Journal of

Personalized
Medicine

Review
Clearing the Fog: A Scoping Literature Review on the Ethical
Issues Surrounding Artificial Intelligence-Based Medical Devices
Alessia Maccaro 1 , Katy Stokes 1 , Laura Statham 1,2 , Lucas He 1,3 , Arthur Williams 1 , Leandro Pecchia 1,4
and Davide Piaggio 1, *

1 Applied Biomedical Signal Processing Intelligent eHealth Lab, School of Engineering, University of Warwick,
Coventry CV4 7AL, UK; [email protected] (A.M.); [email protected] (K.S.);
[email protected] (L.S.); [email protected] (L.H.); [email protected] (A.W.);
[email protected] (L.P.)
2 Warwick Medical School, University of Warwick, Coventry CV4 7AL, UK
3 Faculty of Engineering, Imperial College, London SW7 1AY, UK
4 Intelligent Technologies for Health and Well-Being: Sustainable Design, Management and Evaluation, Faculty
of Engineering, Università Campus Bio-Medico Roma, Via Alvaro del Portillo, 21, 00128 Rome, Italy
* Correspondence: [email protected]

Abstract: The use of AI in healthcare has sparked much debate among philosophers, ethicists,
regulators and policymakers who raised concerns about the implications of such technologies. The
presented scoping review captures the progression of the ethical and legal debate and the proposed
ethical frameworks available concerning the use of AI-based medical technologies, capturing key
themes across a wide range of medical contexts. The ethical dimensions are synthesised in order
to produce a coherent ethical framework for AI-based medical technologies, highlighting how
transparency, accountability, confidentiality, autonomy, trust and fairness are the top six recurrent
ethical issues. The literature also highlighted how it is essential to increase ethical awareness through
interdisciplinary research, such that researchers, AI developers and regulators have the necessary
education/competence or networks and tools to ensure proper consideration of ethical matters in
the conception and design of new AI technologies and their norms. Interdisciplinarity throughout
Citation: Maccaro, A.; Stokes, K.; research, regulation and implementation will help ensure AI-based medical devices are ethical,
Statham, L.; He, L.; Williams, A.;
clinically effective and safe. Achieving these goals will facilitate successful translation of AI into
Pecchia, L.; Piaggio, D. Clearing the
healthcare systems, which currently is lagging behind other sectors, to ensure timely achievement of
Fog: A Scoping Literature Review on
health benefits to patients and the public.
the Ethical Issues Surrounding
Artificial Intelligence-Based Medical
Devices. J. Pers. Med. 2024, 14, 443.
Keywords: artificial intelligence; machine learning; healthcare; medical devices; regulatory affairs;
https://doi.org/10.3390/ ethics
jpm14050443

Academic Editor: Himanshu Arora

Received: 20 March 2024

1. Introduction
Revised: 12 April 2024 The medical device market, including medical software, is one of the fastest growing
Accepted: 16 April 2024 sectors in the world, at much higher rates than adjacent sectors, i.e., biotechnology and
Published: 23 April 2024 pharmaceuticals. Eight thousand new medical device patents were granted in Europe
during 2021 alone [1]. Artificial intelligence (AI) in particular, with its capabilities and
promises, is among the main drivers of the Fourth Industrial Revolution, which is set to
transform all sectors, including that of healthcare (Health 4.0). Health 4.0 relies on Industry
Copyright: © 2024 by the authors.
4.0 concepts and principles, characterised by increased interconnectivity between cyber
Licensee MDPI, Basel, Switzerland.
and physical elements and solutions based on disruptive information and communication
This article is an open access article
technologies (ICTs) (e.g., big data and the Internet of Things), to shift from hospital-centred
distributed under the terms and
to patient-centred organisations and services [2–4]. Developments in medical technology
conditions of the Creative Commons
are occurring faster than ever before, with Healthcare 5.0 already being the new desirable
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
frontier in terms of smart disease control and detection, smart health monitoring and man-
4.0/).
agement and virtual care, relying on emerging digital technologies (e.g., nanotechnology,

J. Pers. Med. 2024, 14, 443. https://doi.org/10.3390/jpm14050443 https://www.mdpi.com/journal/jpm

J. Pers. Med. 2024, 14, 443 2 of 19

5G technologies, robotics, IoT, big data, AI, cloud computing, etc.). Healthcare 5.0, with the
introduction of intelligent sensors, is set to overcome the limits of Healthcare 4.0, i.e., the
lack of emotive recognition [5,6].
AI is a broad term referring to the capacity of computers to mimic human intelligence.
Currently, the majority of healthcare applications of AI relate to machine learning for
specific tasks, known as artificial narrow intelligence. There is also interest and debate on
the future use of artificial general intelligence (able to reason, argue and problem solve)
and artificial superintelligence (cognitive capacity greater than that of humanity), which
are at a much earlier stage of research and development [7]. AI is increasingly infiltrating
healthcare, unlocking novelties that in some cases were difficult to imagine. AI has shown
huge potential across many areas including improved diagnosis and disease monitoring
(e.g., using wearable sensors) and improved operational services (e.g., forecasting of phar-
maceutical needs) [8]. This is reflected in the AI-in-healthcare market that was valued at
USD 8.23 billion in 2020 and is set to reach USD 194.4 billion by 2030 [9]. The number of
approvals for AI-enabled medical devices by the Food and Drug Administration follows
this rapid growth: in the last seven years there was a surge from less than ten approvals
yearly to approximately 100 [10]. Among the other benefits, a reduction in healthcare costs
is envisaged. For example, Bohr et al. claim that US annual healthcare costs can be cut by
USD 150 billion in 2026 with AI applications [11].
This wave of technological progress requires fast-responsive regulations and guide-
lines, to safeguard patients and users. Whilst there are several global initiatives for AI
regulation in development, these are still in progress. For example, currently, AI is not
specifically regulated in the UK, though the UK government recently launched a public
consultation on what an AI regulation might look like. In March 2023, the Office for AI
of the UK Department for Science, Innovation and Technology, published a policy paper
highlighting the envisioned UK regulatory approach with a focus on mitigating risks while
fostering innovation, pinpointing the design and publication of an AI Regulation Roadmap
as one of the paramount future steps [12]. In terms of UK regulations for AI-based medical
devices, there is nothing published yet, as the novel UK medical device regulations are
expected to be laid in Parliament only by mid-2025. For this reason, referring to one of the
most recent regulations for medical devices, i.e., the European ones, is more proper for this
article’s purposes.
In Europe, AI systems that qualify as medical devices are regulated as such by the
medical device Regulations (EU2017/745), which were published in May 2017 and came
into effect in 2021 to ensure greater safety and effectiveness of MDs. The MDR introduced
a more stringent risk-based classification system for medical devices, leading to increased
scrutiny and regulation for higher-risk devices and for devices that were previously not
classed as medical devices (e.g., coloured contact lenses). Despite this positive goal, the
MDR still face related risks and challenges that cannot be underestimated [13]. While the
MDR regulate software with intended medical purpose as a MD, i.e., software as a medical
device (SaMD) or medical device software (MDSW), it does not mention, nor specifically
regulates, AI-based MDs. In fact, the need to regulate AI-based MDSW first stemmed
from the action plan published by the U.S. Food and Drug Administration (FDA). The
section “Evolution of AI regulatory frameworks” summarises the main relevant changes
and milestones in the AI regulatory landscape [14].
Since the medical device sector is rapidly expanding, especially in terms of AI-based
solutions, it becomes imperative to find a proper way of integrating novel frameworks
to regulate such devices in the MDR. This should be led by a joint interdisciplinary ef-
fort, comprising biomedical engineers, computer scientists, medical doctors, bioethicists
and policymakers, who should have the necessary competence, networks and tools to en-
sure [15–26] proper consideration of ethical matters in the conception, design, assessment
and regulation of new AI technologies. In fact, this fast-paced evolution also requires the
continuous training and education of the relevant aforementioned figures. Globally, there
are already some initiatives in this direction, which allow medical students to acquire skills
J. Pers. Med. 2024, 14, 443 3 of 19

and knowledge on the cutting-edge technologies and devices that they will use in their
daily practice [27–31]. Nonetheless, compared to the technical aspects, the ethical compo-
nents are often overlooked and are not currently included in the biomedical engineering
curricula [32].
Currently, the use of AI in healthcare has sparked much debate among philosophers
and ethicists who have raised concerns about the fairness, accountability and workforce
implications of such technologies. Key values relating to the use of AI in healthcare include
human agency and oversight, technical robustness and safety, privacy and data governance,
transparency, diversity, nondiscrimination and fairness, societal and environmental well-
being and accountability [15].
In order to address this, this paper presents the results of a scoping review with the
aim of investigating and “clearing the fog” surrounding the ethical considerations of the
use of AI in healthcare.
This specific project is the natural continuation of previous works, in which we have
already described how the word “ethics” is currently widely (ab)used in scientific texts,
but without real competence, almost as if it were a “humanitarian embellishment” [33,34].
One of the most recent consequences of this is the publication of numerous articles about
COVID-19 that were not scientifically sound and were then retracted [33,34]. This is
perfectly aligned with our previous work that shows how an “ethics by design” and frugal
approach to the design and regulation of medical devices via ad hoc frameworks is key for
their safety and effectiveness [35–37].

