of𝓢(6).the by storing HIN
𝓹fraudulent
ℂproposed
execution. both
𝓹 SHINFDP 𝒔 parties
fraudulent The 𝒓 on
helps
3. 𝚰
insurance the
ℂ
𝓗 to
execution. same
save claim
Breaches: pace
the ∆ ℘
will and
If be
from 𝓗 does ℂ
validated
security
tries not to allow
by
breaches.
manipulate 𝓗 𝓹 with 𝚰 the
, to
the violate
help𝓗 of
datathe
patient
that 𝓬 . Below,
input
is stored datawe 𝚰𝝆 ,
using
𝐦𝐚𝐱 ℘ (𝒙) = ℂ 𝒓 𝒓 ℂ 𝓬 (9)
 Healthcare 2023, 11, x FOR PEER REVIEW 16 of 27
Healthcare 2023, 11, x FOR PEER REVIEW 16 of 27
For efficient ℘ , the transmission of 𝚰𝝆 at maximum speed 𝑺(𝒙) and minimum
For 𝓛(𝒙)
latency efficient ℘ , the transmission
is required, which is only 𝚰𝝆 at maximum
of possible speed 𝑺(𝒙) and
if 𝓖 is implemented andminimum
working
latency 𝓛(𝒙)
properly, is required,
as shown (5).is only possible if 𝓖 is implemented and working
which
in Equation
Healthcare 2023, 11, 2257 properly, as shown in Equation (5). 𝚰 ⋯ 𝓛 16 of 27 𝓗𝓬 ⇔ 𝝆 𝓲
𝕳 𝓖
𝓜𝓵 = 𝕺𝖃 𝓗𝓬 ⇔ ⇔ 𝚰𝝆⋮ ⋯ ⋱ 𝓛⋮𝓲 (6)
𝕳 𝓖
𝓜𝓵 = 𝕺𝖃 ⇔ 𝓓 ⋮𝓬 ⋱ ⋯ 𝓕⋮ 𝓬 (6)
The stored
stored claim 𝓓𝓬H𝒸⋯via𝓕G𝓬 by ensuring its security using
The claimdata
dataO𝕺X will be fetched from
𝖃 will be fetched from 𝓗𝓬 via 𝓖 by ensuring its security
H. This𝕳.input data will bewill
decrypted
𝕺𝖃bewill andfetched
using
andsuitable ML techniques
using𝓗suitable it will be
itsitclassified
using The stored
This claim
input data
data be
decrypted from 𝓬 via 𝓖 ML
by ensuring
techniques security
will be
into
using legitimate
𝕳. This
classified L
into input𝒾 and fraudulent
𝓛 and
data will
legitimate claim
be decryptedF 𝒸 ,
fraudulent and as
claimshown in Equation
𝓕 , assuitable
using shown ML (6).
techniques
in Equation (6). it will be
𝓲 𝓬
classified into legitimate 𝓛𝓲 and fraudulent  claim 𝓕𝓬 , as shownin Equation (6).
𝒚𝒊𝒆𝒍𝒅𝒔 D𝓓 =D 𝓓𝓁𝓵 + + 𝕳,, 𝒙 =𝟎 0 𝓒C𝓶

yields 𝓁 𝓵= x= 𝓂
B𝒸 𝓑 =𝓬−= −− ⎯⎯
→ f (𝒇(𝒙)
x ) = = (7)
(7)
𝒚𝒊𝒆𝒍𝒅𝒔 D 𝓓𝓓 = ==𝟎 0𝓓∩ ∩
𝓵+ +𝕳
+ 𝕳,,,𝒙 x𝒙 ≥ =≥𝟎 0𝟎 𝓒𝓢 S𝓶𝓬𝒸
𝓑𝓬 = ⎯⎯ 𝒇(𝒙) = 𝓁𝓵𝓵 (7)
𝓓𝓵 = 𝟎 ∩ +𝕳 , 𝒙 ≥ 𝟎 𝓢𝓬
According
Accordingto to Equation
Equation (7), (7), the the blockchain
blockchain digital digital ledger ledger 𝓓
D𝓵𝓁 will willadd addthe thehash hashof of the
the
FOR PEER REVIEW 16 of 27
data According
by executing to 𝓒
Equation
𝓶 and (7),
will the
data by executing C𝓂 and will generate a S 𝒸 that will be populated by all the entities blockchain
generate a 𝓢 𝓬 digital
that will ledger be 𝓓
populated 𝓵 will add by theall hash
the of
entitiesthe
For efficient ℘ , the data
involved by executing
transmission
involved to
to secure
secure of 𝓒𝚰𝚰𝝆𝓶 ρat
and
and
and maximum O𝕺 will
X𝖃 .. generate speed a 𝑺(𝒙) 𝓢𝓬 that andwill minimum be populated by all the entities
involved to secure
latency 𝓛(𝒙) is required, which is only possible if 𝓖 is implemented 𝚰 𝝆 and 𝕺 𝖃 .
023, 11, x FOR PEER REVIEW
𝓣𝓸 ⋯ 𝓗 and working
𝓹 16 of 27
𝒚𝒊𝒆𝒍𝒅𝒔
properly, as shown in Equation (5). T
𝚰ℂ = ⋮𝓸 ⋱ ℴ 𝓣 · · ·
⋯ H 𝓗 𝓅 ⋮𝓹 ⎯⎯ 𝓛𝓲 (8)
ealthcare 2023, 11, x FOR PEER REVIEW  .. 𝚰 . . ⋯ ..𝓗 𝒚𝒊𝒆𝒍𝒅𝒔 yields 16 of 27
For efficient ℘ , the transmission 𝓗𝓬 ⇔
of
𝚰𝝆 ⋯ 𝓛𝓲 . 𝚰 C at = 𝚰 ℂ =
maximum ⋮ 𝝆 .⋱ speed . ⋮ 𝓬
 𝑺(𝒙)
− ⎯⎯
−− → 𝓛
and𝓕 L
𝓲𝓬 𝒾 minimum (8)
(8)
𝕳 𝓖 𝚰 ⋯ 𝓗
latency 𝓛(𝒙)For 𝓖
|{z}
is required, 𝓜 which
℘ 𝕺,Healthcare is only possible if
⋮ 𝚰 ρatREVIEW is implemented 𝓕and 𝑺(𝒙) working
𝓵 = ⇔ 2023,⋮11, x⋱FOR
efficient 𝝆 · · ·byH 𝓬 𝓬
The insurance 𝖃the transmission claim will beofvalidated
PEER maximum 𝓗𝒸𝓹 with speed theF 𝒸(6) and minimum
help of patient input data 𝚰𝝆 ,
properly, as shown
latency 𝓛(𝒙) inisEquation
required, (5).which𝓓is𝓬 only ⋯ 𝓕 possible if 𝓖 is implemented and working
The insurance
the obtained treatment claim𝓣will 𝓸 , and be through 𝓬
validated the bydata 𝓗𝓹stored with the onhelp 𝓗𝓬 , of that patient
will inputthe
yield data result𝚰 ,
properly, as shownThe insurance
in Equation claim (5). will be𝚰 validated ⋯ 𝓛 by H
For 𝓅 efficient
with the help℘ , the
of transmission
patient input data of 𝝆𝚰 ρat , maxim
The stored claim data the 𝕺obtained
of legitimate
𝖃 will be treatment
HIN fetched claims 𝓣
𝓗𝓸
from 𝓬 𝓛
⇔, and
𝓲 𝓖 𝓬 𝓗
by through
identifying
𝝆
via 𝓖 by the 𝓲 and data
ensuring stored
reporting its on
security𝓕 𝓗
𝓬 𝓬 , as , that
shown will inyield
Equation the result
(8).
the obtained𝓜 treatment 𝕺𝖃 T
𝕳
ℴ , and through the
⋮latency
data stored 𝓛(𝒙) on H𝒸 , that will
is required, which yieldisthe only result of
possible if 𝓖
using 𝕳. This input data of will legitimate
be decrypted HIN 𝓵 =claims and using 𝓛𝓲 ⇔ by suitable
𝓗
⋮ ⇔ ⋱ML
identifying
𝓬 𝚰𝝆and and ⋯reporting
techniques 𝓛𝓲 𝓓
reporting itFwill 𝓕𝕳 𝓬 ,be as shown (6) in Equation (8).
legitimate HIN claims L 𝒾 by identifying 𝓓𝓬 ⇔ 𝕳 𝓖
⋯in ⋮⎧ 𝓕𝓬 ⋱properly, as shown
𝓵 , as inshownEquation in Equation
(5). (8).
classified
2023, 11, xinto
FOR legitimate 𝓛𝓲 and fraudulent claim 𝓜𝓵 = 𝓕𝓬 ,𝕺as 𝖃 shown Equation 𝚰𝝆 ⋮∩ 𝒸
𝓗𝓬 (6).𝓓 (6)16 of 27
 𝓒𝓶 𝓢𝓬
Healthcare PEER REVIEW 𝕳
 𝓓⎧ ⎪ ⋯ 𝚰 𝓕 ∩ 𝓵
𝓗𝓬 ⇔ 𝚰𝝆 ⋯
The stored claim data 𝕺 will be
𝓓,𝓵 the fetched
= 𝓓transmission from 𝓗 𝓒of via
𝓬 𝓖 𝝆∆ by 𝓬ensuring
D 𝓒𝓁𝓶 𝓢𝓬 H its security
𝓵+ 𝕳, 𝒙℘=(𝒙) 𝟎
𝒚𝒊𝒆𝒍𝒅𝒔 𝖃 𝓬𝓶 𝓗𝓬 𝕳 𝓖
𝓑𝓬 =data For efficient
⎯⎯ will 𝒇(𝒙)be ℘
= decrypted 𝟎and 𝐦𝐚𝐱 = ⎪ 𝓢
𝚰 𝓱 at ⇒→ 𝓗 ∩
maximum 𝓬 {𝓗 𝓹 , 𝓗 speed
𝒔 , 𝓗 𝒓 } 𝑺(𝒙) 𝓜
and =minimum𝕺 (9)⇔ ⋮ ⋱
using 𝕳. ThisThe input 𝒙 𝒏using 

suitable ⎨ 𝓗
ρ ∆
HML C𝓖𝓂bySensuring
techniques (7)
𝒸 it willits besecurity 𝓵 𝖃
stored𝓛(𝒙) claim is data 𝓓𝓵𝕺= 𝖃 𝟎 will ∩ +𝕳
𝐦𝐚𝐱be ,fetched
𝒙℘is≥(𝒙) 𝟎 =𝓢 from 𝓢𝓱∆→𝒸𝓬𝓗 via
latency required, which only 𝓬possible
⎪ 𝓬 {𝓗
∆if 𝓖𝓹is ,𝓗 𝒔 ,𝓗
𝓓implemented 𝕳𝒓 } and working(9) 𝓓𝓬 ⋯

