Cite (APA): Aydın, Ö., Karaarslan, E. (2022).

OpenAI ChatGPT Generated Literature Review: Digital Twin in

Healthcare . In Ö. Aydın (Ed.), Emerging Computer Technologies 2 (pp. 22-31). İzmir Akademi Dernegi.

OpenAI ChatGPT Generated Literature Review:

Digital Twin in Healthcare
Ömer AYDIN Enis Karaarslan
Manisa Celal Bayar University Mugla Sitki Kocman University
Faculty of Engineering Faculty of Engineering
Electrical and Electronics Engineering Computer Engineering
Manisa, Türkiye Mugla, Türkiye
omer.aydin@cbu.edu.tr enis.karaarslan@mu.edu.tr,
0000-0002-7137-4881 0000-0002-3595-8783
Abstract— Literature review articles are essential to academic writing skills. People can improve the quality of
summarize the related work in the selected field. However, their academic writing and effectively communicate their
covering all related studies takes too much time and effort. This ideas to readers by following the following few tips:
study questions how Artificial Intelligence can be used in this
process. We used ChatGPT to create a literature review article • Understand the purpose of people’s writing: Is it to
to show the stage of the OpenAI ChatGPT artificial intelligence inform, persuade, or analyse? Identifying the goal will help
application. As the subject, the applications of Digital Twin in them choose the appropriate tone, style, and structure for their
the health field were chosen. Abstracts of the last three years writing.
(2020, 2021 and 2022) papers were obtained from the keyword
"Digital twin in healthcare" search results on Google Scholar • Use clear and concise language: Avoid jargon and
and paraphrased by ChatGPT. Later on, we asked ChatGPT use simple, straightforward language to convey your ideas
questions. The results are promising; however, the paraphrased effectively.
parts had significant matches when checked with the Ithenticate
tool. This article is the first attempt to show the compilation and • Use evidence to support their claims: It is essential to
expression of knowledge will be accelerated with the help of provide evidence to support your ideas and arguments in
artificial intelligence. We are still at the beginning of such academic writing. Use credible sources and provide proper
advances. The future academic publishing process will require citations to support their claims.
less human effort, which in turn will allow academics to focus
on their studies. In future studies, we will monitor citations to • Use proper structure: Organise their writing into
this study to evaluate the academic validity of the content clear sections and use headings and subheadings to guide the
produced by the ChatGPT. reader through their argument.
Keywords— OpenAI, ChatGPT, Artificial Intelligence, Digital • Edit and proofread: Make sure to proofread their
twin, Healthcare, ChatGPT revolution, Academic Publishing writing for grammar and spelling errors and edit for clarity and
concision.
I. INTRODUCTION
Artificial intelligence (AI) is a technological development
OpenAI ChatGPT [1] is a chatbot based on the OpenAI
that will significantly impact daily life and legal regulations,
GPT-3 language model. It is designed to generate human-like
including civil, criminal, and liability laws. One area where AI
text responses to user input in a conversational context.
will affect is the regulation of intellectual property rights.
OpenAI ChatGPT is trained on a large dataset of human
There is a question of who should own patents for inventions
conversations and can be used to create responses to a wide
made by AI machines [2]. Who owns the AI- generated
range of topics and prompts. The chatbot can be used for
articles? Is it the person who asks the information and
customer service, content creation, and language translation
questions necessary to write the essay, or is it the software and
tasks, creating replies in multiple languages. OpenAI
machine that creates this content? These and many other
ChatGPT is available through the OpenAI API, which allows
questions need to be answered.
developers to access and integrate the chatbot into their
applications and systems. In this study, it is evaluated whether an academic article
can be created by AI and whether the created article carries
OpenAI ChatGPT is a variant of the GPT (Generative Pre-
values such as academic content and originality. All content
trained Transformer) language model developed by OpenAI.
except the “Abstract”, “Introduction”, “Method”, “Findings
It is designed to generate human-like text, allowing it to
and Results” and “Discussion and Conclusion” subsections
engage in conversation with users naturally and intuitively.
was generated by ChatGPT. Another purpose of this study is
OpenAI ChatGPT is trained on a large dataset of human
to analyse the number of citations this article will receive and
conversations, allowing it to understand and respond to a wide
the characteristics of the citations. When the required sample
range of topics and contexts. It can be used in various
size for the citations is reached (at least 100 journal article
applications, such as chatbots, customer service agents, and
citations), we will examine the studies referring to this article
language translation systems. OpenAI ChatGPT is a state-of-
and the effectiveness, and academic validity of the paper will
the-art language model able to generate coherent and natural
be interpreted through citations.
text that can be indistinguishable from text written by a
human. In the first following section, used methods are given. In
the second section one of the new technologies Digital Twin
As an artificial intelligence, ChatGPT may need help to
was defined by ChatGPT. In the third section, information
change academic writing practices. However, it can provide
about the use and features of Digital Twin in HealthCare is
information and guidance on ways to improve people’s