2. Evolution of AI Regulatory Frameworks

This section summarises the main documents published by groups of ad hoc expert
committees on AI in the latest year relative to the use of AI. These documents vary con-
siderably for countries and regions: e.g., the expert group on AI in the Society of the
Organisation for Economic Co-operation and Development (OECD) issued guidelines for
the responsible use of AI [16]. The private sector has also developed policies relating to
AI (i.e., Google in 2018 [17]). Declarations and recommendations have also been issued by
professional associations and nonprofit organisations such as the Association of Computing
Machinery (ACM) [18], Access Now and Amnesty International [19].
As an inherently data-driven technology, data protection regulation is highly relevant
to AI. In the European Union, the General Data Protection Regulation (GDPR) [20] includes
the right to an explanation for automated decision making and the need for data protection
impact assessments. There is an ongoing need for consistency in ethical guidance across
these regulatory frameworks, that allows stakeholders to shape the ethics of AI in ways
that meet their respective priorities [21].
In this landscape, it is recognised that there is a general lack of public understanding
and trust in AI. Bodies such as the High-Level Expert Group on Artificial Intelligence (AI
HLEG) [38], an independent expert group set up by the European Commission in June
2018, have set out to address this. As a result, AI HLEG published the Ethics Guidelines
for Trustworthy Artificial Intelligence in April 2019, outlining guiding principles for trust-
worthy AI and setting out key requirements to be met by these systems [22]. Based on
fundamental rights and ethical principles, these Guidelines list seven key requirements
that AI systems should meet in order to be trustworthy, namely human agency and over-
sight, technical robustness and safety, privacy and data governance, transparency, diversity,
nondiscrimination and fairness, societal and environmental well-being and accountability.
The document also lists some principles, including human autonomy, the prevention of
harm, fairness, explicability and explainability.
On the 17 of July 2020, the same group presented their final Assessment List for
Trustworthy Artificial Intelligence (ALTAI), a tool designed to support the development of
trustworthy AI, in accordance with the key requirements outlined by the Ethics Guidelines
for Trustworthy Artificial Intelligence [23]. In the same year, the European Commission
published a White Paper entitled “On Artificial Intelligence—A European approach to
J. Pers. Med. 2024, 14, 443 4 of 19

excellence and trust” [24]; the paper works on the objectives of “promoting the uptake of AI
and of addressing the risks”. A key idea from the paper is to outline the requirements appli-
cable to high-risk AI uses. These requirements (stated in section 5D of the aforementioned
paper), aim to increase the trust of humans in the system, by ensuring it is ethical. On the
23 of November 2021, UNESCO’s 193 Member States adopted the “UNESCO Recommen-
dation on the Ethics of Artificial Intelligence”, the first global normative instrument on
the ethics of AI, which addresses the concerns about the protection of life during war and
peace when using AI and lethal autonomous robots, outlined in Human Rights Council
Resolution 47/23 [25].
The latest achievements in this field are quite recent. In December 2023, an agreement
was reached between the EU Parliament and the Council on the EU AI Act, the first-
ever comprehensive legal framework on AI worldwide, that was proposed by the EU
Commission in April 2021 [39]. The President of the EU Commission, Ursula von der
Leyen, commented the following: “The AI Act transposes European values to a new era.
By focusing regulation on identifiable risks, today’s agreement will foster responsible
innovation in Europe. By guaranteeing the safety and fundamental rights of people and
businesses, it will support the development, deployment, and take-up of trustworthy AI
in the EU. Our AI Act will make a substantial contribution to the development of global
rules and principles for human-centric AI”. Furthermore, it is also worth mentioning
the World Health Organization (WHO) involvement in this scenario. In 2024, in fact, the
WHO published a document titled “Ethics and governance of artificial intelligence for
health: Guidance on large multi-modal models”, specific to generative AI applications in
healthcare [26].
In this evolving scenario, as AI-based medical devices directly concern human life,
health and wellbeing, it is essential that the relevant ethical declaration and principles
are respected in all regulatory approaches. In fact, any error caused by the use of AI in
healthcare may have severe consequences, either directly, such as failure to diagnose lethal
conditions, or more widely, such as by leading to a deskilling of the healthcare workforce.
See Supplementary Table S1 for further details on the relevant existing laws related to the
main topics of ethical concerns about AI.

3. Methods
3.1. Search Strategy
This scoping literature review was conducted according to PRISMA guidelines (Pre-
ferred Reporting Items for Systematic Reviews and Meta-Analyses) [40]. The protocol
for the review was not registered. The search was run from database inception to April
2022 and updated in September 2023. The three main topics, i.e., ethics, AI, healthcare
technology, were then put together with the AND operator, see Table 1.

Table 1. Terms/string used for systematic search, divided by area. Each area was then put together
with the AND operator. * denotes any character or range of characters in the database search.

Area String
Ethic* OR bioethic* OR ((cod*)AND(ethic*)) OR ((ethical OR
Ethics terms moral) AND (value* OR principle*)) OR deontolog* OR
(meta-ethics) AND (issue or problem or challenge)
(“artificial intelligence”) OR (“neural network”) OR (“deep
learning”) OR (“deep-learning”) OR (“machine learning”) OR
Artificial intelligence terms
(“machine-learning”) OR AI OR iot OR (“internet of things”)
OR (expert system)
Health OR healthcare OR (health care) OR (medical device*)
Healthcare technology terms OR (medical technolog*) OR (medical equipment) OR
((healthcare OR (health care)) AND (technolog*))
J. Pers. Med. 2024, 14, x FOR PEER REVIEW 5 of 19

J. Pers. Med. 2024, 14, 443 5 of 19

3.2. Study Eligibility
Only scientific articles focusing on ethical challenges/issues arising from medical
AI/technology were included. Studies were excluded if the following criteria were met:
3.2. Study Eligibility
non-English language, the full text was not freely accessible, grey literature, letter to edi-
Only scientific articles focusing on ethical challenges/issues arising from medical
tors, editorials, commentaries or review articles.
AI/technology were included. Studies were excluded if the following criteria were met:
non-English language, the full text was not freely accessible, grey literature, letter to editors,
3.3. Study Selection
editorials, commentaries or review articles.
Two authors independently screened the studies by title and abstract while three au-
3.3. completed
thors Study Selection
full text screening against the inclusion and exclusion criteria, with con-
flictingTwo
decisions
authorsbeing mitigated byscreened
independently an additional reviewer.
the studies by Relevant
title and data werewhile
abstract then three
ex-
tracted using an ad hoc extraction table to facilitate the analysis and narrative synthesis.
authors completed full text screening against the inclusion and exclusion criteria, with
conflicting decisions being mitigated by an additional reviewer. Relevant data were then
3.4. Data Extraction
extracted using an ad hoc extraction table to facilitate the analysis and narrative synthesis.
Relevant data were extracted and collected in an ad hoc Excel sheet, organised by
3.4. Data
author, Extraction
main ethical issues raised, technological and medical context and key findings.
Relevant data were extracted and collected in an ad hoc Excel sheet, organised by
author,
3.5. main ethical issues raised, technological and medical context and key findings.
Data Synthesis
To synthesise the extracted data, a narrative synthesis method was used [41]. For each
3.5. Data Synthesis
study, the main ethical issues and their solutions were identified, which were then de-
scribedToin synthesise
the results,the extracted
organised by data, a narrative
principal themes.synthesis method was used [41]. For
each study, the main ethical issues and their solutions were identified, which were then
4. described
Results in the results, organised by principal themes.
4.1.
4. Study
ResultsSelection
4.1.AStudy
total of 2059 studies were returned through the search strategy and 73 of these were
Selection
includedA total offinal
in the 2059review,
studiesconsidering the fulfilment
were returned through of theour inclusion
search andand
strategy exclusion cri-
73 of these
teria.
wereThe study selection
included process
in the final review,isconsidering
summarisedthein fulfilment
Figure 1. of our inclusion and exclusion
criteria. The study selection process is summarised in Figure 1.

Figure
Figure 1. 1.
PRISMA
PRISMA flow
flowdiagram
diagramforforstudy
studyscreening
screeningand
andinclusion.
inclusion.**Refers
Refersto
to studies
studies identified
identified from
from
the the records
records retrieved
retrieved fromfrom
the the database
database which
which were
were not not
heldheld on the
on the database
database searched.
searched.
J. Pers. Med. 2024, 14, 443 6 of 19

4.2. Study Characteristics

The 73 included studies covered a broad spectrum of medical contexts, summarised
with the main ethical issues raised and key findings in Supplementary Table S2. Briefly,
the most frequently addressed areas were as follows: the general use of AI for healthcare
(36 studies) [42–77], the use of AI in decision support systems (eight studies) [78–94],
big data (four studies) [95–98], robotics (seven studies) [99–105] and adaptive AI (one
study) [106]. The remaining studies addressed the following: rehabilitation [107], medical
education [108], monitoring technology for the elderly [109], mental health [110], radiation
technology [111], chatbots [112], health apps [113] and healthcare in low- and middle-
income countries [114].
Across the 73 selected studies, eight main ethical themes concerning medical technol-
ogy were identified including the following: transparency, algorithmic bias, confidentiality,
fairness, trust, autonomy, accountability and informed consent. For the purpose of this
study, the ethical themes are defined in the context of AI, as follows:
Transparency: “the data, system and AI business models should be transparent.
Traceability mechanisms can help achieving this. Moreover, AI systems and their decisions
should be explained in a manner adapted to the stakeholder concerned. Humans need to
be aware that they are interacting with an AI system, and must be informed of the system’s
capabilities and limitations” [22].
Algorithmic bias: “the instances when the application of an algorithm compounds
existing inequities in socioeconomic status, race, ethnic background, religion, gender,
disability or sexual orientation to amplify them and adversely impact inequities in health
systems” [115].
Confidentiality: “the responsibility of those entrusted with those data to maintain
privacy” [116].
Fairness: “a commitment to ensuring equal and just distribution of both benefits and
costs, and ensuring that individuals and groups are free from unfair bias, discrimination
and stigmatisation” [22]
Trust: “a set of specific beliefs dealing with benevolence, competence, integrity, and
predictability; the willingness of one party to depend on another in a risky situation
(trusting intention); or the combination of these elements” [22].
Autonomy: “The right of an individual to make his or her own choice” [117].
Accountability: “Mechanisms should be put in place to ensure responsibility and ac-
countability for AI systems and their outcomes. Auditability, which enables the assessment
of algorithms, data and design processes plays a key role therein, especially in critical
applications. Moreover, adequate an accessible redress should be ensured” [22].
Informed consent: “Consent of the data subject means any freely given, specific,
informed and unambiguous indication of the data subject’s wishes by which he or she, by a
statement or by a clear affirmative action, signifies agreement to the processing of personal
data relating to him or her” [20].
The frequencies of reporting of the ethical themes are displayed in Table 2, along with
the proposed solutions. The selected studies focused on varied technological contexts,
displayed in Figure 2. A summary of the medical contexts covered in the included studies
is provided in Supplementary Table S3.
J. Pers. Med. 2024, 14, 443 7 of 19

Table 2. Proportion of studies reporting common ethical themes and the proposed solutions. Others
included human bias, value, beneficence, nonmaleficence and integrity.