𝓛𝓲 data 𝓵
℘(x𝓕) 𝓬and S⎨ 𝐦𝐚𝐱suitable 𝓖 𝒸Equation
→ 𝓗𝓬𝓅ML

classified into 𝕳.legitimate and fraudulent 𝟎 𝒙
claim 𝒏 , as shown in (6).
using This input will be decrypted using Htechniques it will be
According toREVIEW properly,
Equation (7), the as shown
blockchain in Equation digital
max (5).
ledger
=
𝓓 𝓵 will ⎪⎩𝒽 → 𝚰add
𝝆
H ∆ {H
the  ,𝓓
hash 𝓒 of
𝓶
s, H
𝓵 the 𝓢r𝓬
𝕳 } (9)
Healthcare 2023, 11, x FOR PEER
classified into legitimate 𝒚𝒊𝒆𝒍𝒅𝒔𝓛𝓲 and fraudulent
0≤x≤n
𝓓 = 𝓓 + claim𝕳, 𝒙 𝓕 = 𝐦𝐚𝐱 ,
𝟎 as 𝓒𝓖
The
∆shown→ 𝓗
stored 𝓬 inD claim
Equation H data(6). 𝕺𝖃 will16beof fetched 27 from 𝓗
⎩ 𝚰G 𝓬 𝓶 the 𝓒 𝓶 𝓢𝓬

data by executing 𝓒𝓶 and will 𝓑generate 𝓬 = ⎯⎯ 𝒇(𝒙)
a 𝓢𝓬 = that will 𝓵 be𝓵 populated 𝓗𝓬 ⇔using𝚰𝝆
max 𝝆 by →𝕳. HThis
all ⋯ 𝓛
𝓁
entities