22

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
shared by ChatGPT. In the fourth part, the abstracts of the In manufacturing, we can use digital twins to optimize
articles that were found as a result of the search for "Digital production, reduce downtime, and improve efficiency. For
Twin in Healthcare" and that could be accessed via Google example, a digital twin of a factory floor can be used to
Scholar belonging the last 3 years were paraphrased by simulate and optimize the flow of materials, identify
ChatGPT. In the fifth section, the findings and results are bottlenecks, and minimize waste.
given. Finally, the study ends with the evaluations made by us
In healthcare, we can use digital twins to simulate and
under the title of Discussion and Conclusion.
optimize patient care pathways, reducing the risk of medical
II. METHOD errors and improving outcomes. We can also use digital twins
to optimize the design and operation of healthcare facilities,
It is useful to reiterate the aim of doing this study here.
such as hospitals and clinics.
ChatGPT is an important example of how artificial
intelligence can produce results in academic article writing. In urban planning, we can use digital twins to simulate and
Therefore, we aim to ensure that a review article is written by optimize the design and operation of entire cities, including
artificial intelligence at a basic level and to question its transportation systems, energy networks, and public services.
academic validity. For this purpose, as a research method, it These optimizations can help cities to become more efficient,
was determined that the abstracts of the results obtained by sustainable, and liveable.
searching on a topic (Digital Twin in Healthcare) that we
Overall, digital twins have the potential to transform a wide
determined on Google Scholar should be paraphrased with
range of industries and applications by providing real-time
ChatGPT. Also, "What is Digital Twin?" and "Digital Twin in
data and insights that can be used to optimize the performance
Healthcare" was queried to ChatGPT to generate for the
and behaviour of physical objects and systems. As technology
Digital Twin and Healthcare topics. These two titles are
advances, digital twins will likely become increasingly
included in the article before the literature title.
prevalent in many sectors.
The texts created by ChatGPT by paraphrasing and the
IV. DIGITAL TWIN IN HEALTHCARE
texts produced by giving the answers are checked by
plagiarism tools such as Ithenticate, intihal.net and alike. As a Digital twins have the potential to revolutionize healthcare
result of the evaluation, an evaluation is made on the similarity by providing real-time data and insights that can be used to
rate with the previous studies. optimize patient care pathways and improve outcomes.
Several key themes emerge in a review of the literature on
In order to evaluate the Academic Validity of this study,
digital twins in healthcare.
the following method was determined. This article has been
published on the preprint sharing environments (SSRN, One of the main benefits of digital twins in healthcare is
Arxiv, Techxriv, Researchgate, Academia etc.). Citations to their ability to provide real-time data and insights into patient
this study will be recorded then. Finally, these articles will be care. By simulating and optimizing patient care pathways,
reviewed and evaluated when the citations reach a certain digital twins can help to reduce the risk of medical errors and
number (for example, 100 journal article citations). The improve patient outcomes. For example, a digital twin of a
academic validity of this text produced by ChatGPT will be patient's medical history and current condition can be used to
evaluated from the criticisms and evaluations in these identify potential problems and inefficiencies in the care
citations. process, enabling proactive intervention and optimization of
care.
III. DIGITAL TWIN
Another key benefit of digital twins in healthcare is their
A digital twin is a digital representation of a physical
ability to support remote monitoring and telemedicine. By
object or system that allows for the simulation of the object or
providing real-time data on patient vital signs and other health
system's behaviour, performance, and characteristics. This
indicators, digital twins can enable healthcare providers to
technology has the potential to revolutionize a wide range of
monitor patients remotely and provide care remotely, reducing
industries, from manufacturing and logistics to healthcare and
the need for in-person visits and improving access to care.
urban planning.
We can also use digital twins to optimize the design and
The concept of a digital twin originated in the
operation of healthcare facilities, such as hospitals and clinics.
manufacturing industry, where it was used to optimize the
By simulating and optimizing the flow of patients, staff, and
design and operation of complex systems such as aircraft
resources within a facility, digital twins can help to improve
engines and industrial machinery. Today, digital twins are
efficiency, reduce wait times, and improve patient
being applied to a much more comprehensive range of
satisfaction.
applications, including infrastructure, buildings, and even
entire cities. Overall, using digital twins in healthcare can transform the
way healthcare is delivered, enabling more personalized,
One of the key benefits of digital twins is their ability to
efficient, and effective care. As technology advances, digital
provide real-time data and insights into the performance and
twins will likely become increasingly prevalent in healthcare,
behaviour of a physical object or system. These insights allow
providing valuable data and insights that can be used to
for the identification of problems and inefficiencies in real-
optimize patient care and improve outcomes.
time, enabling proactive maintenance and optimization of the
system.

23

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
V. LITERATURE REVIEW recognition to mimic human reasoning. These core
technologies can also assist in decision-making for a patient
There has been a significant increase in the use of digital
by synthesizing data from multiple real-world sources within
healthcare services, such as personalized medicine,
a virtual domain. Digital twins can be created at various levels,
telemedicine, and smart diet management, in recent years.
including the whole body, a single organ, or the cellular level,
These services often have diverse requirements, and require
and can be used to study specific conditions or settings, such
intelligent analytics and self-sustaining networks in order to
as cancer or critical care. An individual's digital twin can be
function effectively. Digital twins can help to enable
copied and tested with different treatments to determine the
proactive, intelligent analytics and self-sustaining networks in
optimal intervention. By aggregating digital twins from many
healthcare, allowing for the efficient management of resources
individuals with diverse genetic, biomedical, and lifestyle
and minimum interaction from end-users or network
profiles, a population-based database can be created, which
operators. In this article, we provide an overview of digital
can be used for data-driven prevention and patient matching
twins in healthcare, including a proposed architecture and
for clinical trials conducted in silico as an alternative to
several use cases. We also identify open research challenges
lengthy or costly animal and human studies [4].
and potential solutions.
In a paper by Volkow et al., it is noted that as the
The proposed digital twin-based system for monitoring
population grows, the need for high-quality medical services
COVID-19 symptoms uses mobile phones to collect data,
and the use of information technology in medicine also
which is fed into a digital copy of each individual stored in the
increases. The concept of Smart Healthcare aims to address
cloud. This digital twin is continually updated with real-time
the challenges faced in modern healthcare through various
measurements from sensors and other devices, and is analyzed
approaches, including the use of digital twins, the Internet of
using machine learning and artificial intelligence techniques
Things, and mobile medicine. The paper reviews the main
to determine the individual's health status. The system also
problems in modern healthcare and analyses existing
utilizes blockchain technology to ensure the confidentiality of
approaches and technologies in these areas. It also discusses
the data, which can be shared for research purposes with the
the key features of modern platforms that support Mobile
consent of the individual. The goal of the system is to provide
Health Applications. Based on this analysis, the authors
early diagnosis and prevent the spread of the disease by
propose the concept of a Smart Healthcare Platform, focused
establishing a protocol for following up with patients and
on supporting the development of Mobile Health Applications
those at risk [3].
through the organization, management, and sharing of user
Cardiovascular disease (CVD) is a leading cause of data [5].
morbidity and mortality from non- communicable diseases
While Digital Twin technology has been applied in a
worldwide. In many cases, effective acute care and risk factor
variety of fields, including advanced manufacturing, Product
management strategies based on data from epidemiological
Lifecycle Management (PLM), and smart healthcare, this
studies and clinical trials have improved survival and quality
paper discusses the potential use of Digital Twin for full
of life. However, approximately half of all myocardial
lifecycle management of humans. The concept of an
infarction and stroke events occur in patients who do not meet
Augmented Digital Twin is introduced as the foundation for
the criteria for cholesterol-lowering treatment, and costly
the concept of a Human Digital Twin (HDT), which is the
preventive therapies with potential side effects may be given
central focus of the paper. Drawing on the experience of using
unnecessarily. Improving outcomes in cardiology therefore
Digital Twin in smart manufacturing, PLM, and smart
requires a better understanding of individual susceptibility and
healthcare, as well as the development of related technologies
prognosis [4].
such as data mining, data fusion analysis, artificial
Data-driven approaches to risk assessment and treatment intelligence, particularly deep learning, and human-computer
are central to the goals of precision medicine, which aims to science, it is concluded that HDT could be a useful approach
disrupt traditional healthcare practices by providing early, to full lifecycle health management and that it is possible to
customized therapy that considers molecular, genetic, construct a Human Digital Twin, particularly from a
physical, and environmental data profiles. While prediction technological standpoint. A comparison between Digital Twin
and personalization have always been a part of medicine, the and Human Digital Twin demonstrates this possibility. The
convergence of 21st century technologies with advances in paper presents the concept, conceptual model, and
biomedicine is enabling the consolidation and analysis of characteristics of HDT to lay the groundwork for its
large amounts of data from a wide range of sources using construction. It also discusses how to construct a Human
massive data networks. Improvements in clinical data capture, Digital Twin by proposing a Human Digital Twin System
imaging, and molecular phenotyping also contribute to the Architecture and Implementation Approach. However, it is
goal of precision medicine. The ability to process large data noted that there are many challenges to overcome in this
sets digitally has greatly enhanced the utility of this data [4]. process, including extreme complexity and security and social
ethics issues [6].
The health digital twin builds on these capabilities by
using a virtual replica to receive a patient's molecular, Personalized medicine involves using detailed information
physiological, and lifestyle data, allowing for risk prediction about individual people to identify deviations from the
and personalized treatment for the real-life patient. Artificial normal. Digital twins in engineering provide a framework for
intelligence helps link the cyber, physical, and data domains analysing these data-driven healthcare practices and their
to create virtual models by using technologies such as cloud ethical and societal implications for therapy, preventative
computing, big data processing, machine learning, and pattern care, and human enhancement. Digital twins are pairs of