Ethical Themes Number of Studies (%) Proposed Solution/s

J. Pers. Med. 2024, 14, x FOR PEER REVIEW 7 of 19
Transparency 40 (17)
Algorithmic bias 30 (13)
• Multidisciplinary approach to
Confidentiality 33 (14)
development
Trust 34Fairness
(15) 24 (10)
• Standardised framework,
Trust 34 (15)
Autonomy 24Autonomy
(10) 24 (10)
regulation and policy
Informed consent 13Informed
(6) • Education on AI technology for
consent 13 (6)
healthcare professionals and users
Accountability 25Accountability
(11) 25 (11)
Other 11Other
(5) 11 (5)

Adaptive AI 1.2
Radiation technology 1.2
Monitoring technology 1.2
Big data 1.2
Virtual agents/Chatbot 2.4
Apps 4.9
Robotics 8.5
Clinical decision support system (CDSS) 23.2
General 56.1

0 10 20 30 40 50 60
% of studies

Figure 2. Percentage of studies discussing different technological contexts. Multiple technologies

Figure 2. Percentage of studies discussing different technological contexts. Multiple technologies
were discussed in some studies.
were discussed in some studies.
4.3. Transparency
4.3. Transparency
The most frequently addressed ethical theme was transparency (n = 40) [44,46,47,49,55–
The most frequently addressed ethical theme was transparencyconcerning
62,64,66,69,70,75,76,78–82,85,86,88–90,92,94–96,99–101,107,110,111,113,114], (n = 40)a
[44,46,47,49,55–62,64,66,69,70,75,76,78–82,85,86,88–90,92,94–96,99–
wide range of medical contexts but focusing mainly in the area of clinical decision support.
101,107,110,111,113,114],
While the majority of studies concerning
reported a wide range of
a narrative medical contexts
evaluation of ethical but focusing
issues, seven mainly
quali-
in the area of clinical decision support. While the majority of studies
tative studies employed surveys or focus groups that identified transparency as an ethical reported a narrative
evaluation
priority byofeitherethical issues, seven
healthcare qualitative
professionals studies
[44,78], employed
experts in AIsurveys or focus groups
implementation [107], a
that identifiedof transparency
combination AI experts andascitizens an ethical
[65] orpriority by either
a combination healthcare clinician-users,
of designers, professionals
[44,78], experts in
administrators andAIpatients,
implementation
[69,70,89].[107],
For aexample,
combination de Boerof AI experts
et al. and citizens
compared concerns [65]of
or a combination
healthcare of designers,
professionals regardingclinician-users, administrators
clinical decision and patients,
support systems, describing[69,70,89].
theories Forof
‘technomoral
example, de Boerchange’
et al.and ‘technological
compared concerns mediation’
of healthcareas a means to reflect
professionals and identify
regarding key
clinical
concernssupport
decision of machine learning
systems, in healthcare
describing theories[78].
of ‘technomoral change’ and ‘technological
Cagliero
mediation’ as eta al. performed
means a caseand
to reflect study analysis
identify keyandconcerns
collectedof viewpoints
machine from various
learning in
stakeholders,
healthcare [78].noting that while designers appeared focused on the importance of algorithm
validation,
Cagliero clinicians and patients
et al. performed had
a case a desire
study to understand
analysis and collected(at least at a certain
viewpoints from level) how
various
the AI systemnoting
stakeholders, worksthat [89].while
This designers
demonstrates a so-called
appeared focused ‘values-collision’,
on the importance highlighting
of algo-
that itvalidation,
rithm should notclinicians
be taken and for granted
patientsthat
hadall stakeholders
a desire assume (at
to understand theleast
sameatviewpoint
a certain
of what is required for transparency. This can also be seen
level) how the AI system works [89]. This demonstrates a so-called ‘values-collision’, to reflect two sides of the
high-
discussion of transparency, on the one hand, there can be seen
lighting that it should not be taken for granted that all stakeholders assume the samea need for rigorous validation
which is communicated
viewpoint of what is required well, on forthe other hand,This
transparency. regardless
can also of be
theseen
validation, thetwo
to reflect ability
sidesto
understand
of the discussion the AI of processes
transparency, is inherently
on the one important
hand, thereto people.
can beThis seenwas a key
a need fordiscussion
rigorous
point which was raised in depth in several explorations of transparency.
validation which is communicated well, on the other hand, regardless of the validation, Indeed, a critical
and somewhat unique challenge raised by the nature of AI-driven
the ability to understand the AI processes is inherently important to people. This was technologies is the extenta
to which it is possible to understand how the underlying AI
key discussion point which was raised in depth in several explorations of transparency. system works or reaches
Indeed, a critical and somewhat unique challenge raised by the nature of AI-driven tech-
nologies is the extent to which it is possible to understand how the underlying AI system
works or reaches its output [49,60,62,64,66,70,85,86,88,92]. Many studies called for this ‘ex-
plicability’ or ‘explainability’ as a key principle to be met for AI health technologies
[60,64], while others emphasised the need for contextual explainability [88]. Adams et al.
J. Pers. Med. 2024, 14, 443 8 of 19

its output [49,60,62,64,66,70,85,86,88,92]. Many studies called for this ‘explicability’ or

‘explainability’ as a key principle to be met for AI health technologies [60,64], while others
emphasised the need for contextual explainability [88]. Adams et al. provided arguments
for the inclusion of explainability as a principle of bioethics, negating challenges from others
that medical decision making by people themselves is inherently not transparent [64]. It is
not surprising, therefore, given the range of arguments, that it was noted by several authors
that for the final AI-based device to be considered transparent and acceptable for use,
stakeholders such as patients and the public should be involved early in the development
process [58,66,80,89,90,95,111]. This was supported by several studies highlighting the
need for a multidisciplinary approach to the issue of transparency and ethics more widely
(specifically, ethics training or integrating ethicists) [44,47,59,69,75,100,101]. In terms of
barriers, AI development experts identified a lack of transparency to be seen as a limit
to positive relationships between vulnerable patient groups (such as patients of older
generations) and clinicians [107].
More broadly, several studies noted that the public should have access to transparent
information regarding the use of AI in healthcare [61,66,69,76], with some pinpointing that
governments should play a role in raising awareness, especially of prospective vulnerable
populations, such that information provided regarding AI-based devices can be well
understood [96]. Further, to facilitate transparency, AI standardised ethical frameworks
and clear regulations are required, with some studies calling to enforce their use with
legislation [46,56,57,75,82,94,107,110,114].

4.4. Accountability
Accountability emerged as a theme from 25 studies, particularly from studies concern-
ing applications of large language models/chatbots and robotics [42,46,55,56,60,61,69,71,
74,75,78–81,84,88,97,100–104,110–112].
Several authors highlighted the importance of accountability at the patient care level,
underpinning trust in the patient–clinician relationship, which may be changed or chal-
lenged by the use of chatbots and decision support systems [79,81,101,102,112]. Studies
called for clear models for cases of investigation of medical accidents or incidents involving
the use of AI [55,110]; one study emphasised this as a necessity in order to truly prepare
healthcare systems for the use of AI [104]. In this way, legal accountability must be made
clear for applications of AI decisions or decision support [74]. Implementation of AI needs
to also be supported by ethical design and implementation frameworks or guidelines, for
which designers are accountable to meet [56,61,71,97,111]. In some cases, authors advo-
cated for ensuring medical AI is always ‘supervised’ by a healthcare professional, who
ultimately has accountability for the technology [61].
As with transparency, multidisciplinarity (specifically, training and integrating ethi-
cists) was raised as essential in ensuring acceptable levels of accountability in health-related
decisions [42,74,78,80,84,100,101].

4.5. Confidentiality
A total of 33 of the included papers examined the challenges and considerations
surrounding the use of AI, particularly in healthcare, and the imperative of safeguarding
confidentiality in the digital age [42–44,46,47,50,53,54,58–61,66–68,71,72,74–76,84,90,93,94,
96,98,102,104,105,109–111,113].
Broadly, these papers address the complexity of maintaining confidentiality in an era
where AI technologies are increasingly integrated into healthcare systems. Key themes that
emerged include the tension between technological advancement and ethical constraints,
the impact of AI on patient privacy and data security, and the moral obligations of AI
developers and users towards ensuring the confidentiality of sensitive information.
A significant number of papers focus on the use of AI in healthcare, particularly
concerning patient data privacy and security. This group explores the challenges and
ethical considerations in safeguarding patient information in the context of AI-driven
J. Pers. Med. 2024, 14, 443 9 of 19

medical practices and health apps [43,44,46,47,53,60,61,68,71,84,90,102,105,113]. Techno-

logical and practical challenges in ensuring confidentiality was also addressed [53,61,104].
This included discussions on data encryption, secure data handling practices and the
implementation of robust security measures to prevent data breaches.
In summary, the collective insights from these papers underscore the critical need
for robust ethical frameworks and security measures in AI applications, particularly in
healthcare. They highlighted the imperative of balancing technological innovation with the
ethical responsibility to protect confidentiality and maintain the integrity of patient data.
The papers suggested a multidisciplinary approach, involving stakeholders at various
levels, to address these challenges effectively. There was a consensus on the need for
ongoing dialogue, policy development, and ethical guidelines to navigate the complex
landscape of AI and confidentiality.