𝒸 input data (7)
will be decrypted and using s

𝓲
𝓓𝓵 = 𝟎 ∩𝓓+𝕳 of, 𝓓 𝒙 𝚰𝕳≥ρ+ at𝓖𝟎𝕳, 𝓢𝒙𝓬= 𝟎 𝓒C𝓶 S 𝒸 𝑺(𝒙) and minimum

For
𝕺𝖃 .efficient ℘ , the 𝒚𝒊𝒆𝒍𝒅𝒔transmission 𝓵be=compromised
maximum speed
 𝓂
involved to secure 𝚰𝝆 andCorollary. The 𝓑Security of the 𝓜 ℘ =
can 𝕺 ⇔
𝓵 classified ⋮ if ⋱ 𝓕 ⋮
is generated 𝓛and not detected (6)
by ℘. 𝓕𝓬 , as
𝓬 = ⎯⎯ which 𝒇(𝒙) = 𝓵 𝖃 into legitimate and fraudulent
(7) claim
latency 𝓛(𝒙) is(7), required, is 𝓓 only =ledger𝟎possible
∩ +𝕳 𝓓 ,𝓓 if≥ 𝓖𝟎
𝒙will ⋯ is𝓬𝓢𝓕 implemented 𝓲
and working
According toCorollary.
Equation
Furthermore, The 𝓗 the
𝓣𝓸 𝓹in⋯ ,
Security 𝓗 𝒐𝒓
blockchain
𝒔 𝓗𝓹 𝒓 of 𝓗 the ℘
may digital
can be 𝓵be compromised
involved in 𝓵 𝓬HIN if 𝓕
add
fraud.
𝓬
𝓬
isthe𝓬 hash
generated
The 𝓕 𝓬 ofis the
and not
immediately detected by
detected, ℘.
properly, as shown Equation (5).that will be populated by all the entities𝒚𝒊𝒆𝒍𝒅𝒔 𝓓 =𝓓 +𝕳
data by executing According 𝓒𝓶any
Furthermore,
and and
𝚰
to will
manipulation
=
Equation 𝓗 ⋮ , 𝓗
generate⋱ (7), 𝒐𝒓
in ⋮
the𝜤𝓗 a 𝒓the𝒚𝒊𝒆𝒍𝒅𝒔
𝓢⎯⎯
is may
prohibited
𝓬
blockchain
claim data 𝕺𝖃 will𝓲 be fetched from 𝓗𝓬 𝒸via 𝓖(8)
𝓹 𝒔 𝓛 be involvedusing
digital in
theledgerHIN
SHINFDP. 𝓓
fraud.
F will The add 𝓕 𝓬the𝓑is𝓬hash= ⎯⎯not
immediatelyofits 𝒇(𝒙)
the = by𝓵℘. 𝓵
detected,
Corollary.
The stored ℂ The Security of ℂ ℘ can be compromised if 𝓵 is generated by ensuring and detected
security 𝓓𝓵 = 𝟎 ∩ +𝕳
involveddata to secure 𝚰𝝆any and 𝕺𝓶
𝓒 .𝚰 and 𝜤generate 𝓢𝓬𝓗using 𝚰𝝆SHINFDP. ⋯ 𝓛𝓲The Fbyisall
𝝆 𝓅, ⋯
byusing and
executing 𝕳. This
Furthermore, manipulation
𝖃 H
input H will
data in𝓗will
s or H
ℂ
𝓬 r be
is mayprohibited𝓕a𝓬 involved
be
decrypted 𝓬 ⇔ and
that 𝕳 𝓖will inthe HIN be fraud.
using populatedsuitable ML the entities
immediately
𝒸 techniques itdetected,
will beand
involved Proof.
to secure Assume𝚰 and 𝕺
that . a 𝓗 𝓜 ,𝓵 𝓗 = 𝕺
𝒐𝒓 𝖃 𝓗 ⇔
intends ⋮ According
to ⋱ ⋮
manipulate to Equation 𝚰 . The(7), valuethe blockchain of(6)𝕳 is digital le
any
classified manipulation
into 𝝆 legitimate in 𝖃IC 𝓣𝓸is 𝓛𝓲prohibited
⋯and𝓗
𝓹 𝒔
fraudulentusing 𝒓 the claim
𝓹 help of patient 𝓬input data 𝚰 ,
SHINFDP. 𝓕 , as shown in ℂ
Equation (6).
Healthcare 2023,
The insurance claimProof. 11, x FOR
immutable,
PEER
will beAssume REVIEW
validated so it that
doesby 𝓗 a⋮not 𝓹𝓗 with
allow
𝓹 , 𝓗 the
𝒔⋮ 𝒐𝒓
this 𝓗
𝒚𝒊𝒆𝒍𝒅𝒔
change
𝒓 𝓛 intendsdataand 𝓓 𝓬by to ⋯executing
declares 𝓕
manipulate 𝓬 the 𝓒𝝆
altered𝓶 𝚰 ℂand. The
blockwill as 𝓕
generate
value . of
This 𝕳
awill𝓢is𝓬 that will
16
𝚰 = ⋱ 𝓣𝓸 on⋯𝓗𝓬𝓗 ⎯⎯ 𝓹𝓓𝓲involved 𝟎 𝓒 (8) 𝓬
𝚰= 𝓓The + to 𝕳, the 𝒙between
=result
the obtained treatment 𝓣immutable, ℂ 𝒚𝒊𝒆𝒍𝒅𝒔
, and through
𝓸Proof.
prevent 𝓗𝓹so 𝓗it𝒔the
,claim 𝒐𝒓
does data
𝓗a𝚰𝓑 not stored
from
==allow 𝓗this ℘H,rchange
s manipulating
, that will
𝓵𝒚𝒊𝒆𝒍𝒅𝒔
and 𝓵yield
declares
ℂ . manipulate 𝓗𝓢𝓬𝓬secure ofthe 𝚰 Caltered 𝚰The𝓶and 𝓗value 𝕺𝖃of
𝒔block
and 𝓗
. as H𝓹𝓕 . This
𝓬also 𝑺(𝒙) willand minim
keeps
𝕺 𝖃 ⎯⎯ 𝒇(𝒙) = ⋮ 𝓕transmission 𝓖
The stored Assume that For
data H 𝒓efficient
𝓬𝓅 ,⋯Hwill or 𝓬be
the
intends
fetched tofrom via at.by 𝝆maximum
ensuring 16 its speed is immutable,
security (7)
althcare 2023, 11, x FOR PEER REVIEW
of legitimate HIN claims prevent 𝓛 by identifying and 𝝆 𝚰
reporting
ℂ ⋮ 𝓕 ⋱ , as 𝓓
shown 𝓬 ⎯⎯
= 𝟎 in ∩𝓛 +𝕳
Equation
𝓲 , 𝒙 ≥ 𝟎 (8). 𝓢 of 27
(8)
both𝓲
so𝕳.it Thisparties
does 𝓗input
not,𝓗 on 𝒐𝒓 𝓗
𝒔 the
allow same𝓛(𝒙)
this𝒓 from pace
change manipulating
and and 𝓬 doesand 𝓵
not 𝚰ℂusing.allow
The 𝓢anyone
𝓬 between to violate𝓬 𝓗if𝒔F𝓖 and
𝒸 .the 𝓗 𝓢𝓹it𝓬 .will
also
Below, keeps weand work
The insurance
using
claim will
𝓹 latency
be
data
validated
will beis 𝚰decrypted
byof 𝓗
required,
𝝆 ⋯ declares
with 𝓗 the
which
𝓬 help of
the is altered
𝓕𝓬
only
suitable
patient
block
possible
input
ML as
techniques
data 𝚰
isThis
,
implemented will 𝓣prevent
𝓸be ⋯ 𝓗𝓹
𝒚𝒊𝒆𝒍𝒅
H
both , H
parties
According
𝓅 s ℘
classified into legitimate𝚰 𝓛𝓲 and
For discuss
efficient the
or ,H on
possible
the
to
r the
from
properly, same ways
transmission
Equation manipulating
as pace
shown
(7), 𝓓 the and
of𝓹
fraudulent
𝓵 in𝕳
disrupting𝚰 does
at
blockchain . The
Equation not
maximum S 𝒸 allow
between
(5).
digital anyone
claim 𝓕𝓬 , as shown in𝓵 Equation (6).𝚰ℂ𝓬 = ⋮
the security speedledger Hand s 𝑺(𝒙)
and𝓓 to
integrity H
violate
willand𝓅 also
add the
𝝆
of
minimum ℘
keeps
the 𝓢 .
and
hash Below,
both how
of we
the
parties
the ⋱ ⋮ ⎯⎯
𝓣same ⎧ 𝝆 ∩ C
𝓗𝓖
the obtained
latency treatment
The 𝓛(𝒙)
data discuss
insurance
onby
proposedis 𝓸 , SHINFDP
the
theexecuting
required, andpossible
claim pacethrough
𝓒
will
which and ways
be
helps
and theto
validated
𝓗does
𝓬iswill of𝓒not
data
only save
𝓶 𝓢
stored
disruptingby
allow
the
possible
generate 𝓬 𝓗 ℘ on
a
𝓹 𝓢
the
with
anyone
from
if ,the
that that
𝓬security
towill
security
is help will
violate
implemented and
beof yield
breaches.patient
the
populated the
S
integrity result
.input
and Below,of
by ℘ we
data
working
all and
the 𝚰 𝝆 , how
discuss𝚰
entities ⋯the
the𝓗𝓬
⎪ 𝓶 𝒚𝒊𝒆𝒍𝒅𝒔 𝓬
𝓓𝓵𝓕=,integrity 𝓓𝓵 shown + 𝕳, 𝒙 𝓗= 𝓬⇔𝟎 𝓒𝓶𝚰𝝆 (8).
𝒸
⋯ 𝓛𝓲 𝝆
of legitimate HIN
the obtained
properly, as
involvedclaims
possible
proposed
𝐦𝐚𝐱 to𝓛(𝒙)
treatment
shown ℘ 𝓲 by
ways
SHINFDP
in
secure =𝓣
Equationidentifying
of 𝚰𝓢,𝝆and𝓑
𝓸disrupting
∆ through
helps
→𝓬(5).
and 𝓗 =𝕺{𝓗 and
⎯⎯
to the𝒇(𝒙)
𝖃 . save ,𝓗
reporting
security
thethe ,=data
𝓗𝒓𝓓
℘and from
} 𝓵stored
𝓬 assecurity
𝓜 on 𝕺
= 𝓗of in
℘
𝓬 , breaches.
that𝕳 Equation
and
⇔
𝓖will how
(9) ⋮
yieldthe proposed
⋱ be
the resultSHINFDP
⋮ validated (7)
𝓱 𝓬 𝓹 𝒔 = 𝟎 ∩ +𝕳 , 𝒙 ≥ 𝟎 𝓢
of legitimate HIN 1. 𝟎
helps 𝒙 𝓗𝒏
𝓹 claims ⎨𝓛𝓲 by identifying
to Breaches:
save the ℘ Whenever
from security 𝓗 𝓹 tries
breaches.
and 𝓓 to
reporting 𝕳
𝓵
deliberately The 𝓕 ,
𝖃
insurance
as delay
shown orclaim
𝓬 in manipulate will
Equation the
(8). 𝚰 ℂ , theby 𝓢𝓬𝓗𝓹 with
Healthcare⎪2023, 11, ⎧ x∆FOR𝚰𝝆PEER 𝓗 𝓓𝓬𝓵⇔ ∩ 𝓣 𝕳 𝓵𝚰𝝆⋯ ⋯ 𝓗 𝓛𝓲 𝓬 𝓓𝓬 ⋯ 𝓕 𝓬
1.According𝓗 𝓹 Breaches:
between 𝓗𝓹𝐦𝐚𝐱
to Equation Whenever
and 𝓖→ 𝓗𝓗 𝓬helps
REVIEW
𝓗blockchain
𝓹 tries
𝕳 𝓖
to𝓒
𝓸 the
to 𝓢deliberately
solve 𝓹
obtained
this 𝒚𝒊𝒆𝒍𝒅𝒔 problem treatment
delay 𝓓by 𝓵 or
𝓣
manipulate
providing , and through
athe copy 𝚰ofℂ , the the
the
of the 𝓢data stored
=(7),𝕺 𝒔 the 𝓬digital ledger will add
𝓗 𝓸 the hash 𝓬
⎩ The 𝚰𝝆𝓜𝓵 ⎪ 𝓒𝓬=
𝖃 𝚰claim 𝓶 ⇔ 𝓢⋮ 𝓬𝓶 ⋮ ⋱𝕺 ⋱𝓓will ⋮
𝓵 ⋮𝕳 ⎯⎯ 𝓛 𝓗 𝓖(6)by (8)
data1.by between 𝓗 𝓒𝓹𝓶 and stored
𝓗𝒔 between helpsℂ
∆ ⎧ to 𝚰𝝆both
data asolve 𝓢∩to ofthat
𝖃legitimate
this be
problem fetched
beHIN 𝓲 by claimsfrom
providing 𝓛𝓲 𝓬all via
by the aidentifying
copy ensuring and reportin its secu
transparent
executing
H𝓅 𝐦𝐚𝐱 Breaches: ℘ agreement
(𝒙)and =
Whenever will
𝓢 →
generate
𝓗 H{𝓗 𝓅 𝚰tries 𝓗,𝓓 𝓗𝓬⋯
𝓬parties.
𝓬 𝒔 For, ⋯
𝓗𝓒 𝓗 𝓶 𝓕will
efficient
deliberately
} 𝓢
𝓬𝓬
populated
℘delay , the by
ortransmission
manipulate
(9) of of
entities
the 𝚰 Cat,the
themaximum
𝓗 𝟎 𝒙 using 𝒏 𝚰 agreement 𝕳. This 𝓱 input ⎪ 𝓬 data 𝝆𝓹 will 𝒓
be decrypted 𝓬 𝓕 𝓬 andorusing suitable ML techniques it wi
involved 2. transparent
to Breaches:
S𝒔𝒸securebetween 𝝆 and Whenever
H 𝕺⎨
and (𝒙) H𝓗
𝖃 . between 𝒔 tries
s helps
∆𝓢 →latency
both
∆ to toparties.
𝓓𝓲𝓬𝓵solve
hide
𝕳 𝓛(𝒙) thisits isidentity
problem required, bymisuse providing
which his
is only a𝓥copy
𝒊 to of the if 𝓖 is𝓓
obtain
possible
Corollary. The Security Theofstoredthe ℘𝓗 claim
can be data
classified
compromised 𝐦𝐚𝐱
𝕺
𝓅℘
will⎪
into if be𝓕 = fetched
legitimate
𝓗 is generated 𝓗
𝓛
from {𝓗
and
and 𝓗 , 𝓗 via
fraudulent
not , 𝓗 𝓖
detected }
by ensuring
claim
by ℘. 𝓕 , its
as security
shown (9)
𝓥anyin Equation⎧ 𝚰 (6). ∩
2. The illegal Breaches:
𝒔insurance benefits, 𝖃
Whenever
𝒙 the
claim
𝟎agreement𝒏 will 𝕺𝐦𝐚𝐱 𝖃be 𝓒𝓖
, between
𝓬
⎨→ 𝒔and
validated𝓗
𝓱
tries 𝓢 toby hide
hinders 𝓗𝓹as
𝓹 𝓬 𝒔
its
his
with identity
way𝒓
the help and orof 𝓬
misuse
thus patient prevents his
input 𝒊 datato kind 𝚰 , of
obtain 𝝆
𝓒𝓶
transparent
⎩ 𝚰fraud.
𝓶 𝓣 both𝓬⋯
𝓬 properly,
𝓒 parties.
𝓗 𝓢suitable shown in Equation (5). ⎪ 𝝆 ∆ 𝓗𝓬
𝓹 , 𝓗𝒔 𝕳.
Furthermore, 𝓗using 𝒐𝒓the 𝓗2.
This input
may data
be Breaches:
involved will be
in decrypted
HIN 𝝆 and
The 𝓸 𝓕using ∆𝓶 is 𝓹 𝓬 𝓓𝒚𝒊𝒆𝒍𝒅𝒔𝕳
immediately ML techniques
detected, it yield
will be
illegal
𝒓obtainedfraudulent benefits,
treatment 𝚰 the𝓣
execution. , 𝕺 and , 𝓒 ⎪
throughand 𝓢 𝓬 hinders
the data 𝓵
his
stored way on 𝓓 𝓗
and = , thus
𝓓
that + prevents
𝕳,
will 𝒙 = 𝟎 𝓒
any
the kind
result of
𝚰ℂ =⎩𝐦𝐚𝐱 s⋮ 𝓖⋱ → 𝓗𝓬⋮
H Whenever 𝖃 𝓶 H tries 𝓬 to𝒚𝒊𝒆𝒍𝒅𝒔 hide its identity or misuse his V to obtain
s ℂ 𝓸
𝚰 𝝆 𝓑𝓬 𝓕 =𝓬 , ⎯⎯ ⎯⎯ 𝓢 =𝓛 of𝓵 27 𝓬 𝓵 𝐦𝐚𝐱 ℘ (𝒙)𝓗 i𝓶= (8)𝓢 𝚰→ 𝓗⋯{𝓗𝓛
𝓛claims ,𝓒𝒇(𝒙)
R PEER REVIEW 16 Equation
classified 𝜤ℂinto legitimate
𝓗𝒓 Breaches: 𝚰𝓲ℂ and 𝓗 fraudulent claim as shown 𝓬 𝓲𝓓
in (6). 𝓬⇔ 𝓖 𝓱 𝝆 𝓬 𝓹𝓲
𝓗𝚰ℂreporting
and any manipulation in of 3.is prohibited
fraudulent using the SHINFDP.
execution.
If 𝓛𝒓𝓲tries to𝓂manipulate the𝓶 𝓗 𝓕𝓵𝓬 ,and =asthat 𝟎shown
∩ +𝕳 𝟎,in
𝒙𝒙 ≥ 𝒏𝟎 ⎨ 16 of
legitimate HIN Oby X , identifying
C 𝚰𝝆 S ⋯𝒸and 𝓬 data is stored using
Equation 𝓢 𝕳(8).
Healthcare 2023, 11, x FOR PEER REVIEWillegal benefits, the and hinders his 𝓕 way thus
𝓜
prevents
= 𝕺
any 𝓬 kind ⇔
using ⎪𝐦𝐚𝐱 𝓖 → 𝓗𝓬 ⋮ ⋮
𝓬 𝓬 of 27 ∆⋱
Corollary.
For efficient ℘ The 3. 𝓗
, theSecurity
transmission of𝒓 and
the ℘
Breaches:
𝓔fraudulent isof can 𝚰 Cbe
connected If
𝒚𝒊𝒆𝒍𝒅𝒔 𝓗
compromised
atexecution. 𝒓 tries
maximum to the to 𝓑 𝓓 𝓕
if𝓵 using
manipulate
speed =𝓬 𝓓is𝓵𝑺(𝒙) generated
+
𝕳 𝕳, 𝚰becomes
ℂ ,𝒙andthe =𝟎 and𝓗𝓒
minimum 𝓬 notdata detectedthat isbystored ℘. existing
𝓵 𝖃
𝓗 𝓬
, 𝓗insurance
𝒐𝒓 𝓗 𝓑 = ⎯⎯
According 𝒇(𝒙) = to 𝓬
Equation 𝚰∩ℂ if (7),
. 𝚰𝓗 the 𝓓
blockchain 𝓵
𝓶 𝕳
disconnected
𝕳data digital from
ledger 𝓓(7) will 𝚰add ⎩, 𝚰𝝆𝓓 the𝓬 hash ⋯ 𝓕o
𝓬𝚰∩
Proof.𝓛(𝒙)Assume that 𝓗 ,𝓗
a 𝓹which 𝒔 𝒐𝒓 𝒔𝓗 𝒓 ofintends
𝓬claim to manipulate 𝓓 = 𝟎⎧by +𝕳 The 𝓕with value 𝓢of ofisnot by 𝓵and
𝓬, 𝒙 ≥ ,𝟎the
Furthermore, The
𝓹
𝓔𝓑Security