24

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
individual physical artifacts and digital models that demand healthcare. Typically, this large amount of data is
dynamically reflect the status of the artifacts. Ethical and analyzed on centralized servers. However, anomaly detection
societal implications of digital twins are examined, and it is (AD) in centralized healthcare systems can be slow and
noted that they have the potential to deliver significant societal resource-intensive, and there are also privacy concerns related
benefits and function as a social equalizer by allowing for to sending patients' personal health data to a centralized
effective, equalizing enhancement interventions. Digital twins server, which can also introduce security risks such as the
will be important for providing highly personalized treatments possibility of data poisoning. To address these issues with
and interventions, and they rely on explainable artificial centralized AD models, the authors propose a federated
intelligence [7]. learning (FL)-based AD model that uses edge cloudlets to run
AD models locally without sharing patients' data. While
Anticipating the ethical impact of emerging technologies
existing FL approaches perform aggregation on a single
is an important aspect of responsible innovation, and the
server, which limits their scope, they introduce a hierarchical
digital twin is one such technology that is emerging. A digital
FL approach that enables aggregation at different levels and
twin is a living replica of a physical system (human or non-
multi-party collaboration. They also propose a novel disease-
human) that combines various emerging technologies such as
based grouping mechanism for organizing different AD
artificial intelligence, the Internet of Things, big data, and
models based on specific types of diseases. In addition, they
robotics, each of which brings its own ethical issues. This
develop a new Federated Time Distributed (FEDTIMEDIS)
report presents the results of a qualitative study on the ethical
Long Short-Term Memory (LSTM) approach for training the
benefits and risks of using digital twins in healthcare. The
AD model. They present a remote patient monitoring (RPM)
study used insights from the ethics of technology and the
use case to demonstrate our model and provide a proof-of-
Quadruple Helix theory of innovation, and included desk
concept implementation using Digital Twin (DT) and edge
research of white literature and 23 interviews with
cloudlets [10].
representatives from industry, research, policy, and civil
society. The results showed that digital twins have the As the proportion of the aging population increases, the
potential to produce ethical benefits in areas such as the need for sustainable, high-quality, and timely healthcare
prevention and treatment of disease, cost reduction, patient services has become increasingly pressing, particularly in
autonomy and freedom, and equal treatment, but also have light of the COVID-19 pandemic in 2020. To meet this
ethical risks in areas such as privacy and data property, demand, a promising strategy is to introduce cloud computing
disruption of existing societal structures, inequality, and and digital twin techniques into healthcare systems. In this
injustice. The report concludes with a reflection on the approach, a cloud server is used to store healthcare data and
analytical tool used and suggestions for further research [8]. provide efficient query services, while a digital twin is used to
build a digital representation of a patient and leverage the
In a paper by Elayan et al., it is noted that the use of digital
query services of the cloud server to monitor the healthcare
technologies in the healthcare industry, known as digital and
state of the patient. While several cloud computing and digital
smart healthcare, has increased in recent years in an effort to
twin- based healthcare monitoring frameworks have been
improve patient care, increase life expectancy, and reduce
proposed, none have addressed data privacy concerns, even
healthcare costs. One such technology that has the potential to
though the leakage of private healthcare information could
significantly impact the field is digital twin (DT) technology.
have catastrophic consequences for patients. To address this
DT is a virtual replica of a physical asset that reflects its
challenge, this paper proposes an efficient and privacy-
current status through real- time data. The paper proposes and
preserving similarity query-based healthcare monitoring
implements an intelligent context-aware healthcare system
scheme over a digital twin cloud platform, called PSim-DTH.
using DT technology. As part of this system, an
The scheme employs a partition-based tree to index the
electrocardiogram (ECG) heart rhythms classifier model was
healthcare data and uses matrix encryption to propose a
built using machine learning to diagnose heart disease and
privacy-preserving PB-tree-based similarity range query
detect heart problems. The results showed that integrating DT
algorithm. Security analysis and performance evaluation are
with healthcare can improve the healthcare process by
conducted, and the results show that the proposed PSim-DTH
bringing patients and healthcare professionals together in an
scheme is both privacy-preserving and efficient [11].
intelligent, comprehensive, and scalable health ecosystem.
Additionally, the implementation of the ECG classifier In the future, it is possible that a digital twin of each
demonstrated the potential for using machine learning and person's genetic profile will be created at birth. If someone
artificial intelligence with various human body metrics for becomes ill, this virtual representation of themselves could be
continuous monitoring and abnormalities detection. The paper treated with a vast number of drugs using computational
also found that neural network-based algorithms perform methods to determine the most effective medication. The
better on ECG data than traditional machine learning digital twin could also be used to predict which medical
algorithms [9]. conditions a person may be at risk for, allowing them to take
preventive measures to avoid major diseases before they occur
The Internet of Medical Things (IoMT) is becoming
[12].
increasingly prevalent, with the widespread use of smart
medical devices and applications in smart hospitals, home The process of digitization and digitalization has already
healthcare, and nursing homes. It combines smart medical had a significant impact on our world, and further disruptions
devices, cloud computing services, and core Internet of Things are expected as digital transformation continues, with digital
(IoT) technologies to monitor patients' vital signs, track health twins playing a key role. As big data techniques, the Internet
conditions, and generate multivariate data to support on- of Things, cloud computing, and artificial intelligence