4.6. Autonomy
The theme of autonomy was mentioned in 26 of our inclusions, and was reviewed
on various aspects, from philosophical and ethical foundations to practical implications in
healthcare and other domains [43,46,48–50,56,59,61,64,69,71,74,76,79,84,85,87,88,91,99,100,
105,109,113]. The theme intricately linked with issues such as decision making, control,
and the human-centric approach to AI development and implementation.
Sevesson et al. discussed the impact of AI on human dignity and autonomy, empha-
sising the need to maintain the uniqueness and intrinsic value of humanity. Kuhler M et al.
explored paternalism in health apps, including fitness and wellbeing applications, and its
implications for autonomy, particularly in AI-driven tools [113]. Braun M et al. delved into
decision-making complexities in AI contexts and introduced the concept of ‘Meaningful
human control’ as a framework to ensure autonomy in AI systems [79]. Compliance with
universal standards in AI, particularly stressing the importance of maintaining autonomy
in the face of technological advancements, was proposed by Arima et al. [46]. Similarly,
Guan et al. addressed the application of AI in various sectors, advocating for specific
guidelines to ensure autonomy, especially in frontier technologies [48].
The central tenet of these papers called for the imperative to preserve and respect
human autonomy in the face of rapidly advancing AI technologies. The authors collectively
emphasised that AI should be developed and implemented in ways that enhance human
decision making and independence, rather than undermining it.

4.7. Algorithmic Bias in AI-Based Medical Devices

Algorithmic bias refers to the tendency of AI systems to exhibit systematic errors that
disproportionately affect certain patient groups, directly influencing patient treatment and
the efficacy of medical interventions. This bias can emerge from various sources, including
biased training data or flawed algorithm design, leading to unequal treatment outcomes.
Studies from Kerasidou [114] and McLennan et al. [42] highlighted the ethical need to
address these biases in smart healthcare systems. They argued for the development of AI
systems that prioritize equity and fairness, ensuring that these technologies serve as reliable
and unbiased tools in medical diagnostics and treatment. Zhang and Zhang emphasised
the critical role of transparency in AI systems, advocating for measures that prevent the
deepening of existing healthcare disparities through biased algorithms [61]. Similarly,
Liu et al. discussed the ethical challenges posed by digital therapeutics, including AI-
driven devices, stressing the importance of considering diverse patient populations during
development to mitigate bias [98]. Hallowell et al. explored the ethical considerations in the
design of medical IoT devices, also emphasising the need for inclusive and fair algorithms
that cater to diverse patient needs [87].

4.8. Informed Consent in the Era of AI

Informed consent in AI-enhanced medical care has become increasingly complex. The
concept extends beyond the traditional model of patient–physician interaction, incorpo-
J. Pers. Med. 2024, 14, 443 10 of 19

rating the understanding of AI-driven processes that influence patient care. Lorenzini
et al. [88] and Astromskė et al. [82] addressed the complexities involved in obtaining
informed consent when medical decision making is augmented with machine learning,
emphasising the need for clarity in communication. Leimanis and Palkova [108] and Parvi-
ainen and Rantala [112] further discussed the principle of patient autonomy in this context,
highlighting the right of patients to make informed decisions about their care, particularly
when influenced by advanced medical technologies. Astromskė et al. [82] delved into
the practical challenges of ensuring informed consent in the context of AI-driven medical
consultations, suggesting strategies to enhance patient understanding and autonomy. Ho
discussed the ethical considerations in using AI for elderly care, particularly focusing on
the need for clear consent processes tailored to this demographic [109].

4.9. Intersection of Algorithmic Bias and Informed Consent

The intersection of algorithmic bias and informed consent presents unique challenges
in AI-based medical care. Biased algorithms can obscure decision-making processes, con-
sequently affecting the ability of patients to provide informed consent. Transparent AI
systems, as advocated in [106,113], are essential to ensure that patients understand how
biases in AI might impact their healthcare. Education and awareness, as highlighted
in [95,98], play a vital role in enabling informed patient–provider discussions about the
role of AI in healthcare. The ethical deployment of AI in medical devices necessitates a
comprehensive understanding and mitigation of algorithmic bias and informed consent
challenges. Collaborative efforts involving technology design, patient education, regula-
tory frameworks, and ethical considerations are paramount. This collaborative approach
ensures the development of equitable, transparent and patient-centred healthcare solu-
tions. It is through this integrated perspective that AI in healthcare can be effectively
navigated, ensuring that its benefits are maximized while minimizing potential harms and
ethical complexities.

4.10. Trust
Trust was discussed as a theme in 34 of the included studies [42,50,52,54,58–63,66–
68,72,73,76,77,79–82,86,87,89,90,93,94,101,102,105,107,110,112,113], with most focusing on
clinical decision support systems, chatbots and robots. Most of these studies were articles
or qualitative analyses. The main concern raised within this theme was the impact of un-
trustworthy AI on clinician–patient relationships. Several studies described how building
a reliable doctor–patient relationship relies upon the transparency of the AI device [81,107],
as previously discussed. Interviewees of one qualitative study described how the perceived
reliability and trustworthiness of AI technology relies upon validating its results over time,
and bias is a significant problem that may impair this [87]. Arnold also described how AI
devices may erode trust if doctors do not have the autonomy or control of these devices [50].
Braun et al. echoed this, suggesting ‘meaningful human control’ must be developed as
a concept to stand as a framework for AI development, especially in healthcare where
decisions are critical [79].
Medical chatbots were discussed as a mode for increasing rationality but also leading
to automation, which may lead to incompleteness, and, therefore, a loss of trust [112].
De Togni described how human and machine relationships are uncertain in comparison,
and there is a need to rematerialize the boundaries between humans and machines [70].
Other recommendations given to improve trustworthiness included multidisciplinary
collaboration, for example engaging with both clinicians, machine learning experts and
computer program designers [58,98], more precise regulation [60] and specific guidelines
for frontier AI fields [48].

4.11. Fairness
A total of 25 studies concerned the topic of fairness, covering a range of contexts [44,
45,48,54,56,58,59,66–69,71,73,80,81,83,85,89,95,98,99,101,103,106,114]. This theme largely
J. Pers. Med. 2024, 14, 443 11 of 19

discussed justice and resource allocation of AI technology. Pasricha explained how most
vulnerable patients do not have access to AI-based healthcare devices, and that AI should
be designed to promote fairness [98]. Kerasidou specifically discussed the ethical issues
affecting health AI in low- or middle-income countries (LMICs), concluding that further
international regulation is required to ensure fair and appropriate AI [114]. This was echoed
by others indicating that a revision of guidelines is necessary to ensure fair medical AI tech-
nology [56]. Another suggestion was ethicist involvement with AI technology development,
with the view that this may improve the chance that AI is fair and unbiased [98].

5. Discussion
Overall, the ethical considerations surrounding AI are complex and multifaceted and
will continue to evolve as the technology itself advances, although it seems that traditional
issues are not yet fully overcome, since they are still a matter of consideration and concern.
There is an ongoing need to assess the ethical issues and proposed solutions and to identify
gaps and best routes for progress. In particular, common concerns include the following:
• The lack of transparency in relation to data collection, use of personal data, explain-
ability of AI and its effects on the relationship between the users and the service
providers;
• The challenge of identifying who is responsible for medical AI technology. As AI
systems become increasingly advanced and autonomous, there are questions about
the level of agency and control that should be afforded to them and about how to
ensure that this technology acts in the best interests of human beings;
• The pervasiveness, invasiveness and intrusiveness of technology that is difficult for the
users to understand and therefore challenges the process of obtaining a fully informed
consent;
• The lack of the establishment of a trust framework that ensures the protection/security
of shared personal data, enhanced privacy and usable security countermeasures on
the personal and sensitive data interchange among IoT systems;
• The difficulty of creating fair/equitable technology without algorithmic bias;
• The difficulty of respecting autonomy, privacy and confidentiality, particularly when
third parties may have a strong interest in getting access to electronically recorded and
stored personal data.
Starting from the aforementioned AI HLEG (EU Commission) Ethics Guidelines
for Trustworthy Artificial Intelligence and its four principles, namely respect for human
autonomy, prevention of harm, fairness and explicability, it can be noted that, upon closer
inspection, they are comparable to the classic principles of bioethics, namely beneficence,
nonmaleficence, autonomy and justice. The latter are considered the framework for ethics
and AI by Floridi et al., who further adds “explicability, understood as incorporating both
intelligibility and accountability”. Autonomy clearly features both lists [118]. Prevention of
harm could be seen as parallel to nonmaleficence (i.e., to avoid bias and respect security
and privacy). Fairness includes beneficence and justice, not only relative to the individual
but to society as well. Findings from this scoping review strongly support the proposition
of Floridi et al. to include explainability as a principle of modern bioethics.
The topic of explicability/explainability is also addressed by the AI HLEG document
and is related to the ethical theme of transparency, which was addressed in over half
of all the studies included in this review. The transparency of AI may also be seen to
underpin other ethical concerns including trust, fairness and accountability. In particular,
the appropriate selection and use of medical devices relies on an understanding of how they
work, which is key to mitigating any possible risks or biases. However, in some cases, it
could be challenging or impossible to determine how an AI system reaches an output (e.g.,
black boxes) and this is well interwoven with the concept of ‘explainability’ of AI, referring
to the level of understanding in the way a system reaches its output. The most extreme case
is the so-called ‘black box’ systems, where no information is available on how the output
is reached. Increasing the explainability of AI algorithms is an active research field and
J. Pers. Med. 2024, 14, 443 12 of 19