H may be
theinvolved ℘can
℘Ifwill H bein
𝓖 HIN
validated
𝓑 fraud. The
𝕳 𝓕 is immediately
the help detected,
patient 𝑺(𝒙)
isinput data
℘. using
latency Corollary.
is required, 3. The 𝓬𝓬is
For
and
rwhich efficient
Breaches:
𝒓only is possible
connected ,ifto
r the
betries
the transmission
to
compromised
is manipulate
implemented
𝓬𝓵 using
𝝆of 𝓹
becomes
𝓗 and
is at
generated
𝓬 C 𝓒𝓶activities.
maximum
working 𝓬H 𝓢𝓢𝒸𝓬and
disconnected speedthat
detected
from stored
existing 𝝆 E𝒸
minimum
,
𝜤thistreatment data immediatelyby executing highlights 𝓒
the𝓶altered and the
⎪Honly willfraudulent generate 𝓕𝓬on a that will
𝕺𝖃 yield be populated by from 𝓗𝓬ent
all the
immutable,
properly, asand so
anyitFurthermore,
shown does notobtained
manipulation
the
in Equation allow in
𝓗
latency 𝓹,,ℂ𝓗
and
(5). change
𝒔𝓛(𝒙)
isis prohibited
𝒐𝒓 𝓗
connected is and𝓣using
required,
may declares
𝓸 ,to and
be the
the SHINFDP.
through
involved B 𝒸 using
which in theisHIN data block
The
becomes stored
∆ possible
fraud. as
stored The .claim
𝓕
disconnected This
if 𝓗 𝓬immediately
𝓬 is𝓖𝓬add
,will
data
is that will
from
implemented will
existing be
detected, theand fetched
Bresult
𝒸 , working
which vi
prevent 𝓗𝓹 , 𝓗𝒔and 𝒐𝒓 𝓗 According 𝓑 to
The which
Equation
𝓬 above involved 𝒓(7),
immediately
𝚰
discussion the to 𝐦𝐚𝐱
secure
𝓢 highlights
blockchain
shows ℘ (𝒙)𝚰 that =
anddigitalthe
𝓗 𝓢 𝕺
properly fraudulent
→
. ledger𝓗 𝓗{𝓗 𝓓
implementing , activities.
𝓗 will , 𝓗 } the
SHINFDP hash of
makes the the (9)
HIN
of
any from
legitimate
𝒓 manipulation manipulating
HIN
immediately
properly, inas 𝜤claims
shown ℂ 𝓛in
. The
highlights
iswill
prohibited 𝓲 𝟎by
between
𝒙 𝓬identifying
Equation 𝝆
the fraudulent
𝒏using (5).using
the ⎨and
SHINFDP.
𝖃 𝕳.
and
𝒔Corollary.
𝓱 reportingThis𝓹input
activities.𝓬 𝓹 𝓵
also
The 𝓕𝓬Security
𝒔 keeps
, data
as shown 𝒓 will of thebe ℘ can be compromised
decrypted
in Equation (8). and using if 𝓕𝓬
suita
𝓒 𝚰generate ⋯ 𝓛a𝓲 that
both partiesProof.ondata by executing
the same
Assume paceThe
process
that a and 𝓗 above
fair 𝓶,𝓗
𝓹does
and 𝓗𝓬ℂ⇔
and
𝒔 𝒐𝒓
discussion
transparent.
𝕳not 𝓗allow
𝓖 𝝆
𝒓 intends
shows
anyone to 𝓢to
Furthermore, 𝓬 that properly
violate
manipulate⎪
classified will
thethe
Furthermore, be
proposed
∆
into
ℂ 𝓢populated
𝚰implementing
. 𝓬𝓣 .The 𝓓
Below,
legitimate 𝓗 𝓵⋯SHINFDP
value , 𝓗𝕳 𝓗
we bySHINFDP
𝓛𝒐𝒓all𝓗
of 𝕳
andthe
provides
is entities
may makes
fraudulent be the
flexibility
involved claimHIN to𝓕HIN
in , as frau
sho
involved to 𝓜
secure =fair
The 𝕺
𝚰above and 𝕺 ⇔
discussion ⋮
. security ⋱shows ⋮ that 𝐦𝐚𝐱
properly 𝓗𝓖 →ML 𝓗𝓓𝓬𝓵algorithms
implementing𝚰𝓸𝝆𝓒𝕳SHINFDP 𝓹
⋯(6) 𝒔 𝓲
𝓛𝓲𝓹SHINFDP 𝒚𝒊𝒆𝒍𝒅𝒔
𝒓
makes the HIN 𝓬
discuss immutable,
the possible soways process
the
itAssume
does ofstakeholders
𝓵disrupting
notthat 𝝆𝖃and
allow a this
transparent.
𝓗the
in 𝖃 selecting
change and Furthermore,
cloud
anddeclares ⎧⎩the
services
integrity 𝚰𝚰𝝆and
𝝆the and
of𝓬𝚰any
altered ⇔
∩℘=
proposed and
block ⋮
𝓖 manipulation how
𝓶as ⋱ 𝓢
𝓕 the𝓬𝓬 .⋮ inaccording
This provides
𝜤𝒚𝒊𝒆𝒍𝒅𝒔
⎯⎯ will 𝓛
to flexibility
theirusing choice. to SHINFD
𝓹 , 𝓗𝓓 𝒔 𝒐𝒓 ⋯ 𝓗 𝚰 𝕳
Proof. intends to manipulate 𝕳
𝓒 𝓢 . The value ℂ is prohibited
of is 𝓓 = the
𝓓to + 𝕳, 𝒙 =
process
the stakeholdersfair and transparent.
in selecting 𝓬 cloud 𝓕𝓬𝓜of
𝒓 Furthermore, =
⎪
services
𝓵 𝕺 𝖃
𝓗
the
and 𝓬 ℂ
ML⇔
proposed 𝓶 ⋮
algorithms
ℂ ⋱SHINFDP
𝓬can ⋮ according provides 𝓲
to flexibility
their 𝓵 HIN
choice. 𝓵(6)
proposed SHINFDP
prevent 𝓗𝓹 ,helps
𝓗𝒔 𝒐𝒓tosoThe limited
𝓗save𝒓it from the notand ℘ from expensive
manipulating security storage
𝚰ℂ𝓣. 𝓸breaches.
The⋯and 𝓢𝓬𝓗 the between blockchain
∆ 𝓗𝒔 altered DTL
𝚰𝝆 ⋯
and 𝓗block 𝓹 𝓗
be𝓑𝓬 as
also =keeps
overcome
𝓕⎯⎯. This𝓕𝒇(𝒙)
by =
storing
immutable, the limited does
stakeholders allow in this
selecting𝐦𝐚𝐱
change
℘cloud(𝒙) = servicesdeclares
𝓹
𝒚𝒊𝒆𝒍𝒅𝒔
𝓢𝓱blockchain
→ 𝓗 and the
{𝓗 ML ,𝓓 𝓗algorithms ⋯ 𝓕□ 𝓬 𝓬
accordingbytostoring will 𝓓
their𝓵(9) =HIN +𝕳
𝟎
choice. ∩ ,𝒙 ≥
The stored claim
both parties data 𝕺𝖃,same
The
data
onCorollary.
the on the
will be
pace and
cloud
fetched
and expensive
and
doesfrom 𝚰just =
not 𝓗
storage
saving⋮
allowvia⋱be 𝓖
of
theanyonethe
byhash
⋮ of
ensuring⎯⎯
Proof.
to the 𝓬𝓛
violate data𝓹DTL
its
Assume 𝓬on𝒔 ,can
security
the 𝓗
DTL.
𝓢 𝒓.}be
that 𝓬overcome
Below, a𝓗and 𝓗we , 𝓗detected 𝒐𝒓 𝓗𝒓 byintends to man
prevent 𝓗 The 𝓗
𝓹 limited 𝒔 𝒐𝒓 The 𝓗 and
𝒓 from
Security expensive of 𝟎
manipulating
ℂthe 𝒙 ℘
𝒏 𝓬can
storage 𝚰 of .⎨ The
compromised
ℂ the 𝓢
blockchain between
𝓬 validated if
𝓲 𝓕 𝓬DTLby is 𝓗 generated
can
𝒔 𝓗𝓬 and be overcome 𝓹
also
not
𝓹 the help 𝒔keeps (8)
by storing ℘.HINinput dat
1. 𝕳.𝓗This 𝓗 The insurance 𝕺notclaim will be 𝓓to 𝚰so 𝕳
,Equation
𝓗 𝓢□ with of its patient
is𝓖
Breaches: Whenever data on thetriescloud and just saving
𝚰𝝆security ⋯the hash
𝓗 of the
According ∆ data on DTL. 𝓹 (7), the blockchain digital ledger
using input
𝓹discuss the
bothdata will
possible be
Furthermore,
parties data on onthe
The
decrypted
ways 𝓹
the of
𝓗 stored
same
cloud
𝓹the,to
𝓗and
disrupting deliberately
pace
𝒔
claim
𝒐𝒓
and
using
𝓗and
just
𝒓
thedata
may
does
saving
delay
suitable be or
will
𝖃involved
the
ML⎪
allow
manipulate
and
𝐦𝐚𝐱
hash 𝓬be in fetched
techniques
immutable,
integrity
𝓖
anyone
of →
HINthe 𝓕𝓗
the
𝓬 fraud.
datatofrom
itof 𝓵ℂ℘
will
violate
on itthe
TheDTL.
be
does
and𝓬𝓕 via
the𝓬
𝓬how
 not
𝓬
by
.allow
the
𝓢immediately ensuring
Below, thiswe change
detected, security
and declares
𝓗discussand𝓛 𝕳. obtained 𝓕from treatment ⎩ 𝓣 , and through 𝓬
𝓒 the 𝓒 𝓢data stored on 𝓗 , that 𝓢 will yield the re
𝓲𝓗and
𝚰 𝓗 𝓹 , 𝓗of 𝒐𝒓 of𝓗℘𝒓input 𝓬 𝚰ℂ . The 𝓢
classified into
between legitimate using fraudulent
helps This
totosave solve input
claim ℘ data , as will
shown be in data
decrypted𝓸
Equation by executing
and (6). using suitable and will
ML generate
techniques 𝓬 a it that
will bewill be
proposed 𝓹 SHINFDP
The andthe 𝒔any
insurance
helps
possible
manipulation
claim ways
of will
the inthis
of
legitimate be
problem
𝓬is
𝜤disrupting
ℂvalidated
security
prohibited
HIN
the
by
claims
byusing
𝓗
providing
breaches.
security
𝓛
𝝆 prevent
withthe
by
and
SHINFDP.
the
identifying
a integrity
help
copy 𝓶
of
𝒔𝓶
patient
and
𝓬 the
reporting
from
and data manipulating
how
𝓕 , 𝚰
as
the ,shown in Equation
𝓬 b
transparentproposed
agreement classifiedhelps
between into parties.
both legitimate 𝓛℘ 𝓲= and 𝓹
involved
fraudulent
𝓒security
𝓲 both claim to
partiessecure 𝓕𝓬on , as𝚰𝝆the and
shown same 𝕺𝖃in.pace Equation 𝓬and does 𝝆 (6). not allow any
𝒚𝒊𝒆𝒍𝒅𝒔
SHINFDP 𝓣to 𝓓,𝓵to =save 𝓓through
𝓵+ the 𝕳, 𝒙 from𝟎data 𝓶 storedbreaches. 𝓗
2. 𝓗𝒔 1. 𝓗the
Breaches: obtained
𝓑
Whenever
Breaches: = ⎯⎯treatment
𝓗
Whenever 𝒇(𝒙) =
tries 𝓗 𝓸 and hide
tries its
to identity
deliberately the or misuse
delay or on
his
manipulate 𝓥 𝓬 , to that will𝚰 yield
(7)
obtain
the , the 𝓓 the
𝓢 result
𝕳
𝓹℘ if𝓵 𝓕 𝓣𝓸ofby⋯𝕳℘.the
𝓬 𝓗+𝕳 , 𝓗can 𝓒not
𝒔 𝒐𝒓 𝓗 𝚰ℂ ∩
𝓓𝓛𝓹𝓵 =by a𝟎the ∩ ,𝒙 ≥𝒚𝒊𝒆𝒍𝒅𝒔
be𝟎 intends 𝓓to discuss the possible ways of disrupting is 𝓹security a
𝓗
𝓕=𝓬𝓬, 𝓓 + 𝕳,𝚰𝝆𝒙 = inℂ𝟎
𝓹 Corollary.
Proof. Assume 𝒔
The Security that of 𝒓𝓢reporting
compromised 𝒊is generated
manipulate and
. The detected
𝓶𝓵 𝓬value
of legitimate 𝕺 HIN 𝓒 claims 𝓢 𝓲 identifying 𝓑 = ⎯⎯and 𝓬
𝒇(𝒙) = as𝓵 ⎧ shown Equation
𝓒 𝓢 (8). 𝒚𝒊𝒆𝒍𝒅𝒔
(7)
illegal benefits,
between
1. 𝓗 the 𝓗 𝖃and𝓶𝓗𝓗
, and ,𝓗
helps hinders
𝒔 𝒐𝒓not
to𝓗 solve
𝓗 his way
𝓬
this and
problem thus prevents
by
proposed 𝓓 providing
= 𝟎or any
SHINFDP kind
+𝕳aaltered
∩manipulate ,𝓕copy
𝒙𝓬≥𝓬is𝓗of helps of
𝟎immediatelythe
to𝚰 save
, =the
𝓬𝓕 the 𝓢This ℘ from security ⎯⎯ 𝓛𝓲b
𝓢𝓶 𝓬⋮.detected,
𝓬 ⋱ will⋮
Furthermore, 𝓬does 𝓹may be toinvolved
changein HIN ⎪
𝓹 𝓹 Breaches:
immutable, 𝒔Whenever
so 𝓹it 𝒓allow tries this deliberately and 𝓵 fraud.
delay
declares The
the the
block 𝓬 ℂ ℂas
According
fraudulent to Equation
𝚰 ℂ execution.
(7), the blockchain digital ledger 𝓓 𝓵 will add 𝓓 the 𝕳 hash of the ∆
transparent and
between agreement
prevent 𝓗
any 𝓗
manipulation between
𝓹𝓹 , 𝓗𝒔 𝒐𝒓 𝓗
and 𝓗 in 𝜤
both
helps is parties.
prohibited
𝒓ℂ from manipulatingto solve using
this
𝚰the 𝐦𝐚𝐱
the
∩ problem 𝓵℘
𝚰ℂ . The 𝓢𝓬 between
SHINFDP. (𝒙) by = 𝓢𝓱 → 𝓗
providing 𝓗𝓬 {𝓗 a
𝒔 and 𝓹 , 𝓗𝓗
copy 𝒔 ,𝓹𝓗also
of 𝚰 𝝆} keeps 𝓬
the ⋯ 𝓗 𝓕𝓬
data3. by𝓗executing 𝓗𝒔 𝓒 and
𝓗both will According
generate a 𝓢𝒔tries 𝓬 to
𝚰that Equationwill 𝓗both be⎧ (7),
populated 𝝆 blockchain 𝟎 𝒙 𝓒by𝒏 all𝓢
𝓗using digital
the ⎨hisledger
entities 𝓓 𝓵 will add 𝓗.𝓹Below,
𝒓the
trieshash of the
Breaches: If transparent
𝒓 tries to on 𝓗
manipulate ℂ , the data that 𝓥𝒊violate
𝒓 2. 𝓶
Breaches: Whenever to hide its 𝓗𝓬 is
identity stored
1. or 𝓶 misuse 𝓬 Breaches: Whenever
to towe deliberately d
parties agreement the 𝒔 same between pace 𝓬and ⎪ parties.
does not allow 𝓹anyone to ∆ obtain the 𝓓 𝓢 𝕳
involved to secure 𝚰𝝆 benefits,
and 𝕺𝖃 .data by ,executing 𝓒𝓶 and will generate
∆ The athus 𝓢𝓬 that
insurance ⎪will
𝐦𝐚𝐱𝓗claim be 𝓖 →populated
will 𝓗 𝓗 be 𝓬
𝓵 by all the
validated by 𝓗
entitieswith the h
𝓔𝓬 and is connected
illegal
2. 𝓗 Proof. to the
𝒔 Breaches:
discuss the
Assume
the 𝕺
𝓑𝓬 𝖃Whenever
using𝓒
possible 𝓶 𝕳
that and 𝐦𝐚𝐱 𝓢
𝓗 𝓗
abecomes
ways ℘𝓹(𝒙)
𝓬 ,𝓗
hinderstries
of 𝒔= 𝒐𝒓to𝓗 his
𝓢𝓱hide
disconnected
disrupting way
𝒓 →intends 𝓗its and
𝓬 {𝓗
the from 𝓹, 𝓗
identity between
to existing
security 𝒔 ,𝓗
prevents
manipulate
orand 𝒓 }⎩ misuse 𝓹 𝚰ℂhis
any
𝚰integrity
and kind
. The 𝓥𝓬
of𝒊 𝒔𝓒of helps
℘𝓶 obtain
value
to and 𝓢(9)of 𝕳 is
to solve 𝓹this proble
𝟎 𝒙 𝒏 𝚰𝝆𝒔 and 𝕺 𝝆 𝓬 how the
fraudulent involved
𝚰 execution.
immutable,
to
so𝓣it does
secure
⋯ not 𝓗allow this⎨𝖃change
. theand obtaineddeclares transparent treatment
the altered agreement𝓣𝓸block , and as through
𝓕 . This
between the
both willdata stored on
parties.
Healthcare 2023, 11, 2257 17 of 27