25

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
algorithms advance, digital twin technology is experiencing One of the thesis investigates the potential use of digital
rapid development and has been identified as one of the top twins in the healthcare industry as a means of predicting and
ten most promising technologies. Although it is still in its early preventing diseases. It examines the values and perspectives
stages, digital twins are already being widely used in various of individuals towards the use of digital twins in diagnosis,
fields, particularly in industry, smart cities, and smart health, and proposes different forms or concepts that a digital twin
which are the areas that attract the most research. There are could take in order to be meaningful to the user. The research
numerous articles and reviews on digital twins published found that the meaningfulness of a digital twin may vary
every year in these three fields, making it timely and necessary significantly based on factors such as age, life situation,
to provide an analysis of the published work. This article aims preferences, and individual personality. It also found that the
to do just that, providing a survey of the major research and time, place, and frequency at which an individual interacts
application areas of digital twins. It begins by analyzing the with a digital twin may be important considerations. The
recent developments of digital twins, then summarizes the thesis concludes with guidelines for developing a more user-
theoretical foundations of the technology, and concludes with centered and meaningful digital twin that takes into account
specific developments in various application areas of digital the varied and dynamic nature of user values [17].
twins. The article also discusses the challenges that may be
BOATĂ and other's paper summarizes the progress made
encountered in the future [13].
in using the concept of digital twins to select the most
Digital Twins (DTs) are becoming more widespread in appropriate medication for an individual, based on genetic
various industries due to advances in computing power and expressions related to a particular disease and the genes
data science. There are many sensors and connected devices targeted by therapy. The paper presents a subnet model that
that generate data, which can be used to create a digital model predicts existing drugs for a specific pathology. Finally, the
of a system or object and predict and simulate various paper discusses how the proposed method could be used to
scenarios. DTs have recently been adapted for use in the search for a medication for the disease caused by the Covid-
healthcare industry for accurate medical simulations and 19 virus, referencing the results of clinical trials conducted
resource management, but the technology is still relatively after the start of the pandemic [18].
new to the healthcare system and there are concerns about
Digital twin technology, which creates virtual
security threats. This paper proposes a Blockchain-based
representations of people, products, or buildings, has the
secure Digital Twin framework for a smart healthy city as a
potential to revolutionize the healthcare industry by
solution to these threats. The paper also discusses the current
improving the diagnosis and treatment of patients,
COVID-19 pandemic as a case study and suggests that DTs
streamlining preventative care, and facilitating new
can be used to control the situation, prevent future cases, and
approaches for hospital planning. In the future, it could be
personalize treatment [14].
used to create a digital twin of each person's genetic profile at
De Maeyer and Markopoulos study how experts in the birth, allowing healthcare professionals to computationally
digital transformation of healthcare envision the use of Digital test the most effective medications for an individual and
Twins. Digital Twins are digital replicas of physical assets, predict their risk for certain medical conditions. This could
processes, people, places, systems, and devices that can be enable more personalized and preventative healthcare
used for scientific experiments, simulations, and predictions approaches [19].
of intervention outcomes. They are a new technology with
Madubuike and Anumba’s research aims to examine how
different implications and possibilities for different
digital twin technology can be used to improve healthcare
applications, and may evolve in various ways. The authors
facilities management. Currently, healthcare facilities
used the Delphi method to reach a consensus among experts
management often relies on a reactive approach, where issues
on three research questions related to the materialization,
are only addressed when they arise. Digital twins, which are
expectations, and implementation of Digital Twins in
virtual representations of physical assets created using real-
healthcare. The main conclusion is that Digital Twins are seen
time data, can be used to monitor and manage critical
as a way to enable preventive healthcare and trial-and-error
healthcare facilities in a more preventive way. The research
approaches to support personalized medicine and patient-
uses literature review and interviews to address its objectives
centered care [15].
and identify issues with critical healthcare facilities that can
De Maeyer and Markopoulos explore how experts in the be addressed using digital twin technology. It also reviews the
digital transformation of healthcare envision the use of digital technologies that enable digital twins and presents a system
twins (DTs), which are digital replicas of physical assets, architecture for their use in healthcare facilities management.
processes, people, places, systems, and devices that can be The research concludes by discussing the potential benefits of
used for scientific experiments, simulations, and prediction of using digital twins in healthcare facilities management [20].
intervention outcomes. DTs are a new and evolving
Voigt et al. discuss the potential for digital twins to
technology, and can be applied in different ways and for
improve the management of multiple sclerosis (MS) by
different purposes. To understand how DTs might develop
providing an individualized, innovative approach to disease
and impact healthcare, the authors conducted a Delphi study
management. They propose that digital twins, which are
and survey with experts to reach a consensus on how DTs
virtual representations of a patient's characteristics created
might be materialized, what expectations there are for their
using artificial intelligence-based analysis of various disease
use, and how they might be implemented in healthcare. The
parameters, could help healthcare professionals handle large
study aimed to identify the top ranked topics for further
amounts of patient data and provide more personalized,
consideration [16].
effective care by integrating data from multiple sources,