there is a growing number of methods aiming to offer insight as to how AI predictions are
reached [119]. However, significant debate remains as to whether it is ever appropriate to
deploy algorithms which are unexplainable in healthcare settings. The question of whether
(or to what degree) AI must be explainable, and to who, is complex. Poor communication
between stakeholders has been identified in previous literature as a limiting factor in the
successful development of AI health technologies, with calls for increased representation
of diverse ethnic socioeconomic and demographic groups and promotion of open science
approaches to prevent algorithmic bias from occurring. Involving interdisciplinary and
cross-sector stakeholders (including healthcare professionals, patients, carers and the
public) in the design and deployment of AI will help to ensure the technologies are designed
with transparency, that they meet clinical needs and that they are ultimately acceptable to
users [120,121].
Transparency also relates to autonomy and consent; if a clinician cannot describe the
details involved in the AI’s decision-making process, the relevant information may not be
communicated to a patient effectively, preventing fully informed consent from taking place.
Also, accountability is noteworthy; who can be held responsible for decisions made via
clinical decision support systems when the developers cannot explain the decision-making
process that has occurred? Leimanis et al., therefore, suggested that AI systems cannot yet
be the primary decision maker, rather they should act only as an assistant to clinicians [108].
As demonstrated by the findings of this review, a frequent theme in the debate on
ethics, AI and the IoT entails issues related to the sharing and protection of personal data.
It has been argued that one key characteristic of the use of the “things” in the IoT is that the
collection of information is passive and constantly ongoing, making it difficult for users to
control the sharing and use of data. Andrejevic and Burdon described this phenomenon as
the “sensor society”, where sensor driven data collection takes place in a complex system,
where the collection and analysis target a pattern of data rather than the individual persons
and where processes of data collection and analysis are opaque. As a consequence, it is
difficult for an individual to anticipate how their individual data will be used [122]. The
above discussion highlights the way in which the evolution and roll out of IoT applications
is taking place against the backdrop of discussions around trust, transparency, privacy
and security.
Health-related data are considered personal and classed as sensitive information
throughout the lifecycle (acquisition, storage, transfer and destruction). Due to the sen-
sitivity of the data and the potential consequences for the users, human control over
algorithms and decision-making systems is paramount for these applications. For example,
as noted in projects related to the IoT and Active and Healthy Ageing (EU Large-Scale
Pilot, GATEKEEPER [123]), while the continuous monitoring of personal health data can be
very beneficial to improve and personalise treatment, some may worry about ethical issues
like constant surveillance and lack of control over the data collected, hindering autonomy
and confidentiality. Ho (2020) described how monitoring technology for older adults may
be effective in reducing caregivers’ burden and improve the quality of care but may be
viewed as an invasion of privacy and can affect family dynamics [109]. This situation is also
complicated in cases where patients, for example older people with cognitive impairments,
may not be in a position to participate in the decision-making process around privacy
settings, but can be supported by either health information counsellors or some AI-based
tools (e.g., assistive technologies).
Hence, an urgent need has emerged for a universal (recognised by law) ethical frame-
work that can support all the individuals involved with the use of AI in healthcare. For
example, in the medical field, it will assist medical professionals, carers and other health
service providers in meeting their moral responsibilities in providing healthcare and man-
agement. Likewise, users will be empowered and protected from potential exploitation and
harm via the AI technology. By creating and adopting an ethical framework and guidelines,
developers could demonstrate a serious commitment to meeting their legal and moral
responsibilities to users, care providers and other stakeholders. Furthermore, this may
J. Pers. Med. 2024, 14, 443 13 of 19

prevent many foreseeable ethical problems in the design and roll out of IoT devices and
protocols, for which developers would be legally or morally liable. In ongoing discussions
on forming an ethical framework for AI and IoT, trust is a recurring theme. All stakeholders
involved in the development, deployment and use of AI and IoT applications need to be
ensured that the systems demonstrate their trustworthiness from social, technical and legal
perspectives.
In accordance with this principle, as seen in the results of this review, the debate
proposes some solutions to develop a framework of ethical guidelines on AI in health-
care. In primis, a potential solution includes the consideration of a multidisciplinary
approach [44,85,100,107], or more specifically involving experts from ethics [47,80,111],
bioethics [54] and policy [84], encouraging the involvement of the stakeholders [58] and
their communication [80]. Multidisciplinarity is intended not only at the theoretical debate
level, but also practically, for example, involving physicians in the designs of AI-based
medical technology [55,58], along with bioethics and policy experts [54,84] and other stake-
holders [58,80]. Other authors referred to embedded ethics [42] as a means of integrating
ethics in technology design, development and deployment to minimise risks and fears.
For example, Smith proposed an integrated life-cycle approach to AI, integrating ethics
throughout transparency, replicability and effectiveness (TREE) [42,111].
Another important point is the standardisation of regulatory frameworks at the inter-
national level [46,48,114], in particular offering better guidance for low- and middle-income
countries [83]. The main debate considers the choice between improving the existing ethical–
legal solutions [46,56,57,81,95,96,102] or proposing new ethical–political approaches and
policy decisions [43,112]. In relation to this, it is noteworthy to mention that certain basic
ethical principles are indisputable. Therefore, when updating existing guidelines with the
latest technological advancements, existing frameworks cannot be disregarded.
Finally, the improvement of training and education on technology for profession-
als [47,50,55,82,99,107] and the general public [96,107] is paramount. It is essential to
not only create cross-sectoral expertise encouraging basic ethical training at schools and
universities, but also on the basic elements of technologies. This does not mean that pro-
fessionals in a field should be experts in all the relevant disciplines. Rather, this basic
multidisciplinary knowledge is key to promoting and facilitating the communication on
common topics among experts from different disciplines. Creating multidisciplinary teams
helps constructive dialogue and prepares citizens for technological advancement without
unnecessary fears, but with a full sense of responsibility. In light of this, some authors
referred to “health information counsellors” [57,99], who can support patient autonomy
regarding healthcare decisions. It is essential to reflect on figures such as ethic counsellors
or the ethical committees in research and clinical practice, which are aimed at supporting
patients and medical staff with ethical queries and technologies.
In light of this, the authors of this manuscript believe that it is neither necessary
nor useful to rethink the basic principles of ethics in order to propose a framework that
responds to the new needs emerging from the use of AI in medicine. However, they believe
that a specific, context-aware and internationally harmonised approach to the regulation of
AI for medical applications is required urgently to “clear the fog” around this topic. Such
an approach could be built starting from the principles listed above (i.e., respect for human
autonomy, prevention of harm, fairness and explainability or the parallel bioethical ones,
i.e., autonomy, nonmaleficence, beneficence and justice with the addition of explainability).
Many of the issues raised here exist more widely in the regulation of medical devices,
as some of the authors of this paper have highlighted in previous work [37]. On a similar
thread, some of the authors of this project have already proposed the need for frugal
regulations for medical devices, declaring that the current regulatory frameworks for
medical devices are not aware of peculiar contexts or responsive to their specific needs [37].
All in all, this regulation for the use of AI in the medical field will only be possible
through the combination of solutions: defining a unique ethical–legal framework involv-
ing multidisciplinary teams and intercultural and international perspectives, involving
J. Pers. Med. 2024, 14, 443 14 of 19

stakeholders and the public through education in ethics and technology as well as the
consultation in the development of guidelines and technology.

6. Conclusions
This paper presents the results of a scoping literature review on ethics and AI-based
medical technology. The objectives of this review were as follows:
• Clarifying the ethical debate on AI-based solutions and identifying key issues;
• Fostering the ethical competence of biomedical engineering students, who are coau-
thors of this paper, introducing them to interdisciplinarity in research as a good
practice;
• Enriching our already existing framework with the need for considerations of ethical–
legal aspects of AI-based medical device solutions, awareness of the existing debates
and an innovative and interdisciplinary approach. Such a framework could support
AI-based medical device design and regulations at an international level.
The ethics of AI is a complex and multifaceted topic that encompasses a wide range
of recurring issues (for example, transparency, accountability, confidentiality, autonomy,
trust and fairness), which are not yet addressed by a single and binding legal reference
at the international level. For this, the authors of this paper propose several solutions
(interdisciplinarity, legal strength and citizenship involvement/education) in order to
reinforce the theories presented in their legal–ethical framework. This tool, intended to
support the development of future health technologies, is adaptable and versatile and in
continuous refinement.
In conclusion, this work is a step forward in understanding the ethical issues raised by
novel AI-based medical technologies and what guidance is required to face these challenges
and prevent patient/user’s harm. Although this work is focused on the ethical debate on
AI-based medical technologies, it sits well in the wider context of that relative to ethics
and technology, in order to “clear” the existing fog and shed a light on the next steps into
the future.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/jpm14050443/s1, Table S1: This table contains a summary of the
relevant law/directives related to the main topics of ethical concerns about AI; Table S2. Summary of
study technological and medical contexts and outcomes; Table S3. Percentage of studies discussing
different medical contexts (n = 41). Certain studies addressed more than one medical context.
Author Contributions: Conceptualization, D.P. and A.M.; methodology, D.P., A.M. and L.P.; formal
analysis, K.S., L.S., L.H. and A.W.; data curation, K.S., L.S., L.H. and A.W; writing—original draft
preparation, K.S., L.S., L.H., A.W., A.M. and D.P.; writing—review and editing, K.S., A.M., D.P. and
L.P.; visualization, K.S. and L.S.; supervision, D.P. and A.M.; project administration, D.P. and A.M.;
funding acquisition, L.P. All authors have read and agreed to the published version of the manuscript.
Funding: K.S. is funded by the MRC Doctoral Training Partnership [grant number MR/N014294/1].
A.M. received funding from UKRI Innovate UK grant (grant number 10031483). L.S. received funding
for her internship from the Beacon Academy, University of Warwick.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The datasets used and/or analysed during this study are available
from the corresponding author on reasonable request.
Conflicts of Interest: The authors have no conflicts of interest to disclose.
J. Pers. Med. 2024, 14, 443 15 of 19