4.2. Industrial Evaluation

A focus group technique was used to evaluate the ease of use, security, understanding,
and structure of the proposed SHINFDP. Numerous factors, such as cost-effectiveness,
observation of body language, and enhanced consumer engagement, contribute to the
selection of the focus group technique for evaluation. Focus groups are comparatively
inexpensive to conduct, and the moderator is able to observe the interviewee and modify
the conversation accordingly. It is also more engaging than impersonal research methods
such as questionnaires and surveys. However, sizable samples are difficult to target when
using focus groups. To measure the ease of use, security, and structure of the SHINFDP,
closed-end questions were designed and asked to the focus group respondents. In contrast,
the concept mapping technique was used to evaluate the understanding of the SHINFDP.
Seven professionals were included in the focus group, from which four belonged to a
well-known HIN provider company in Saudi Arabia, while three of them were healthcare
professionals dealing with insurance claim handling in a well-known public sector hospital.
Below, we describe the details of each measure used to evaluate SHINFDP.

4.2.1. Evaluating Ease of Use for SHINFDP

A desktop meeting was arranged with respondents of the focus group in which
the author presented the workings of the SHINFDP to the respondents via PowerPoint
presentation. At the end, the respondents were asked five closed-ended questions to
evaluate the feasibility of the SHINFDP. Table 1 shows the responses of focus group
members where SA refers to strongly agree, A→agree, D→disagree, SD→strongly disagree,
and N→Neutral. The threshold value was set to 50%, which means that if more than
50% of the respondents agree on any statement, it will be considered acceptable. The total
was calculated by adding the value of SA and A.

Table 1. Evaluation results for “ease of use” of SHINFDP.

Ease of Learning SA A D SD N Total Mean

Comprehending the information flow and
E1 the function of cutting-edge technologies 2 4 1 0 0 5 71%
in SHINFDP is straightforward.
It is easy to handle the end-to-end HIN
E2 4 3 0 0 0 7 100%
process using SHINFDP
It is easy to use SHINFDP to assess
E3 2 2 1 0 2 4 71%
organizations’ readiness for SHINFDP
Each individual technology is easy to
E4 2 3 1 0 1 5 71%
understand and implement
Some training needs to be provided for the
E5 1 2 2 1 1 3 43%
use of SHINFDP

Figure 6 depicts the respondents of the focus group’s opinions regarding the usability
features of SHINFDP. E1, E2, E3, and E4 were positive statements, in which a higher
agreement percentage refers to more usability, while E5 was negatively assessing the
usability of SHINFDP. Figure 6 shows that the agreement percentage of E1, E2, E3, E4 is
more than the threshold value, while the agreement percentage of E5 is 43%. This shows
that SHINFDP is easy to use and is adaptable by healthcare organizations.

4.2.2. Evaluating the Security Features of SHINFDP

Using the same pattern mentioned in “ease of use” evaluation, respondents of the focus
group were asked five close-ended questions for the security evaluation of SHINFDP. The
questions and responses are given in Table 2, along with the total and average percentage
of agreement.
Healthcare 2023, 11, 2257 18 of 27

Healthcare 2023, 11, x FOR PEER REVIEW 18 of 27

Ease of use for SHINFDP

120%

Responses of Focus group member

100%

80%

60%

40%

20%

0%
E1 E2 E3 E4 E5
Usability evaluation of SHINFDP

Figure 6.
Figure 6. SHINFDP
SHINFDPevaluation
evaluationregarding “ease
regarding of use”
“ease criteria.
of use” criteria.

4.2.2. 2.
Table Evaluating the Security
Security evaluation ofFeatures
SHINFDP.of SHINFDP
Using the same pattern mentioned in “ease of use” evaluation, respondents of the
Security focus group were asked SAfive close-ended
A questions
D for
SDthe security
N evaluation
Total of Mean
SHINFDP. The questions and responses are given in Table 4, along with the total and
Using SHINFDP makes the HIN process
S1 3
average percentage of agreement. 3 1 0 0 6 85.7%
end-to-end secure.
Multilevel security is provided in
Table 4. Security evaluation of SHINFDP.
SHINFDP by first handling it at the
S2 Security cloud level using encryption 5 2SA A 0 D 0
SD N 0 Total 7Mean 100%
healthcare
Using SHINFDP
and then using BLC. makes the HIN process end-to-end
S1 3 3 1 0 0 6 85.7%
secure.
It is difficult for any stakeholder to
S3 Multilevel security 2 3 0 0 2 4 71%
manipulate the HINisclaim.
provided in SHINFDP by first
S2 handling it at the healthcare cloud level using encryption 5 2 0 0 0 7 100%
ML algorithms further improve security by
S4 and then using BLC. 3 3 0 0 1 5 71%
filtering HIN claims.
It is difficult for any stakeholder to manipulate the HIN
S5 S3 5G makes the HIN process more secure
claim. 2 22 3 1 0 10 2 2 4 371% 57%
ML algorithms further improve security by filtering HIN
S4 3 3 0 0 1 5 71%
claims. The security evaluation results are depicted in Figure 7; the agreement percentage
S5 5G makes the HIN process more secure
regarding the five mentioned elements 2 2 higher
was 1 1 the threshold
than 2 3 value.
57%This shows
that SHINFDP helps to improve the security of the HIN process.
Healthcare 2023, 11, x FOR PEER REVIEW 19 of 27
The security evaluation results are depicted in Figure 7; the agreement percentage
regarding the five mentioned elements was higher than the threshold value. This shows
that SHINFDP helps to improve the
Security security of
evaluation of SHINFDP
the HIN process.

57%
85.70%

S1
S2
71% S3
S4
S5
100%

71%

Figure 7.
Figure 7. Security
Securityevaluation
evaluationof SHINFDP.
of SHINFDP.