26

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
implementing individualized clinical pathways, and that need to be met through proactive intelligent analytics and
facilitating communication and shared decision-making self-sustaining networks. Proactive intelligent analytics
between patients and physicians. The authors also discuss the allows for the efficient management of resources in response
potential benefits of using digital twins in MS management, to user requests, while self-sustaining networks operate with
including improved diagnosis, monitoring, and therapy, minimal interaction from end-users or network operators. To
enhanced patient well- being, reduced economic costs, and enable Healthcare 4.0 with these capabilities, digital twins can
prevention of disease progression. They argue that digital be utilized. The paper provides an overview and recent
twins can help make precision medicine and patient-centered advancements in the use of digital twins for Healthcare 4.0,
care a reality in everyday practice [21]. presents an architecture for digital twins in healthcare, and
discusses several use cases. Additionally, the paper identifies
A digital twin is a virtual representation of a real-life object
open research challenges and possible solutions [25].
or system that is connected to its physical counterpart in a way
that allows for real-time data exchange and analysis. Digital In recent years, Digital Twins technology has been
twins are being used in various industries, and when applied introduced into the healthcare industry, bringing significant
to the fields of healthcare and public health, they have the changes to the way medical care is provided. In this study, the
potential to revolutionize traditional electronic health records authors propose a new disease diagnosis algorithm called the
by making them more precise and accurate. This technology Factorization Machine Combine Product- based Neural
allows for learning, hypothesis testing, and in silico Network (FMCPNN), which is an improved version of the
experiments, and has the potential to enable highly Product-based Neural Network (PNN). PNN is an end-to-end
personalized treatments and interventions in the future. This Factorization Machine algorithm that is effective at handling
paper discusses the history and main concepts of digital twins, sparse data, but it lacks the ability to handle low-order feature
provides examples of their use in personal medicine, public interactions, which can limit its generalization capabilities.
health, and smart cities, and explores the technical and ethical FMCPNN adds a second-order interaction component to
challenges involved in their application [22]. PNN, improving its performance. FMCPNN can be
effectively used in Digital Twin medical systems to improve
Hussain et al. aims to demonstrate the use of a digital twin
the accuracy and speed of disease diagnosis. Their tests have
in healthcare, specifically for stroke patients, using EEG data
shown that FMCPNN outperforms some other advanced
and machine learning techniques. The authors collected EEG
models [26].
data from 48 stroke patients and 75 healthy individuals using
portable EEG devices and analyzed the data to identify Human digital twin (HDT) is a technology that has the
characteristics that could be used to distinguish between the potential to revolutionize the healthcare system by enabling
two groups. They found that certain brain activity personalized healthcare services (PHS) through the use of
measurements were useful for classifying stroke patients and tools such as artificial intelligence and blockchain. Its
healthy individuals in different mental states, and used a implementation is expected to be similar to digital twins in
machine learning model (SVM) to classify the data with 76% other sectors, such as manufacturing and aviation, and will
accuracy. The authors suggest that this digital twin framework consist of three key dimensions: the physical entity, virtual
could be used to help make clinical decisions related to stroke model, and connection that characterizes the physical- virtual
prevention and post-stroke treatment [23]. interactions. However, the complexity of the human body,
with its constant molecular and physiological changes, makes
Healthcare technologies have seen increased use during
it difficult to extract precise medical data and model HDTs.
the COVID-19 pandemic, including remote patient care and
As a result, HDT is more complex than other types of digital
virtual follow-up. These technologies are likely to continue to
twins and the methods for implementing it are not yet clear,
be adapted in the future, both as a way to prepare for future
which requires further investigation. This paper presents the
pandemics and due to the advancement of artificial
architectural framework and key design requirements for
intelligence. This study looks at the potential healthcare
HDT, discusses the key technologies and technical challenges,
applications of digital twin technology, which involves
and suggests future directions for research. It is believed that
creating a virtual counterpart to a physical model and studying
this paper will stimulate new research efforts towards the
the relationship between the two. It is proposed that digital
development of HDT for PHS [27].
twin technology could be used to analyze the relationship
between a physical cancer patient and their corresponding Emotion recognition (ER) in healthcare has gained
digital twin in order to predict and prevent neurological significant attention due to recent advances in machine-
complications related to the disease. This approach is based learning and deep-learning techniques. ER systems, combined
on the idea that data science can be used to inform the with digital twins of individuals in real-time, can facilitate the
diagnosis, treatment planning, and prognosis of cancer care. monitoring, understanding, and improvement of physical
Digital twin technology could be used in precision medicine, capabilities, as well as provide constant input to improve
cancer care and treatment modeling, predictive analytics and quality of life and well-being for personalized healthcare.
machine learning, and to bring together different perspectives However, building such ER systems in real-time presents
from clinicians [24]. technical challenges such as limited datasets, occlusion and
lighting issues, identifying important features, false
In a paper by Alazab et al., it is noted that the landscape of
classification of emotions, and high implementation costs. To
digital healthcare, also known as Healthcare 4.0, has seen a
address these challenges, a simple, efficient, and adaptable ER
significant increase in services such as personalized
system was developed that acquires and processes images in
healthcare, intelligent rehabilitation, telemedicine, and smart
real-time using a web camera. An end-to-end framework was
diet management. These services have diverse requirements

27

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
also proposed that combines an ER system with a digital twin intelligent, and proactive healthcare. With the evolution of
setup, allowing for the prediction and testing of results prior personalized healthcare, there is a need to create a virtual
to providing the best possible personal healthcare treatment. replica for individuals to provide the right type of care in the
The proposed ER system achieved promising results in a right way and at the right time. In this paper, the concept of a
shorter training time without sacrificing accuracy. It is personal digital twin (PDT) as an enhanced version of a DT
expected that in real-time, the system will be useful in with actionable insight capabilities is examined. PDT can
healthcare centres for monitoring patients' health conditions, bring value to patients by enabling more accurate decision-
early diagnosis of life-threatening diseases, and obtaining the making and proper treatment selection and optimization. The
most effective treatment for patients during emergencies [28]. progression of PDT as a revolutionary technology in
healthcare research and industry is also explored. However,
The digital world has seen significant growth in recent
although there have been numerous research studies on the use
decades, and the COVID-19 pandemic has further accelerated
of DTs in smart healthcare, PDT is still in its early stages.
this trend. Digital transformations are affecting all aspects of
Therefore, this work aims to guide the design of industrial
our lives, and new technologies are constantly emerging and
personalized healthcare systems by introducing a reference
attracting attention and investment. Digital twin technology,
framework that brings together existing advanced
which has gained significant popularity in recent years, is now
technologies such as DTs, blockchain, and artificial
being applied in the healthcare sector, which has been under
intelligence to enable smart personalized healthcare using
scrutiny due to the COVID-19 pandemic. This paper aims to
PDTs. Several use cases are described, including the
provide a better understanding of digital twin technology,
mitigation of COVID-19 contagion, follow-up care for
address common misconceptions, and review the current state
COVID-19 survivors, personalized COVID-19 medicine,
of digital twin applications in healthcare. It also summarizes
personalized osteoporosis prevention, personalized cancer
the functions of digital twins at different stages of their use in
survivor follow-up care, and personalized nutrition. Finally,
the context of digital twin technology in healthcare. The paper
the challenges that need to be addressed to move the PDT
proposes a model for using digital twin technology in
paradigm towards the smart personalized healthcare industry
healthcare based on the Internet of Things as a service concept
are identified [31].
and digital twinning as a service model supporting Industry
4.0. It also identifies different groups of physical entities that Digital twins, virtual representations that are essentially a
can be represented by digital twins in healthcare. The research real-time digital version of a physical object or process, were
discusses the value of digital twin technology in healthcare, as first developed by NASA in 2010 to improve the simulation
well as the current challenges and future research directions of spacecraft physical models. With the growth of the Internet
[29]. of Things (IoT), digital twins have become more cost-
effective and have been successfully used in the
The concept of digital twins has the potential to transform
manufacturing and service sectors to transform these
the current healthcare system by making it more personalized.
industries. Given the benefits that digital twins have brought
As a convergence of healthcare, artificial intelligence, and
to these sectors, it is not surprising that they could potentially
information and communication technologies, personalized
be used to bring similar benefits to the healthcare industry.
healthcare services developed under the concept of digital
This chapter discusses how and why digital twins could be
twins raise a range of ethical issues. While some of these
conceptualized for healthcare [32].
ethical issues are known to researchers in the field of digital
health and personalized medicine, there is currently no Digital Twin (DT) is a technology that creates a digital
comprehensive review that examines the major ethical risks of replica of any physical phenomenon in a digital space, in sync
digital twins for personalized healthcare services. This study with the physical state. However, creating a Healthcare DT
aims to address this gap by identifying the major ethical risks model for patient care is seen as a challenging task due to the
of digital twins for personalized healthcare services. A lack of a structured data collection system. There are also
working definition of digital twins for personalized healthcare security and privacy concerns, as healthcare data is sensitive
services is proposed, and a process-oriented ethical map is and can be misused. Given these research gaps, it is important
developed to identify the major ethical risks in each stage of to find a way to properly acquire structured data and manage
data processing. A literature review of eHealth, personalized it securely. In this article, the authors present a mathematical
medicine, precision medicine, and information engineering is data model to collect patient-relevant data in a structured and
conducted to identify potential issues. The ethical map allows predefined way with clear boundaries. They also describe the
for the systematic identification of ethical concerns that data model in the context of real-life scenarios. Then, they use
emerge during the transformation of raw data into valuable the patient-centric mathematical data model to formally define
information. Developers of digital twins for personalized the semantics and scope of our proposed Healthcare Digital
healthcare services can use this map to proactively identify Twin (HDT) system based on blockchain technology. They
and address ethical risks during the development stage. This describe the proposed system with all of its key components,
paper provides a working definition of digital twins for as well as the detailed protocol flows and an analysis of its
personalized healthcare services, identifies 10 major various aspects. Finally, they provide a feasibility analysis of
operational problems and the corresponding ethical risks, and the proposed model and compare it with other relevant
suggests a process-oriented ethical map to assist in the research works [33].
analysis of ethical risks [30].
Ferdousi et al. provide an overview of well-being digital
Digital twins (DTs) are playing a vital role in transforming twins (WDT) in the healthcare industry, discussing their
the healthcare industry by enabling more personalized, potential scope, architecture, and impact. They also discuss