References
1. MedTechEurope. The European Medical Technology Industry in Figures 2022; MedTechEurope: Brussels, Belgium, 2022.
2. Digital Health News. Healthcare 5.0 Technologies at Healthcare Automation and Digitalization Congress 2022. Available on-
line: https://www.digitalhealthnews.eu/events/6737-healthcare-5-0-technologies-at-healthcare-automation-and-digitalization-
congress-2022 (accessed on 23 January 2023).
3. Tortorella, G.L.; Fogliatto, F.S.; Mac Cawley Vergara, A.; Vassolo, R.; Sawhney, R. Healthcare 4.0: Trends, challenges and research
directions. Prod. Plan. Control 2020, 31, 1245–1260. [CrossRef]
4. Corti, L.; Afferni, P.; Merone, M.; Soda, P. Hospital 4.0 and its innovation in methodologies and technologies. In Proceedings of
the 2018 IEEE 31st International Symposium on Computer-Based Medical Systems (CBMS), Karlstad, Sweden, 18–21 June 2018.
5. Mbunge, E.; Muchemwa, B.; Jiyane, S.e.; Batani, J. Sensors and healthcare 5.0: Transformative shift in virtual care through
emerging digital health technologies. Glob. Health J. 2021, 5, 169–177. [CrossRef]
6. Mohanta, B.; Das, P.; Patnaik, S. Healthcare 5.0: A paradigm shift in digital healthcare system using artificial intelligence, IOT and
5G communication. In Proceedings of the 2019 International Conference on Applied Machine Learning (ICAML), Bhubaneswar,
India, 25–26 May 2019.
7. Korteling, J.E.H.; van de Boer-Visschedijk, G.C.; Blankendaal, R.A.M.; Boonekamp, R.C.; Eikelboom, A.R. Human-versus Artificial
Intelligence. Front. Artif. Intell. 2021, 4, 622364. [CrossRef] [PubMed]
8. Healthcare IT News. 3 Charts Show Where Artificial Intelligence Is Making an Impact in Healthcare Right Now. 2018. Available
online: https://www.healthcareitnews.com/news/3-charts-show-where-artificial-intelligence-making-impact-healthcare-right-
now (accessed on 23 January 2023).
9. Allied Market Research. Artificial Intelligence in Healthcare Market|Global Report—2030. 2021. Available online: https:
//www.alliedmarketresearch.com/artificial-intelligence-in-healthcare-market (accessed on 23 January 2023).
10. Reuter, E. 5 Takeaways from the FDA’s List of AI-Enabled Medical Devices. 2022. Available online: https://www.medtechdive.
com/news/FDA-AI-ML-medical-devices-5-takeaways/635908/ (accessed on 23 January 2023).
11. Bohr, A.; Memarzadeh, K. The rise of artificial intelligence in healthcare applications. In Artificial Intelligence in Healthcare;
Academic Press: Cambridge, MA, USA, 2020; pp. 25–60.
12. A Pro-Innovation Approach to AI Regulation. Available online: https://www.gov.uk/government/publications/ai-regulation-a-
pro-innovation-approach/white-paper (accessed on 10 March 2024).
13. Bini, F.; Franzò, M.; Maccaro, A.; Piaggio, D.; Pecchia, L.; Marinozzi, F. Is medical device regulatory compliance growing as fast as
extended reality to avoid misunderstandings in the future? Health Technol. 2023, 13, 831–842. [CrossRef]
14. FDA. Artificial Intelligence and Machine Learning in Software as a Medical Device. Available online: https://www.fda.
gov/medical-devices/software-medical-device-samd/artificial-intelligence-and-machine-learning-software-medical-device (ac-
cessed on 10 March 2024).
15. FEAM. Summary Report: Digital Health and AI: Benefits and Costs of Data Sharing in the EU. In FEAM FORUM Annual Lecture;
Federation of European Academies of Medicine: Brussels, Belgium, 2022.
16. OECD. OECD Legal Instruments. 2019. Available online: https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-04
49 (accessed on 1 February 2023).
17. Google. Our Principles—Google AI. Available online: https://ai.google/principles/ (accessed on 1 February 2023).
18. ACM. SIGAI—Artificial Intelligence. 2022. Available online: https://www.acm.org/special-interest-groups/sigs/sigai (accessed
on 10 March 2024).
19. Alexia Skok, D. The EU Needs an Artificial Intelligence Act That Protects Fundamental Rights. Available online: https://www.
accessnow.org/eu-artificial-intelligence-act-fundamental-rights/ (accessed on 1 February 2023).
20. Consulting, I. General Data Protection Regulation (GDPR). Available online: https://gdpr-info.eu (accessed on 1 February 2023).
21. Jobin, A.; Ienca, M.; Vayena, E. The global landscape of AI ethics guidelines. Nat. Mach. Intell. 2019, 1, 389–399. [CrossRef]
22. Lemonne, E. Ethics Guidelines for Trustworthy AI. 2018. Available online: https://ec.europa.eu/futurium/en/ai-alliance-
consultation.1.html (accessed on 1 February 2023).
23. European Commission. Assessment List for Trustworthy Artificial Intelligence (ALTAI) for Self-Assessment|Shaping Europe’s
Digital Future. 2020. Available online: https://www.frontiersin.org/articles/10.3389/frai.2023.1020592/full#:~:text=In%20July%
202020,%20the%20European,Principles%20for%20Trustworthy%20AI.%E2%80%9D%20Prior (accessed on 9 February 2023).
24. European Commission. White Paper on Artificial Intelligence: A European Approach to Excellence and Trust; European Commission:
Brussels, Belgium, 2020.
25. UNESCO. UNESCO’s Input in Reply to the OHCHR Report on the Human Rights Council Resolution 47/23 entitled “New and
Emerging Digital Technologies and Human Rights”. 2021. Available online: https://www.ohchr.org/sites/default/files/2022-0
3/UNESCO.pdf (accessed on 15 April 2024).
26. Ethics and Governance of Artificial Intelligence for Health: Guidance on Large Multi-Modal Models. Available online: https:
//www.who.int/publications/i/item/9789240084759 (accessed on 10 March 2024).
27. Hunimed. 6-Year Degree Course in Medicine and Biomedical Engineering, Entirely Taught in English, Run by Humanitas
University in Partnership with Politecnico di Milano. 2022. Available online: https://www.hunimed.eu/course/medtec-school/
(accessed on 3 November 2022).
J. Pers. Med. 2024, 14, 443 16 of 19