4.2.3. Evaluating Understanding Feature of SHINFDP

In order to evaluate the understanding level of respondents regarding SHINFDP, a
concept mapping technique was used. Concept maps are diagrams that express
information visually. Charts, graphic organizers, tables, flowcharts, Venn diagrams,
 S4
S5
100%

71%
Healthcare 2023, 11, 2257 19 of 27

Figure 7. Security evaluation of SHINFDP.

4.2.3. Evaluating Understanding Feature of SHINFDP

4.2.3. Evaluating Understanding Feature of SHINFDP
In order to evaluate thetounderstanding
In order level of respondents
evaluate the understanding regarding
level of respondents SHINFDP,
regarding a
SHINFDP, a
concept mapping technique
concept was used.
mapping Concept
technique wasmaps areConcept
used. diagrams that express
maps information
are diagrams that express
visually. Charts,information
graphic organizers, tables, graphic
visually. Charts, flowcharts, Venn diagrams,
organizers, timelines,
tables, flowcharts, anddiagrams,
Venn T-
charts are a fewtimelines,
examplesand of T-charts
how they aremight be presented.
a few examples of howConcept maps
they might are particularly
be presented. Concept maps
helpful for assessing how well helpful
are particularly a phenomenon is understood.
for assessing how well a Figure 8 shows
phenomenon the exampleFigure 8
is understood.
concept map forshows the examplewhich
the SHINFDP, conceptshows
map for
howthe SHINFDP, which showsof
various components how
thevarious components
SHINFDP
are related. of the SHINFDP are related.

Figure
Figure 8. A partial 8. Amap
concept partial
ofconcept map of the SHINFDP.
the SHINFDP.

In order
In order to evaluate thetofocus
evaluate the focus
group group respondents’
respondents’ understanding
understanding of SHINFDP,
of SHINFDP, we we
extracted eight relationships between concepts and asked the respondents
extracted eight relationships between concepts and asked the respondents to evaluate to evaluate
whether this
whether this relationship relationship
should should
exist or not. exist
Tableor3not. Table
shows the5 shows therelationships
selected selected relationships
and and
respondents’ responses.
respondents’ responses.

Table 3. Evaluating relationships between concepts.

S.No Relationship between Concepts Yes No Percentage (Yes)

Approaches
1 HIP −−−−−−→ HSP 7 0 100%
Uses
2 HSP −−→ SHS 7 0 100%
have
3 BLC −−→ DTL 6 1 85.7%
Keep Hash
4 DTL −−−−−−→ HCC 6 1 85.7%
Facilitate storage
5 5G −−−−−−−−−−→ HCC 6 1 85.7%
Store info
6 SC −−−−−→ HIP 5 2 71%
Maintain
7 BLC −−−−−→ SC 5 2 71%
Fetch data for prediction
8 PL −−−−−−−−−−−−−−→ HCC 6 1 85.7%

Table 3 shows that the respondents’ evaluation of SHINFDP is more than 70% for all
relationships. This shows that SHINFDP is easy to understand for healthcare professionals.
 8 PL ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ HCC 6 1 85.7%

Table 5 shows that the respondents’ evaluation of SHINFDP is more than 70% for all
relationships. This shows that SHINFDP is easy to understand for healthcare
Healthcare 2023, 11, 2257 professionals. 20 of 27

4.2.4. Evaluating the Structure of SHINFDP

SHINFDP the
4.2.4. Evaluating is a Structure
layered framework
of SHINFDPconsisting of six layers, and different cutting-edge
technologies
SHINFDPareis a proposed at various
layered framework layers. of
consisting Three questions
six layers, were used
and different to evaluate the
cutting-edge
structure of SHINFDP which are given in Table 6.
technologies are proposed at various layers. Three questions were used to evaluate the
structure of SHINFDP which are given in Table 4.
Table 6. Evaluating the structure of SHINFDP.
Table 4. Evaluating the structure of SHINFDP.
S.NO The Structure of SHINFDP SA A D SD N Total Mean
S.NO All the
The layers
Structure of SHINFDP
of the SHINFDP are self- SA A D SD N Total Mean
S1 All the layers
explanatory andof require
the SHINFDP are
no further explanation 4 2 1 0 0 6 85.7%
S1 self-explanatory and require no further 4 2 1 0 0 6 85.7%
to be used effectively
explanation to be used effectively
SHINFDP is practical and applicable in the HIN
S2 S2 SHINFDP is practical and applicable in
2 23 31 1 0 0 1 1 5 5 71% 71%
industry
the HIN industry
TheThe
distribution
distributionofof technologies
technologies at at various
S3 S3 3 32 21 1 0 0 1 1 5 5 71% 71%
layers is correct
various layers is correct

Figure9 9depicts
Figure depictsthethe responses
responses of focus
of focus group
group respondents
respondents regarding
regarding the evaluation
the evaluation
of SHINFDP’s structure. The agreement percentage for all three questions
of SHINFDP’s structure. The agreement percentage for all three questions was more was more than
70 percent,
than showing
70 percent, showingthatthat
SHINFDP
SHINFDP has
hasa simple
a simpleand
andeasy-to-understand structure.
easy-to-understand structure.

SHINFDP's Structure Evaluation

71%
85.70%

S1
S2
S3
71%

Figure 9. Respondents’ evaluation regarding the structure of SHINFDP.

4.3. Evaluation Using a Case Study

Because of the COVID-19 epidemic, individuals have become more sensitive to
positive living choices. They have started adopting healthier lifestyles using cutting-edge
technologies such as wearable devices (WRD) and smartphone healthcare applications.
Wearable technology can be any electronic device worn on the user’s body [57–59].
Smartphone healthcare applications provide medical-related services. Clinical research
may take advantage of the data collected by healthcare applications and WRDs. Some
HIP businesses utilize the data collected from WRDs for future predictions [60,61].
They employ WRDs as part of their marketing strategy to engage with HINSs. HIP
enterprises may give prizes or discounts on premium pricing and free WRDs under
specific situations as part of their marketing strategy. Conditions include the HINS
using a WRD, providing authorization for WRD tracker data sharing, and demonstrating
healthy habits such as walking a specific amount of steps each week, counting calories,
Healthcare 2023, 11, 2257 21 of 27

and keeping a healthy heart rate using WRD [62]. In 2016, a US-based HIN startup
established a platform to inspire and reward HINSs to partake in healthy activities using
a WRD. According to data collected from WRDs, the company offers HINSs challenges
and competitions, and in return, HINSs receive workout equipment, gift vouchers, and
lower HIN insurance rates [54]. Another US-based partner company provides free MI-fit
fitness trackers to all HINSs. This business is one of the first HIP companies to analyze
insurance premium rates for HINSs using WRD data. They connected their HIN account
to the HINS’s biometric data [54]. An HIN corporation located in France encourages
HINSs to exchange health-related personal data from a WRD. By sharing their WRD data,
HINS may earn fit points. This information is used in the firm’s insurance underwriting
procedure [63]. An Australian-based HIP startup also collaborated to incorporate HINS
data from different WRD trackers into their accounts. This information evaluates HIN
risks and sets rates [64]. As previously stated, the HIN business utilizes WRD data to
evaluate premium pricing and insurance risk assessment; however, it has not yet been
used for HIN fraud detection.
HIN fraudsters become smarter daily and develop new methods to commit fraud.
As a result, every HIN company must discover a new and enhanced future strategy to
prevent fraud utilizing a WRD’s data. As a result, we provide a case study to meet the
criteria mentioned above for HIN companies. Figure 10 depicts the working of the case
study that addresses HIN security and privacy concerns. The HINS is wearing several
WRD/sensors in Figure 10. The HIN company may provide the WRD for marketing and
fraud detection in HIC. The HINS WRD communicates with HINS smartphones through
Wi-Fi or Bluetooth. The data produced by the WRD is encrypted and digitally signed
before being saved in IPFS. The data from the HIP and HSP is encrypted, digitally signed,
and transferred to the fraud detection system. Data preparation follows data production
and involves data cleansing, integration, transformation, and reduction methods. The
HINS distributes permission for sharing personal healthcare data using a preset smart
contract before data preparation. The HIP business provides the public key to the HINS
and HSP. If an HINS approves exchanging personal health data, the HIN corporation
will reward the HINS. An AI algorithm is used to preprocess the data to determine if
the HIC is fraudulent or authentic. If a claim is genuine, the BLC network consensus
permits automatic HIC coverage payment to the HINS or HSP. However, if the HIC is
false, the HINS will pay a penalty.
The use of WRDs in a HIC has various advantages. It enhances the HINS’s involve-
ment with the HIP regularly and in real time. The WRD will be used to track the HINS’s
whereabouts. HINS location data and real-time health status may be used to determine
what occurred before, during, and after a hospitalization emergency. HIN corporations
may use this data to identify pain levels, prevent medication-seeking behavior, and
uncover inconsistent conduct with a HIC. Furthermore, BLC technology provides new
avenues for improving inefficiencies in the HIN market. It may benefit HIPs by improv-
ing security, lowering costs, and increasing confidence. BLC in the HIN business also
offers real-time data that may be utilized for risk evaluation, claims settlement, and
other purposes.
Based on the case study and existing state-of-the-art review, it can be concluded that
a combination of cutting-edge technologies, including BLC, 5G, cloud, AI, and wearable
devices, can improve the HIC process and make it more transparent. When HIC fraud
reduces, HIP firms will benefit from it, and HINSs will also benefit in the form of lower
HIN premium amounts. Academia and practitioners should provide easy solutions for
implementing these solutions in reality so that more people around the globe could
benefit from healthcare facilities.
Healthcare 2023, 11, x FOR PEER REVIEW 22 of 27

Healthcare 2023, 11, 2257 22 of 27

business also offers real-time data that may be utilized for risk evaluation, claims
settlement, and other purposes.

Figure 10.
Figure 10. A
A case
case study
study of
of HIC
HIC fraud detection.
fraud detection.