28

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
the definition and benefits of WDT, the evolution of digital city applications due to concerns about trust and privacy in
twin frameworks, and the challenges, types, drawbacks, and data sharing. Federated learning (FL), a technology that
potential application areas of WDT. Finally, the authors preserves privacy and trustworthiness, could be used in
present the requirements for a WDT framework as identified conjunction with digital twins to improve adoption in real-
in the literature [34]. time and life-critical scenarios and facilitate governance in
smart city applications. This paper presents a survey of the
This study aims to examine the potential and challenges of
various smart city applications of FL models in digital twins,
using digital twins, virtual representations of physical entities,
identifies challenges and future directions for improving FL-
to manage and predict outbreaks of Covid-19. By providing a
digital twin integration in future applications [38].
detailed model of each patient, digital twins can be used to
determine the most effective method of care for that Designing, constructing, and operating healthcare
individual, which can improve patient experience and care facilities is a complex process in the
delivery and reduce demand on healthcare services. The architectural/engineering/construction (AEC) industry.
review analyzed 18 papers that met the inclusion criteria and Creating a systematic and transparent digital twin (DT)
were published in English on the use of digital twins in architecture with practical use cases would be an important
healthcare. However, none of the included papers specifically first step in effectively managing these facilities. This study
examined the use of digital twins in the context of Covid-19 presents a DT system architecture specifically designed for
or infectious disease outbreaks. The review identified a need healthcare facility management, with five layers: data
for more research on the use of digital twins in healthcare, acquisition, transmission, integrated middle platform, service,
particularly in the context of infectious disease outbreaks. The and target. The authors demonstrate how this architecture was
paper presents a preliminary conceptual framework for the use used to develop a DT-based smart ward in a Shanghai hospital,
of digital twins in hospital management during the Covid-19 which integrates heterogeneous data sources to support real-
outbreak to address this research gap [35]. time data analysis and decision-making, improves healthcare
service, and helps bridge the gap between humans and
Electronic Health Records (EHRs) have collected a large
healthcare assets and spaces. The paper also outlines future
amount of data in recent years, but there is currently no
directions and opportunities for DT-based healthcare facility
comprehensive approach to using artificial intelligence to
management [39].
understand a patient's overall health. Digital Twin refers to a
comprehensive physical and functional representation of an Digital twins, virtual representations of physical systems
item or system, including all relevant information for all life created using technologies such as Cyber-Physical Systems
cycle phases. This paper presents a platform that uses (CPS), the Internet of Things (IoT), Big Data, Edge
advanced technologies such as Microservice Architecture Computing (EC), Artificial Intelligence (AI), and Machine
(MSA), containerization (Docker), orchestration Learning (ML), have the potential to optimize a wide range of
(Kubernetes), and Machine Learning Operations (MLOps) to applications in industries such as healthcare and smart cities.
create a Digital Twin of Patient platform, inspired by However, digital twins are still in the early stages of
Leonardo DaVinci's Vitruvian man. The platform's development. This paper provides an overview of the
architecture includes clusters of Docker containers and technologies used to create digital twins in industry and
Kubernetes orchestration to represent specific parts or organs healthcare, including the characteristics of digital twins,
of the body as "digital_twin_components," which come communication technologies and tools, reference models,
together to form the "patient_digital_twin" cluster. Pipelines standards, and challenges and open issues. The authors also
are used to define and monitor the construction of the patient's discuss their own recent work in the field of smart
digital twin in real time [36]. manufacturing and healthcare [40].
Digital twins, originally defined as virtual representations A digital twin is a virtual model designed to accurately
of physical assets, systems, or processes, have the potential to reflect a physical object or system. In radiology, a digital twin
improve medical decision-making through the acquisition, of a radiological device can be used to test its characteristics,
management, analysis, prediction, and interpretation of data. make changes to the design or materials, and assess the impact
However, there are many challenges and barriers to using of those changes in a virtual environment. Innovative
digital twins in healthcare. In this paper, the authors propose a technologies such as artificial intelligence and -omics sciences
dynamic digital twin approach to optimize individual patient may be used to create digital twins for patients that can be
healthcare journeys, specifically for women at risk for continually updated based on real-time health and lifestyle
cardiovascular complications during preconception and data. This could be used to improve personalized medicine. In
pregnancy and throughout the lifespan. They discuss the addition, the creation of digital twins based on real-world
potential for multiple domains to contribute to the deployments could enable the use of large virtual patient
development of this digital twin and outline future goals, populations for virtual clinical trials and population studies.
trade-offs, and methods for guiding its development and Digital twins have the potential to transform how diseases are
implementation in healthcare [37]. detected and managed, but there are still significant challenges
to be addressed in their development. Radiologists can play a
Artificial intelligence (AI) and the Internet of Things (IoT)
role in introducing digital twins into healthcare as this
have improved smart city applications such as smart
technology evolves [41].
healthcare, transportation, and environmental management.
Digital twins, which are AI-based virtual replicas of real- Digital twins, virtual representations of physical assets
world entities, have been adopted in manufacturing and such as devices, structures, and patients, are being explored
industrial sectors but are still in the early stages of use in smart and applied in the healthcare industry. While much of the