28. Uo, P. Nasce Meet: Grazie All’incontro Di Quattro Prestigiosi Atenei, Una Formazione Medica All’altezza Delle Nuove Tecnologie.
2019. Available online: http://news.unipv.it/?p=45400 (accessed on 3 November 2022).
29. Çalışkan, S.A.; Demir, K.; Karaca, O. Artificial intelligence in medical education curriculum: An e-Delphi Study for Competencies.
PLoS ONE 2022, 17, e0271872. [CrossRef] [PubMed]
30. Civaner, M.M.; Uncu, Y.; Bulut, F.; Chalil, E.G.; Tatli, A. Artificial intelligence in medical education: A cross-sectional needs
assessment. BMC Med. Educ. 2022, 22, 772. [CrossRef] [PubMed]
31. Grunhut, J.; Marques, O.; Wyatt, A.T.M. Needs, challenges, and applications of artificial intelligence in medical education
curriculum. JMIR Med. Educ. 2022, 8, e35587. [CrossRef] [PubMed]
32. Skillings, K. Teaching of Biomedical Ethics to Engineering Students through the Use of Role Playing. Ph.D. Thesis, Worcester
Polytechnic Institute, Worcester, MA, USA, 2017.
33. Maccaro, A.; Piaggio, D.; Pagliara, S.; Pecchia, L. The role of ethics in science: A systematic literature review from the first wave of
COVID-19. Health Technol. 2021, 11, 1063–1071. [CrossRef] [PubMed]
34. Maccaro, A.; Piaggio, D.; Dodaro, C.A.; Pecchia, L. Biomedical engineering and ethics: Reflections on medical devices and PPE
during the first wave of COVID-19. BMC Med. Ethics 2021, 22, 130. [CrossRef]
35. Piaggio, D.; Castaldo, R.; Cinelli, M.; Cinelli, S.; Maccaro, A.; Pecchia, L. A framework for designing medical devices resilient to
low-resource settings. Glob. Health 2021, 17, 64. [CrossRef] [PubMed]
36. Di Pietro, L.; Piaggio, D.; Oronti, I.; Maccaro, A.; Houessouvo, R.C.; Medenou, D.; De Maria, C.; Pecchia, L.; Ahluwalia, A. A
Framework for Assessing Healthcare Facilities in Low-Resource Settings: Field Studies in Benin and Uganda. J. Med. Biol. Eng.
2020, 40, 526–534. [CrossRef]
37. Maccaro, A.; Piaggio, D.; Leesurakarn, S.; Husen, N.; Sekalala, S.; Rai, S.; Pecchia, L. On the universality of medical device
regulations: The case of Benin. BMC Health Serv. Res. 2022, 22, 1031. [CrossRef] [PubMed]
38. High-Level Expert Group on Artificial Intelligence. Available online: https://digital-strategy.ec.europa.eu/en/policies/expert-
group-ai (accessed on 10 March 2024).
39. European Council. Artificial Intelligence Act: Council and Parliament Strike a Deal on the First Rules for AI in the World. 2023.
Available online: https://www.consilium.europa.eu/en/press/press-releases/2023/12/09/artificial-intelligence-act-council-
and-parliament-strike-a-deal-on-the-first-worldwide-rules-for-ai/ (accessed on 15 April 2024).
40. Moher, D.; Liberati, A.; Tetzlaff, J.; Altman, D.G.; PRISMA Group. Preferred reporting items for systematic reviews and
meta-analyses: The PRISMA statement. Ann. Intern. Med. 2009, 151, 264–269. [CrossRef] [PubMed]
41. Popay, J.; Roberts, H.; Sowden, A.; Petticrew, M.; Arai, L.; Rodgers, M.; Britten, N.; Roen, K.; Duffy, S. Guidance on the Conduct of
Narrative Synthesis in Systematic Reviews; A Product from the ESRC Methods Programme Version; Citeseer: Online; University of
Lancaster: Lancaster, UK, 2006; Volume 1, p. b92.
42. McLennan, S.; Fiske, A.; Tigard, D.; Müller, R.; Haddadin, S.; Buyx, A. Embedded ethics: A proposal for integrating ethics into the
development of medical AI. BMC Med. Ethics 2022, 23, 6. [CrossRef] [PubMed]
43. Svensson, A.M.; Jotterand, F. Doctor ex machina: A critical assessment of the use of artificial intelligence in health care. J. Med.
Philos. A Forum Bioeth. Philos. Med. 2022, 47, 155–178. [CrossRef] [PubMed]
44. Martinho, A.; Kroesen, M.; Chorus, C. A healthy debate: Exploring the views of medical doctors on the ethics of artificial
intelligence. Artif. Intell. Med. 2021, 121, 102190. [CrossRef] [PubMed]
45. Donia, J.; Shaw, J.A. Co-design and ethical artificial intelligence for health: An agenda for critical research and practice. Big Data
Soc. 2021, 8, 20539517211065248. [CrossRef]
46. Arima, H.; Kano, S. Integrated Analytical Framework for the Development of Artificial Intelligence-Based Medical Devices. Ther.
Innov. Regul. Sci. 2021, 55, 853–865. [CrossRef] [PubMed]
47. Racine, E.; Boehlen, W.; Sample, M. Healthcare uses of artificial intelligence: Challenges and opportunities for growth. In
Healthcare Management Forum; SAGE Publications: Los Angeles, CA, USA, 2019.
48. Guan, J. Artificial intelligence in healthcare and medicine: Promises, ethical challenges and governance. Chin. Med. Sci. J. 2019,
34, 76–83. [PubMed]
49. Quinn, T.P.; Senadeera, M.; Jacobs, S.; Coghlan, S.; Le, V. Trust and medical AI: The challenges we face and the expertise needed
to overcome them. J. Am. Med. Inform. Assoc. 2021, 28, 890–894. [CrossRef] [PubMed]
50. Arnold, M.H. Teasing out artificial intelligence in medicine: An ethical critique of artificial intelligence and machine learning in
medicine. J. Bioethical Inq. 2021, 18, 121–139. [CrossRef] [PubMed]
51. Karmakar, S. Artificial Intelligence: The future of medicine, or an overhyped and dangerous idea? Ir. J. Med. Sci. 2022, 191,
1991–1994. [CrossRef] [PubMed]
52. Montemayor, C.; Halpern, J.; Fairweather, A. In principle obstacles for empathic AI: Why we can’t replace human empathy in
healthcare. AI Soc. 2021, 37, 1353–1359. [CrossRef] [PubMed]
53. Adlakha, S.; Yadav, D.; Garg, R.K.; Chhabra, D. Quest for dexterous prospects in AI regulated arena: Opportunities and challenges
in healthcare. Int. J. Healthc. Technol. Manag. 2020, 18, 22–50. [CrossRef]
54. Ho, A. Deep ethical learning: Taking the interplay of human and artificial intelligence seriously. Hastings Cent. Rep. 2019, 49,
36–39. [CrossRef] [PubMed]
55. Whitby, B. Automating medicine the ethical way. In Machine Medical Ethics; Springer: Cham, Switzerland, 2015; pp. 223–232.
J. Pers. Med. 2024, 14, 443 17 of 19

56. Buruk, B.; Ekmekci, P.E.; Arda, B. A critical perspective on guidelines for responsible and trustworthy artificial intelligence. Med.
Health Care Philos. 2020, 23, 387–399. [CrossRef] [PubMed]
57. de Miguel, I.; Sanz, B.; Lazcoz, G. Machine learning in the EU health care context: Exploring the ethical, legal and social issues.
Inf. Commun. Soc. 2020, 23, 1139–1153. [CrossRef]
58. Johnson, S.L. AI, machine learning, and ethics in health care. J. Leg. Med. 2019, 39, 427–441. [CrossRef] [PubMed]
59. Pasricha, S. Ethics for Digital Medicine: A Path for Ethical Emerging Medical IoT Design. Computer 2023, 56, 32–40. [CrossRef]
60. Reddy, S. Navigating the AI Revolution: The Case for Precise Regulation in Health Care. J. Med. Internet Res. 2023, 25, e49989.
[CrossRef] [PubMed]
61. Zhang, J.; Zhang, Z.-M. Ethics and governance of trustworthy medical artificial intelligence. BMC Med. Inform. Decis. Mak. 2023,
23, 7. [CrossRef] [PubMed]
62. Pruski, M. Ethics framework for predictive clinical AI model updating. Ethics Inf. Technol. 2023, 25, 48. [CrossRef]
63. Schicktanz, S.; Welsch, J.; Schweda, M.; Hein, A.; Rieger, J.W.; Kirste, T. AI-assisted ethics? Considerations of AI simulation for the
ethical assessment and design of assistive technologies. Front. Genet. 2023, 14, 1039839. [CrossRef] [PubMed]
64. Adams, J. Defending explicability as a principle for the ethics of artificial intelligence in medicine. Med. Health Care Philos. 2023,
26, 615–623. [CrossRef] [PubMed]
65. Love, C.S. “Just the Facts Ma’am”: Moral and Ethical Considerations for Artificial Intelligence in Medicine and its Potential to
Impact Patient Autonomy and Hope. Linacre Q. 2023, 90, 375–394. [CrossRef] [PubMed]
66. Couture, V.; Roy, M.C.; Dez, E.; Laperle, S.; Bélisle-Pipon, J.C. Ethical Implications of Artificial Intelligence in Population Health
and the Public’s Role in Its Governance: Perspectives From a Citizen and Expert Panel. J. Med. Internet Res. 2023, 25, e44357.
[CrossRef]
67. Yves Saint James, A.; Stacy, M.C.; Nehmat, H.; Annette, B.-M.; Khin Than, W.; Chris, D.; Lei, W.; Wendy, A.R. Practical, epistemic
and normative implications of algorithmic bias in healthcare artificial intelligence: A qualitative study of multidisciplinary expert
perspectives. J. Med. Ethics 2023. [CrossRef]
68. Chikhaoui, E.; Alajmi, A.; Larabi-Marie-Sainte, S. Artificial Intelligence Applications in Healthcare Sector: Ethical and Legal
Challenges. Emerg. Sci. J. 2022, 6, 717–738. [CrossRef]
69. Cobianchi, L.; Verde, J.M.; Loftus, T.J.; Piccolo, D.; Dal Mas, F.; Mascagni, P.; Garcia Vazquez, A.; Ansaloni, L.; Marseglia, G.R.;
Massaro, M.; et al. Artificial Intelligence and Surgery: Ethical Dilemmas and Open Issues. J. Am. Coll. Surg. 2022, 235, 268–275.
[CrossRef] [PubMed]
70. De Togni, G.; Krauthammer, M.; Biller-Andorno, N. Beyond the hype: ‘Acceptable futures’ for AI and robotic technologies in
healthcare. AI Soc. 2023, 1–10. [CrossRef] [PubMed]
71. Iqbal, J.D.; Krauthammer, M.; Biller-Andorno, N. The Use and Ethics of Digital Twins in Medicine. J. Law. Med. Ethics 2022, 50,
583–596. [CrossRef] [PubMed]
72. Lewanowicz, A.; Wiśniewski, M.; Oronowicz-Jaskowiak, W. The use of machine learning to support the therapeutic process—
Strengths and weaknesses. Postep. Psychiatr. Neurol. 2022, 31, 167–173. [CrossRef]
73. Martín-Peña, R. Does the COVID-19 Pandemic have Implications for Machine Ethics? In HCI International 2022—Late Breaking
Posters; Springer Nature: Cham, Switzerland, 2022.
74. Papadopoulou, E.; Exarchos, T. An Ethics Impact Assessment (EIA) for AI uses in Health & Care: The correlation of ethics and
legal aspects when AI systems are used in health & care contexts. In Proceedings of the 12th Hellenic Conference on Artificial
Intelligence, Corfu, Greece, 7–9 September 2022; p. 14.
75. Pasricha, S. AI Ethics in Smart Healthcare. IEEE Consum. Electron. Mag. 2022, 12, 12–20. [CrossRef]
76. Refolo, P.; Sacchini, D.; Raimondi, C.; Spagnolo, A.G. Ethics of digital therapeutics (DTx). Eur. Rev. Med. Pharmacol. Sci. 2022, 26,
6418–6423. [PubMed]
77. Smallman, M. Multi Scale Ethics—Why We Need to Consider the Ethics of AI in Healthcare at Different Scales. Sci. Eng. Ethics
2022, 28, 63. [CrossRef] [PubMed]
78. De Boer, B.; Kudina, O. What is morally at stake when using algorithms to make medical diagnoses? Expanding the discussion
beyond risks and harms. Theor. Med. Bioeth. 2021, 42, 245–266. [CrossRef] [PubMed]
79. Braun, M.; Hummel, P.; Beck, S.; Dabrock, P. Primer on an ethics of AI-based decision support systems in the clinic. J. Med. Ethics
2021, 47, e3. [CrossRef]
80. Rogers, W.A.; Draper, H.; Carter, S.M. Evaluation of artificial intelligence clinical applications: Detailed case analyses show value
of healthcare ethics approach in identifying patient care issues. Bioethics 2021, 35, 623–633. [CrossRef]
81. Lysaght, T.; Lim, H.Y.; Xafis, V.; Ngiam, K.Y. AI-assisted decision-making in healthcare: The application of an ethics framework
for big data in health and research. Asian Bioeth. Rev. 2019, 11, 299–314. [CrossRef] [PubMed]
82. Astromskė, K.; Peičius, E.; Astromskis, P. Ethical and legal challenges of informed consent applying artificial intelligence in
medical diagnostic consultations. AI Soc. 2021, 36, 509–520. [CrossRef]
83. Fletcher, R.R.; Nakeshimana, A.; Olubeko, O. Addressing fairness, bias, and appropriate use of artificial intelligence and machine
learning in global health. Front. Artif. Intell. 2021, 3, 561802. [CrossRef]
84. Nabi, J. How bioethics can shape artificial intelligence and machine learning. Hastings Cent. Rep. 2018, 48, 10–13. [CrossRef]
[PubMed]
J. Pers. Med. 2024, 14, 443 18 of 19