4.4. Discussion
Based on the case study and existing state-of-the-art review, it can be concluded that
a combination
A reducedofmortality
cutting-edgerate,technologies,
better healthincluding
outcomes, BLC,
and5G, cloud, AI,
increased and wearable
productivity are
devices,
all relatedcan
to improve the HIC
having health process and
insurance. make
Health it more also
insurance transparent.
makes itWheneasierHIC fraud
to receive
reduces, HIP
treatment. firms
Even will benefit
though from it,advantages,
it has many and HINSsmany will also benefit
people stillindothe form
not haveof health
lower
insurance,
HIN premium which puts their
amounts. physical, and
Academia emotional, and financial
practitioners shouldhealth
providein danger. One of the
easy solutions for
challenges
implementing thatthese
the HIN procedure
solutions must contend
in reality so that with
moreispeople
fraud with
aroundthe HIN. In this
the globe sort
could
of fraud,
benefit a health
from insurance
healthcare company is given information that is either false or deceptive
facilities.
to trick them into paying unauthorized benefits to the policyholder, another party, or
the
4.4. organization
Discussion delivering services. It is possible for the insured person or the supplier
of health services
A reduced to commit
mortality rate,the offence.
better health The majorityand
outcomes, of fraud is conducted
increased by HINSs
productivity are all
and HSPs; however, HIPs are also guilty of conducting fraud on
related to having health insurance. Health insurance also makes it easier to receiveoccasion, although their
involvement
treatment. Even in fraud
thoughis often
it haslimited
many to intentionally
advantages, delaying
many people thestill
processing of claims
do not have or
health
artificially understating their value. We presented a six-layered framework
insurance, which puts their physical, emotional, and financial health in danger. One of the that utilizes the
potentials
challengesofthatcutting-edge technologymust
the HIN procedure for safeguarding
contend withhealthcare datathe
is fraud with andHIN.
combating
In thisHIN
sort
fraud in order to make the HIN process smart and secure. An assessment of the suggested
of fraud, a health insurance company is given information that is either false or deceptive
framework was carried out with the assistance of mathematical modelling, an evaluation
to trick them into paying unauthorized benefits to the policyholder, another party, or the
of the industry that made use of the focus group approach, and a case study. The outcomes
organization delivering services. It is possible for the insured person or the supplier of
indicate that the suggested framework is straightforward, useful, secure, and adaptable to
health services to commit the offence. The majority of fraud is conducted by HINSs and
various circumstances.
HSPs; however, HIPs are also guilty of conducting fraud on occasion, although their
The proposed approach is different from the existing solutions [5,15,24,53–56] in the
involvement in fraud is often limited to intentionally delaying the processing of claims or
following ways:
artificially understating their value. We presented a six-layered framework that utilizes
• potentials
the Existing solutions mainlytechnology
of cutting-edge focus on afor particular technology
safeguarding and data
healthcare try toand
determine
combating its
potential in improving the HIN process;
Healthcare 2023, 11, 2257 23 of 27

• Existing solutions discuss the benefits of using BLC in the HIN process, but they do not
evaluate the solution due to the limitation of there being less use of BLC in healthcare;
• The existing studies used a single method of evaluation, while the proposed solution
is evaluated using hybrid approach (mathematical modeling, focus group study, and
industrial case study), which shows the generalizability of research findings;
• Existing studies mainly emphasize the adaptability of the HIN process and provide
various techniques for it. Furthermore, there exist no solutions that specifically address
the end-to-end HIN process cycle security.

4.5. Limitations of the Study

Below are some limitations of this research work.
1. We evaluated the proposed framework SHINFDP using a focus group technique
in which only seven professionals were involved. However, we tried to choose a
representative sample.
2. The use of mentioned technologies will add to the cost for HINP and HSP, but the
enormous costs associated with insurance fraud will be avoided. The use of cutting-
edge technology requires expenses in terms of hardware and training, but it is a
one-time investment on the part of healthcare organizations and HINP, with long-
term rewards. Other advantages of this automation will include convenience for the
parties engaged in the HIN process, no need for manual visits, fast communication
between health insurance subscribers and health insurance providers, audit and
checking, lack of fraudulent activities, easy claim verification, and many more that
will also indirectly save on costs.
3. ML and cloud computing are widely used in healthcare for prediction and storage.
However, blockchain technology is somewhat new and is not commonly adopted in
the HIN process. Therefore, we could not implement the proposed framework in a
real setting.

5. Conclusions and Future Work

We presented a comprehensive summary of HIN fraud detection and related security
vulnerabilities in HIC. First, we discussed the background of HIN, including the HIN
revolution, varieties of HIN fraud, and the HIC process. Then, we discussed the many
parties in the HIC process and the accompanying fraudulent acts that lead to HIN fraud.
We discussed the role of BLC and AI in HIN. Given the significance of HIN fraud, we
presented SHINFDP, an end-to-end solution for safe HIC processing to prevent HIN fraud.
SHINFDP is made up of six layers: the user layer, the communication layer, the storage
layer, the security layer, the prediction layer, and the HIP layer. Each layer contributes
to the improvement of the HIC process. The proposed framework was refined with the
assistance of a case study. The results indicate that the proposed method may assist in
identifying and mitigating HIN fraud and make the HIC process more transparent. In
the future, we want to verify the suggested framework by deploying it in a real-world
healthcare context or through simulation. The proposed framework might be improved
further by using more cutting-edge tools and technologies to optimize the HIC process.

Author Contributions: Conceptualization, F.A.-Q.; data curation, S.T.; formal analysis, M.H.; funding
acquisition, F.A.-Q.; investigation, M.H.; methodology, F.A.-Q. and M.H.; project administration,
F.A.-Q. and S.T.; visualization, S.T.; writing—original draft, M.H. All authors have read and agreed to
the published version of the manuscript.
Funding: This work was funded by the Deanship of Scientific Research at Jouf University under
grant No. DSR2023-NF-09.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Healthcare 2023, 11, 2257 24 of 27

Acknowledgments: This work was funded by the Deanship of Scientific Research at Jouf University
under grant No. DSR2023-NF-09.
Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations

Acronyms Used for

AI Artificial intelligence
BLC Blockchain
DL Deep learning
DTL Digital ledger
HC Healthcare center
HIC Healthcare insurance claim
HIN Healthcare insurance
HIP Healthcare insurance provider
HCC Healthcare cloud
SHINFDP Smart healthcare insurance framework for fraud detection and prevention
CET Cutting-edge technologies
IPFS Interplanetary File System
ML Machine language
WRD Wearable devices
SHS Smart healthcare system
SL Supervised learning
SSL Semi-supervised learning
USL Unsupervised learning
HINS Healthcare insurance subscriber
PL Prediction layer
Notations Used for
Iρ Patient data input
To Obtain treatment
H Hash/unique ID
IC Insurance claim
℘ HIN process
Sh Smart healthcare system
Vi Valid insurance
Hs HIN subscriber
G 5G
Fc Fraudulent claim
Li Legitimate IC
Hc Healthcare cloud
Ml ML algorithm
Bc Blockchain
Dl Digital ledger
Cm Consensus mechanism
Sc Smart contract
OX Store claim data
Hr Healthcare professional
Ec Encryption
Dc Decryption
H√ HIN provider

References
1. Rajawat, A.S.; Bedi, P.; Goyal, S.B.; Shaw, R.N.; Ghosh, A.; Aggarwal, S. Ai and blockchain for healthcare data security in smart
cities. In AI and IoT for Smart City Applications; Springer: Singapore, 2022; pp. 185–198.
2. Li, J.; Lan, Q.; Zhu, E.; Xu, Y.; Zhu, D. A Study of Health Insurance Fraud in China and Recommendations for Fraud Detection
and Prevention. J. Organ. End User Comput. 2022, 34, 1–19. [CrossRef]
Healthcare 2023, 11, 2257 25 of 27

3. Lavanya, S.; Kumar, S.M.; Kumar, P.M. Machine Learning Based Approaches for Healthcare Fraud Detection: A Comparative
Analysis. Ann. Rom. Soc. Cell Biol. 2021, 25, 8644–8654.
4. Huang, X.; Wu, B. Impact of urban-rural health insurance integration on health care: Evidence from rural China. China Econ. Rev.
2020, 64, 101543. [CrossRef]
5. Saldamli, G.; Reddy, V.; Bojja, K.S.; Gururaja, M.K.; Doddaveerappa, Y.; Tawalbeh, L. Health care insurance fraud detection using
blockchain. In Proceedings of the 2020 Seventh International Conference on Software Defined Systems (SDS), Paris, France, 20–23
April 2020; pp. 145–152.
6. van der Wielen, N.; Channon, A.A.; Falkingham, J. Does insurance enrolment increase healthcare utilisation among rural-dwelling
older adults? Evidence from the National Health Insurance Scheme in Ghana. BMJ Glob. Health 2018, 3, e000590. [CrossRef]
[PubMed]
7. Sheshasaayee, A.; Thomas, S.S. A purview of the impact of supervised learning methodologies on health insurance fraud
detection. In Information Systems Design and Intelligent Applications, Proceedings of Fourth International Conference INDIA 2017;
Springer: Singapore, 2018; pp. 978–984.
8. Michaela, S.M.; Nurmalasari, M.; Hosizah, H. Fraud in healthcare facilities: A Narrative Review. Public Health Indones. 2021, 7, 166–171.
[CrossRef]
9. Okuzu, O.; Malaga, R.; Okereafor, K.; Amos, U.; Dosunmu, A.; Oyeneyin, A.; Adeoye, V.; Sambo, M.N.; Ebenso, B. Role of digital
health insurance management systems in scaling health insurance coverage in low- and Middle-Income Countries: A case study
from Nigeria. Front. Digit. Health 2022, 4, 1008458. [CrossRef]
10. Chakravarthi, D.S.; Barot, H.; Kalkar, P.; Aute, K.; Gangurde, P. Health Insurance-It’s Potential and Distribution Avenues in India.
J. Posit. Sch. Psychol. 2022, 6, 7148–7158.
11. Tien, C. Elements Influencing The Acceptance Of Digital Transformation in Health Insurance Management Of Students. Ir. J.
Educ. Pract. 2022, 5, 1.
12. R. a. markets, Global Health Insurance Market (2022–2027) by Plan Type, Duration, Application, Coverage, Type, Age Group,
Geography, Competitive Analysis and the Impact of COVID-19 with Ansoff Analysis. 2022. Available online: https://www.
researchandmarkets.com/reports/5669313/global-health-insurance-exchange-market-2022 (accessed on 12 July 2023).
13. Gee, J.; Button, M. The Financial Cost of Healthcare Fraud 2014: What Data from Around the World Shows. 2014. Available
online: https://pure.port.ac.uk/ws/portalfiles/portal/17778625/The_Financial_Cost_of_Healthcare_Fraud_Report_2014_11.
3.14a.pdf (accessed on 12 July 2023).
14. Amponsah, A.A.; Adebayo, F.A.; Weyori, B.A. Blockchain in Insurance: Exploratory Analysis of Prospects and Threats. Int. J. Adv.
Comput. Sci. Appl. 2021, 12, 1. [CrossRef]
15. Thaifur, A.Y.B.R.; Maidin, M.A.; Sidin, A.I.; Razak, A. How to detect healthcare fraud? A systematic review. Gac. Sanit.
2021, 35, S441–S449. [CrossRef]
16. Shi, Y.; Sun, C.; Li, Q.; Cui, L.; Yu, H.; Miao, C. A fraud resilient medical insurance claim system. In Proceedings of the Thirtieth
AAAI Conference on Artificial Intelligence, Phoenix, AZ, USA, 12–17 February 2016.
17. Kapadiya, K.; Patel, U.; Gupta, R.; Alshehri, M.D.; Tanwar, S.; Sharma, G.; Bokoro, P.N. Blockchain and AI-Empowered Healthcare
Insurance Fraud Detection: An Analysis, Architecture, and Future Prospects. IEEE Access 2022, 10, 79606–79627. [CrossRef]
18. Ai, J.; Russomanno, J.; Guigou, S.; Allan, R. A Systematic Review and Qualitative Assessment of Fraud Detection Methodologies
in Health Care. North Am. Actuar. J. 2021, 26, 1–26. [CrossRef]
19. Thesmar, D.; Sraer, D.; Pinheiro, L.; Dadson, N.; Veliche, R.; Greenberg, P. Combining the Power of Artificial Intelligence with the
Richness of Healthcare Claims Data: Opportunities and Challenges. Pharmacoeconomics 2019, 37, 745–752. [CrossRef]
20. Cao, H.; Zhang, R. Using PCA to improve the detection of medical insurance fraud in SOFM neural networks. In Proceedings of
the 2019 3rd International Conference on Management Engineering, Software Engineering and Service Sciences, Wuhan, China,
12–14 January 2019; pp. 117–122.
21. Shamitha, S.; Ilango, V. A time-efficient model for detecting fraudulent health insurance claims using Artificial neural net-
works. In Proceedings of the 2020 International Conference on System, Computation, Automation and Networking (ICSCAN),
Pondicherry, India, 3–4 July 2020; pp. 1–6.
22. Liu, W.; Yu, Q.; Li, Z.; Li, Z.; Su, Y.; Zhou, J. A blockchain-based system for anti-fraud of healthcare insurance. In Proceedings of
the 2019 IEEE 5th International Conference on Computer and Communications (ICCC), Chengdu, China, 6–9 December 2019;
pp. 1264–1268.
23. Mendoza-Tello, J.C.; Mendoza-Tello, T.; Mora, H. Blockchain as a Healthcare Insurance Fraud Detection Tool. In Springer
Proceedings in Complexity; Springer: Cham, Switzerland, 2021; pp. 545–552.
24. Ismail, L.; Zeadally, S. Healthcare Insurance Frauds: Taxonomy and Blockchain-Based Detection Framework (Block-HI). IT Prof.
2021, 23, 36–43. [CrossRef]
25. Humayun, M.; Jhanjhi, N.Z.; Niazi, M.; Amsaad, F.; Masood, I. Securing Drug Distribution Systems from Tampering Using
Blockchain. Electronics 2022, 11, 1195. [CrossRef]
26. Yeow, C.-N.; Tay, L.-Y.; Lye, C.-T.; Fauzi, E.P. Why do Malaysian Young Adults Buy Health Insurance? J. Int. Business Econ. Entrep.
2021, 6, 46. [CrossRef]
27. Brahmana, R.K.; Brahmana, R.K.M.; Memarista, G. Planned behaviour in purchasing health Insurance. South East Asian J.
Manag. SEAM 2018, 12, 53–64.
Healthcare 2023, 11, 2257 26 of 27