29

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
focus has been on creating digital twins as standalone
applications, in reality, these assets are often interconnected
and part of a larger physical ecosystem. In this article, the
authors propose the use of digital twins to create virtual
representations of contexts and situations involving multiple
related assets within a healthcare organization, resulting in an
ecosystem of digital twins. They use trauma management as
an example, but suggest that the concept could be applied to
virtualize complex physical realities in a variety of contexts
[42].
VI. FINDINGS AND RESULTS Figure 2. Plagiarism Tool Match Screenshots for ChatGPT
In the analysis made with the Ithenticate Plagiarism Tool, Paraphrased Abstracts
the similarity rate of the entire article (except the
bibliography) was determined as 40%. This ratio occurs when
the texts created by the authors and all the texts created by
ChatGPT are evaluated. It is better to analyse the overall paper
in three different sections:
• Section A: The texts written by the paper's authors
• Section B: The abstract texts of the article
paraphrased by ChatGPT
• Section C: The content generated by ChatGPT,
which are answers to specific questions
Section A had no noteworthy match, as seen in Figure 1.
On the other hand, Section B showed a clear match, as seen in Figure 3. Plagiarism Tool Match Screenshot
Figure 2. The check of Section C is shown in Figure 3. The
matches of these texts are low, but still more than the Section
A results. VII. CONCLUSION AND DISCUSSION
In this study, we first used ChatGPT to create texts for
Digital Twin and Digital Twin in Healthcare titles. The
literature review part of the article is created with these texts.
The articles published from 2020 to 2022 were selected by
searching Google Scholar with the keyword "Digital Twin in
Healthcare". Then the abstracts of these studies were
paraphrased with ChatGPT. Later on, we asked ChatGPT
questions.
In light of the findings and results, the plagiarism tool
matching rates of the texts written by the study authors were
very low. On the other hand, it was seen that the answers given
by ChatGPT for the questions asked were relatively low. On
the contrary, the match rates of the abstract paraphrase tests
created by ChatGPT are very high. From this information, it
Figure 1. Plagiarism Tool Match Screenshot for The Authors’ can be concluded that ChatGPT does not produce original
writings texts after paraphrasing.
This article is a beautiful and striking example of Artificial
Intelligence to demonstrate the point it has reached. It seems
possible for the academic literature to reach a different point
with this and similar applications. The compilation and
expression of knowledge will be accelerated with the help of
artificial intelligence, and these processes will require less
human-driven procedures.
In future work, citations to our study will be monitored. A new
study will be put forward to evaluate the academic validity
after we reach a high number (ex. 100) of citations to this
study from journal articles. These citations will be grouped as
positive, negative, or neutral. Then the evaluations of the
article's adequacy will be considered. In this way, we will
understand whether the academic community accepts the