28. Bovbjerg, R.; Hadley, J. Why health insurance is important. In Health Policy Briefs; The Urban Institute: Washington, DC, USA, 2007.
29. Nayak, B.; Bhattacharyya, S.S.; Krishnamoorthy, B. Application of digital technologies in health insurance for social good of
bottom of pyramid customers in India. Int. J. Sociol. Soc. Policy 2019, 39, 752–772. [CrossRef]
30. Agustina, R.; Dartanto, T.; Sitompul, R.; Susiloretni, K.A.; Suparmi, M.K.M.; Achadi, E.L.; Taher, A.; Wirawan, F.; Sungkar, S.;
Sudarmono, P.; et al. Universal health coverage in Indonesia: Concept, progress, and challenges. Lancet 2019, 393, 75–102.
[CrossRef]
31. He, X.; Alqahtani, S.; Gamble, R. Toward privacy-assured health insurance claims. In Proceedings of the 2018 IEEE International
Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber,
Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData), Halifax, NS, Canada, 30 July–3 August 2018;
pp. 1634–1641.
32. Khezr, S.; Moniruzzaman, M.; Yassine, A.; Benlamri, R. Blockchain Technology in Healthcare: A Comprehensive Review and
Directions for Future Research. Appl. Sci. 2019, 9, 1736. [CrossRef]
33. Greco, L.; Percannella, G.; Ritrovato, P.; Tortorella, F.; Vento, M. Trends in IoT based solutions for health care: Moving AI to the
edge. Pattern Recognit. Lett. 2020, 135, 346–353. [CrossRef]
34. Peng, Y.; Kou, G.; Sabatka, A.; Chen, Z.; Khazanchi, D.; Shi, Y. Application of clustering methods to health insurance fraud
detection. In Proceedings of the 2006 International Conference on Service Systems and Service Management, Troyes, France,
25–27 October 2006; pp. 116–120.
35. Rawte, V.; Anuradha, G. Fraud detection in health insurance using data mining techniques. In Proceedings of the
2015 International Conference on Communication, Information & Computing Technology (ICCICT), Mumbai, India,
15–17 January 2015; pp. 1–5.
36. Verma, C.; Stoffova, V.; Illes, Z.; Tanwar, S.; Kumar, N. Machine Learning-Based Student’s Native Place Identification for Real-Time.
IEEE Access 2020, 8, 130840–130854. [CrossRef]
37. Settipalli, L.; Gangadharan, G.R. Healthcare fraud detection using primitive sub peer group analysis. Concurr. Comput. Pr. Exp.
2021, 33, e6275. [CrossRef]
38. Dua, P.; Bais, S. Supervised learning methods for fraud detection in healthcare insurance. In Machine Learning in Healthcare
Informatics; Springer: Berlin/Heidelberg, Germany, 2014; pp. 261–285.
39. Waghade, S.S.; Karandikar, A.M. A comprehensive study of healthcare fraud detection based on machine learning. Int. J. Appl.
Eng. Res. 2018, 13, 4175–4178.
40. Palacio, S.M. Abnormal Pattern Prediction: Detecting Fraudulent Insurance Property Claims with Semi-Supervised Machine-
Learning. Data Sci. J. 2019, 18, 35. [CrossRef]
41. Gajendran, N. Blockchain-Based secure framework for elearning during COVID-19. Indian J. Sci. Technol. 2020, 13, 1328–1341.
42. Nasir, M.H.; Arshad, J.; Khan, M.M.; Fatima, M.; Salah, K.; Jayaraman, R. Scalable blockchains—A systematic review.
Future Gener. Comput. Syst. 2022, 126, 136–162. [CrossRef]
43. Saranya, R.; Murugan, A. A systematic review of enabling blockchain in healthcare system: Analysis, current status, challenges
and future direction. Mater. Today Proc. 2021, 80, 3010–3015. [CrossRef]
44. Al Asad, N.; Elahi, M.T.; Al Hasan, A.; Yousuf, M.A. Permission-based blockchain with proof of authority for secured healthcare
data sharing. In Proceedings of the 2020 2nd International Conference on Advanced Information and Communication Technology
(ICAICT), Dhaka, Bangladesh, 28–29 November 2020; pp. 35–40.
45. Hölbl, M.; Kompara, M.; Kamišalić, A.; Nemec Zlatolas, L. A systematic review of the use of blockchain in healthcare. Symmetry
2018, 10, 470. [CrossRef]
46. McGhin, T.; Choo, K.-K.R.; Liu, C.Z.; He, D. Blockchain in healthcare applications: Research challenges and opportunities. J. Netw.
Comput. Appl. 2019, 135, 62–75. [CrossRef]
47. Javaid, M.; Haleem, A.; Singh, R.P.; Suman, R. 5G technology for healthcare: Features, serviceable pillars, and applications.
Intell. Pharm. 2023, 1, 2–10. [CrossRef]
48. Li, D. 5G and intelligence medicine—How the next generation of wireless technology will reconstruct healthcare? Precis. Clin.
Med. 2019, 2, 205–208. [CrossRef]
49. Nirjhor, M.K.I.; Yousuf, M.A.; Mhaboob, M.S. Electronic medical record data sharing through authentication and integrity
management. In Proceedings of the 2021 2nd International Conference on Robotics, Electrical and Signal Processing Techniques
(ICREST), DHAKA, Bangladesh, 5–7 January 2021; pp. 308–313.
50. Ahad, A.; Tahir, M.; Sheikh, M.A.; Ahmed, K.I.; Mughees, A.; Numani, A. Technologies Trend towards 5G Network for Smart
Health-Care Using IoT: A Review. Sensors 2020, 20, 4047. [CrossRef]
51. Chen, M.; Yang, J.; Zhou, J.; Hao, Y.; Zhang, J.; Youn, C.-H. 5G-Smart Diabetes: Toward Personalized Diabetes Diagnosis with
Healthcare Big Data Clouds. IEEE Commun. Mag. 2018, 56, 16–23. [CrossRef]
52. Hartmann, M.; Hashmi, U.S.; Imran, A. Edge computing in smart health care systems: Review, challenges, and research directions.
Trans. Emerg. Telecommun. Technol. 2022, 33, e3710. [CrossRef]
53. Amponsah, A.A.; Adekoya, A.F.; Weyori, B.A. A novel fraud detection and prevention method for healthcare claim processing
using machine learning and blockchain technology. Decis. Anal. J. 2022, 4, 10012. [CrossRef]
Healthcare 2023, 11, 2257 27 of 27

54. Gera, J.; Palakayala, A.R.; Rejeti, V.K.K.; Anusha, T. Blockchain technology for fraudulent practices in insurance claim process. In
Proceedings of the 2020 5th International Conference on Communication and Electronics Systems (ICCES), Coimbatore, India,
10–12 June 2020; pp. 1068–1075.
55. Goyal, A.; Elhence, A.; Chamola, V.; Sikdar, B. A blockchain and machine learning based framework for efficient health insurance
management. In Proceedings of the 19th ACM conference on embedded networked sensor systems, New York, NY, USA,
15–17 November 2021; pp. 511–515.
56. Amponsah, A.A.; Adekoya, A.F.; Weyori, B.A. Improving the Financial Security of National Health Insurance using Cloud-Based
Blockchain Technology Application. Int. J. Inf. Manag. Data Insights 2022, 2, 100081. [CrossRef]
57. Humayun, M.; Jhanjhi, N.Z.; Almotilag, A.; Almufareh, M.F. Agent-Based Medical Health Monitoring System. Sensors 2022, 22, 2820.
[CrossRef]
58. Ates, H.C.; Nguyen, P.Q.; Gonzalez-Macia, L.; Morales-Narváez, E.; Güder, F.; Collins, J.J.; Dincer, C. End-to-end design of
wearable sensors. Nat. Rev. Mater. 2022, 7, 887–907. [CrossRef]
59. Humayun, M.; Jhanjhi, N.Z.; Almotilag, A. Real-Time Security Health and Privacy Monitoring for Saudi Highways Using
Cutting-Edge Technologies. Appl. Sci. 2022, 12, 2177. [CrossRef]
60. Nayak, B.; Bhattacharyya, S.S. The Changing Narrative in the Health Insurance Industry: Wearables Technology in Health
Insurance Products and Services for the COVID-19 World. J. Health Manag. 2020, 22, 550–558. [CrossRef]
61. Gupta, R.; Thakker, U.; Tanwar, S.; Obaidat, M.S.; Hsiao, K.-F. Bits: A blockchain-driven intelligent scheme for telesurgery
system. In Proceedings of the 2020 International Conference on Computer, Information and Telecommunication Systems (CITS),
Hangzhou, China, 5–7 October 2020; pp. 1–5.
62. McCrea, M.; Farrell, M. A Conceptual Model for Pricing Health and Life Insurance Using Wearable Technology. Risk Manag.
Insur. Rev. 2018, 21, 389–411. [CrossRef]
63. Humayun, M.; Jhanjhi, N.; Alamri, M. Smart secure and energy efficient scheme for e-health applications using IoT: A review. Int.
J. Comput. Sci. Netw. Secur. 2020, 20, 55–74.
64. Akande, A. Disruptive Power of Blockchain on the Insurance Industry. Master’s Thesis, Universiti of Tartu, Institute of Computer
Science, Tartu, Estonia, 2018.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.