30

Electronic copy available at: https://ssrn.com/abstract=4308687

 Aydin and Karaarslan
study we put forward. The contribution of this ChatGPT [21] Voigt, I., Inojosa, H., Dillenseger, A., Haase, R., Akgün, K., &
Ziemssen, T. (2021). Digital twins for multiple sclerosis. Frontiers in
implementation to the academic literature will be analysed immunology, 12, 669811.
then. [22] Kamel Boulos, M. N., & Zhang, P. (2021). Digital twins: from
personalised medicine to precision public health. Journal of
REFERENCES Personalized Medicine, 11(8), 745.
[1] ChatGPT (2022). ChatGPT: Optimizing Language Models for [23] Hussain, I., Hossain, M. A., & Park, S. J. (2021, December). A
Dialogue. OpenAI. Retrieved from https://openai.com/blog/chatgpt/ , Healthcare Digital Twin for Diagnosis of Stroke. In 2021 IEEE
Access Date: 16.12.2022. International Conference on Biomedical Engineering, Computer and
[2] Bozkurt Yüksel, D. A. E. (2018). Yapay Zekanın Buluşlarının Information Technology for Health (BECITHCON) (pp. 18-21). IEEE.
Patentlenmesi . Uyuşmazlık Mahkemesi Dergisi , 11.Sayı , 585-622 . [24] Thiong’o, G. M., & Rutka, J. T. (2021). Digital Twin Technology: The
DOI: 10.18771/mdergi.437298 Future of Predicting Neurological Complications of Pediatric Cancers
[3] Aydın, Ö., & Karaarslan, E. (2020, June). An Artificial Intelligence and Their Treatment. Frontiers in Oncology, 11.
Based Decision Support and Resource Management System for [25] Alazab, M., Khan, L. U., Koppu, S., Ramu, S. P., Iyapparaja, M.,
COVID-19 Pandemic (Covid-19 Belirtilerinin Tespiti İçin Dijital İkiz Boobalan, P., ... & Aljuhani, A. (2022). Digital twins for healthcare
Tabanlı Bir Sağlık Bilgi Sistemi). Online International Conference of 4.0-recent advances, architecture, and open challenges. IEEE
COVID- 19 (CONCOVID) Consumer Electronics Magazine.
[4] Coorey, G., Figtree, G. A., Fletcher, D. F., & Redfern, J. (2021). The [26] Yu, Z., Wang, K., Wan, Z., Xie, S., & Lv, Z. (2022). FMCPNN in
health digital twin: advancing precision cardiovascular medicine. Digital Twins Smart Healthcare. IEEE Consumer Electronics
Nature Reviews Cardiology, 18(12), 803-804. Magazine.
[5] Volkov, I., Radchenko, G., & Tchernykh, A. (2021). Digital Twins, [27] Okegbile, S. D., Cai, J., Yi, C., & Niyato, D. (2022). Human digital
Internet of Things and Mobile Medicine: A Review of Current twin for personalized healthcare: Vision, architecture and future
Platforms to Support Smart Healthcare. Programming and Computer directions. IEEE Network.
Software, 47(8), 578-590. [28] Subramanian, B., Kim, J., Maray, M., & Paul, A. (2022). Digital Twin
[6] Shengli, W. (2021). Is human digital twin possible?. Computer Model: A Real-Time Emotion Recognition System for Personalized
Methods and Programs in Biomedicine Update, 1, 100014. Healthcare. IEEE Access, 10, 81155-81165.
[7] Garg, H. (2021). Digital twin technology: Revolutionaryto improve [29] Hassani, H., Huang, X., & MacFeely, S. (2022). Impactful Digital
personalized healthcare: https://doi. org/10.52152/spr/2021.105. Twin in the Healthcare Revolution. Big Data and Cognitive
Science Progress and Research (SPR), 1(1), 32-34. Computing, 6(3), 83.
[8] Popa, E. O., van Hilten, M., Oosterkamp, E., & Bogaardt, M. J. (2021). [30] Huang, P. H., Kim, K. H., & Schermer, M. (2022). Ethical Issues of
The use of digital twins in healthcare: socio-ethical benefits and socio- Digital Twins for Personalized Health Care Service: Preliminary
ethical risks. Life sciences, society and policy, 17(1), 1-25. Mapping Study. Journal of Medical Internet Research, 24(1), e33081.
[9] Elayan, H., Aloqaily, M., & Guizani, M. (2021). Digital twin for [31] Sahal, R., Alsamhi, S. H., & Brown, K. N. (2022). Personal digital
intelligent context-aware IoT healthcare systems. IEEE Internet of twin: a close look into the present and a step towards the future of
Things Journal, 8(23), 16749-16757. personalised healthcare industry. Sensors, 22(15), 5918.
[10] Gupta, D., Kayode, O., Bhatt, S., Gupta, M., & Tosun, A. S. (2021, [32] Wickramasinghe, N. (2022). The case for digital twins in healthcare. In
December). Hierarchical federated learning based anomaly detection Digital Disruption in Health Care (pp. 59-65). Springer, Cham.
using digital twins for smart healthcare. In 2021 IEEE 7th International [33] Akash, S. S., & Ferdous, M. S. (2022). A Blockchain Based System for
Conference on Collaboration and Internet Computing (CIC) (pp. 16- Healthcare Digital Twin. IEEE Access.
25). IEEE.
[34] Ferdousi, R., Laamarti, F., Hossain, M. A., Yang, C., & El Saddik, A.
[11] Zheng, Y., Lu, R., Guan, Y., Zhang, S., & Shao, J. (2021, June). (2022). Digital twins for well-being: an overview. Digital Twin, 1(7),
Towards private similarity query based healthcare monitoring over 7.
digital twin cloud platform. In 2021 IEEE/ACM 29th International
Symposium on Quality of Service (IWQOS) (pp. 1-10). IEEE. [35] Khan, A., Milne-Ives, M., Meinert, E., Iyawa, G. E., Jones, R. B., &
Josephraj, A. N. (2022a). A scoping review of digital twins in the
[12] James, L. (2021). Digital twins will revolutionise healthcare: Digital context of the Covid-19 pandemic. Biomedical Engineering and
twin technology has the potential to transform healthcare in a variety Computational Biology, 13, 11795972221102115.
of ways–improving the diagnosis and treatment of patients,
streamlining preventative care and facilitating new approaches for [36] Kleftakis, S., Mavrogiorgou, A., Mavrogiorgos, K., Kiourtis, A., &
hospital planning. Engineering & Technology, 16(2), 50-53. Kyriazis, D. (2022). Digital Twin in Healthcare Through the Eyes of
the Vitruvian Man. In Innovation in Medicine and Healthcare (pp. 75-
[13] Yang, D., Karimi, H. R., Kaynak, O., & Yin, S. (2021). Developments 85). Springer, Singapore.
of digital twin technologies in industrial, smart city and healthcare
sectors: a survey. Complex Engineering Systems, 1(1), 3. [37] Mulder, S. T., Omidvari, A. H., Rueten-Budde, A. J., Huang, P. H.,
Kim, K. H., Bais, B., ... & Steegers-Theunissen, R. (2022). Dynamic
[14] EL Azzaoui, A., Kim, T. W., Loia, V., & Park, J. H. (2021). Digital Twin: Diagnosis, Treatment, Prediction, and Prevention of
Blockchain-based secure digital twin framework for smart healthy city. Disease During the Life Course. Journal of Medical Internet Research,
In Advanced Multimedia and Ubiquitous Engineering (pp. 107-113). 24(9), e35675.
Springer, Singapore.
[38] Ramu, S. P., Boopalan, P., Pham, Q. V., Maddikunta, P. K. R., Huynh-
[15] De Maeyer, C., & Markopoulos, P. (2021a, July). Future outlook on The, T., Alazab, M., ... & Gadekallu, T. R. (2022). Federated learning
the materialisation, expectations and implementation of Digital Twins enabled digital twins for smart cities: Concepts, recent advances, and
in healthcare. In 34th British HCI Conference 34 (pp. 180-191). future directions. Sustainable Cities and Society, 79, 103663.
[16] De Maeyer, C., & Markopoulos, P. (2021b). Experts’ View on the [39] Song, Y., & Li, Y. Digital Twin Aided Healthcare Facility
Future Outlook on the Materialization, Expectations and Management: A Case Study of Shanghai Tongji Hospital. In
Implementation of Digital Twins in Healthcare. Interacting with Construction Research Congress 2022 (pp. 1145-1155).
Computers, 33(4), 380-394.
[40] Khan, S., Arslan, T., & Ratnarajah, T. (2022b). Digital Twin
[17] Strasser, C. (2021). Towards Digital Twins in Healthcare: How would Perspective of Fourth Industrial and Healthcare Revolution. IEEE
a meaningful Digital Twin for the user look like?. Access, 10, 25732-25754.
[18] Boată, A., Angelescu, R., & Dobrescu, R. (2021). Using digital twins [41] Pesapane, F., Rotili, A., Penco, S., Nicosia, L., & Cassano, E. (2022).
in health care. UPB Scientific Bulletin, Series C: Electrical Digital Twins in Radiology. Journal of Clinical Medicine, 11(21),
Engineering and Computer Science, 53-62. 6553.
[19] Benson, M. (2021). Digital twins will revolutionise healthcare. [42] Ricci, A., Croatti, A., & Montagna, S. (2022). Pervasive and Connected
Engineering & Technology, 16(2), 50-53. Digital Twins–A Vision for Digital Health. IEEE Internet Computing,
[20] Madubuike, O. C., & Anumba, C. J. (2021) Digital Twin Application 26(5), 26 - 32.
in Healthcare Facilities Management. In Computing in Civil
Engineering 2021 (pp. 366-373).

31

Electronic copy available at: https://ssrn.com/abstract=4308687