85. Amann, J.; Blasimme, A.; Vayena, E.; Frey, D.; Madai, V.I.; Consortium, P.Q. Explainability for artificial intelligence in healthcare:
A multidisciplinary perspective. BMC Med. Inform. Decis. Mak. 2020, 20, 310. [CrossRef]
86. Chen, A.; Wang, C.; Zhang, X. Reflection on the equitable attribution of responsibility for artificial intelligence-assisted diagnosis
and treatment decisions. Intell. Med. 2023, 3, 139–143. [CrossRef]
87. Hallowell, N.; Badger, S.; McKay, F.; Kerasidou, A.; Nellåker, C. Democratising or disrupting diagnosis? Ethical issues raised by
the use of AI tools for rare disease diagnosis. SSM Qual. Res. Health 2023, 3, 100240. [CrossRef] [PubMed]
88. Lorenzini, G.; Arbelaez Ossa, L.; Shaw, D.M.; Elger, B.S. Artificial intelligence and the doctor-patient relationship expanding the
paradigm of shared decision making. Bioethics 2023, 37, 424–429. [CrossRef] [PubMed]
89. Cagliero, D.; Deuitch, N.; Shah, N.; Feudtner, C.; Char, D. A framework to identify ethical concerns with ML-guided care
workflows: A case study of mortality prediction to guide advance care planning. J. Am. Med. Inf. Assoc. 2023, 30, 819–827.
[CrossRef] [PubMed]
90. Redrup Hill, E.; Mitchell, C.; Brigden, T.; Hall, A. Ethical and legal considerations influencing human involvement in the
implementation of artificial intelligence in a clinical pathway: A multi-stakeholder perspective. Front. Digit. Health 2023, 5,
1139210. [CrossRef] [PubMed]
91. Ferrario, A.; Gloeckler, S.; Biller-Andorno, N. Ethics of the algorithmic prediction of goal of care preferences: From theory to
practice. J. Med. Ethics 2023, 49, 165–174. [CrossRef] [PubMed]
92. Lorenzini, G.; Shaw, D.M.; Arbelaez Ossa, L.; Elger, B.S. Machine learning applications in healthcare and the role of informed
consent: Ethical and practical considerations. Clin. Ethics 2023, 18, 451–456. [CrossRef]
93. Sharova, D.E.; Zinchenko, V.V.; Akhmad, E.S.; Mokienko, O.A.; Vladzymyrskyy, A.V.; Morozov, S.P. On the issue of ethical aspects
of the artificial intelligence systems implementation in healthcare. Digit. Diagn. 2021, 2, 356–368. [CrossRef]
94. Wellnhofer, E. Real-World and Regulatory Perspectives of Artificial Intelligence in Cardiovascular Imaging. Front. Cardiovasc.
Med. 2022, 9, 890809. [CrossRef] [PubMed]
95. Ballantyne, A.; Stewart, C. Big data and public-private partnerships in healthcare and research: The application of an ethics
framework for big data in health and research. Asian Bioeth. Rev. 2019, 11, 315–326. [CrossRef]
96. Howe, E.G., III; Elenberg, F. Ethical challenges posed by big data. Innov. Clin. Neurosci. 2020, 17, 24.
97. De Angelis, L.; Baglivo, F.; Arzilli, G.; Privitera, G.P.; Ferragina, P.; Tozzi, A.E.; Rizzo, C. ChatGPT and the rise of large language
models: The new AI-driven infodemic threat in public health. Front. Public Health 2023, 11, 1166120. [CrossRef] [PubMed]
98. Liu, T.Y.A.; Wu, J.H. The Ethical and Societal Considerations for the Rise of Artificial Intelligence and Big Data in Ophthalmology.
Front. Med. 2022, 9, 845522. [CrossRef] [PubMed]
99. Fiske, A.; Henningsen, P.; Buyx, A. Your robot therapist will see you now: Ethical implications of embodied artificial intelligence
in psychiatry, psychology, and psychotherapy. J. Med. Internet Res. 2019, 21, e13216. [CrossRef] [PubMed]
100. Steil, J.; Finas, D.; Beck, S.; Manzeschke, A.; Haux, R. Robotic systems in operating theaters: New forms of team–machine
interaction in health care. Methods Inf. Med. 2019, 58, e14–e25. [CrossRef] [PubMed]
101. De Togni, G.; Erikainen, S.; Chan, S.; Cunningham-Burley, S. What makes AI ‘intelligent’and ‘caring’? Exploring affect and
relationality across three sites of intelligence and care. Soc. Sci. Med. 2021, 277, 113874. [CrossRef] [PubMed]
102. Weber, A. Emerging medical ethical issues in healthcare and medical robotics. Int. J. Mech. Eng. Robot. Res. 2018, 7, 604–607.
[CrossRef]
103. Bendel, O. Surgical, therapeutic, nursing and sex robots in machine and information ethics. In Machine Medical Ethics; Springer:
Cham, Switzerland, 2015; pp. 17–32.
104. Shuaib, A.; Arian, H.; Shuaib, A. The Increasing Role of Artificial Intelligence in Health Care: Will Robots Replace Doctors in the
Future? Int. J. Gen. Med. 2020, 13, 891–896. [CrossRef] [PubMed]
105. Boch, A.; Ryan, S.; Kriebitz, A.; Amugongo, L.M.; Lütge, C. Beyond the Metal Flesh. Beyond the Metal Flesh: Understanding the
Intersection between Bio- and AI Ethics for Robotics in Healthcare. Robotics 2023, 12, 110. [CrossRef]
106. Hatherley, J.; Sparrow, R. Diachronic and synchronic variation in the performance of adaptive machine learning systems: The
ethical challenges. J. Am. Med. Inform. Assoc. 2023, 30, 361–366. [CrossRef] [PubMed]
107. Lanne, M.; Leikas, J. Ethical AI in the re-ablement of older people: Opportunities and challenges. Gerontechnology 2021, 20, 1–13.
[CrossRef]
108. Leimanis, A.; Palkova, K. Ethical guidelines for artificial intelligence in healthcare from the sustainable development perspective.
Eur. J. Sustain. Dev. 2021, 10, 90. [CrossRef]
109. Ho, A. Are we ready for artificial intelligence health monitoring in elder care? BMC Geriatr. 2020, 20, 358. [CrossRef] [PubMed]
110. Luxton, D.D. Recommendations for the ethical use and design of artificial intelligent care providers. Artif. Intell. Med. 2014, 62,
1–10. [CrossRef] [PubMed]
111. Smith, M.J.; Bean, S. AI and ethics in medical radiation sciences. J. Med. Imaging Radiat. Sci. 2019, 50, S24–S26. [CrossRef]
[PubMed]
112. Parviainen, J.; Rantala, J. Chatbot breakthrough in the 2020s? An ethical reflection on the trend of automated consultations in
health care. Med. Health Care Philos. 2022, 25, 61–71. [CrossRef] [PubMed]
113. Kühler, M. Exploring the phenomenon and ethical issues of AI paternalism in health apps. Bioethics 2022, 36, 194–200. [CrossRef]
[PubMed]
J. Pers. Med. 2024, 14, 443 19 of 19

114. Kerasidou, A. Ethics of artificial intelligence in global health: Explainability, algorithmic bias and trust. J. Oral Biol. Craniofacial
Res. 2021, 11, 612–614. [CrossRef]
115. Panch, T.; Mattie, H.; Atun, R. Artificial intelligence and algorithmic bias: Implications for health systems. J. Glob. Health 2019, 9,
020318. [CrossRef] [PubMed]
116. Balthazar, P.; Harri, P.; Prater, A.; Safdar, N.M. Protecting your patients’ interests in the era of big data, artificial intelligence, and
predictive analytics. J. Am. Coll. Radiol. 2018, 15, 580–586. [CrossRef] [PubMed]
117. Beauchamp, T.L. Methods and principles in biomedical ethics. J. Med. Ethics 2003, 29, 269–274. [CrossRef] [PubMed]
118. Floridi, L.; Cowls, J.; Beltrametti, M.; Chatila, R.; Chazerand, P.; Dignum, V.; Luetge, C.; Madelin, R.; Pagallo, U.; Rossi, F.; et al.
AI4People—An Ethical Framework for a Good AI Society: Opportunities, Risks, Principles, and Recommendations. Minds Mach.
2018, 28, 689–707. [CrossRef] [PubMed]
119. Sheu, R.K.; Pardeshi, M.S. A Survey on Medical Explainable AI (XAI): Recent Progress, Explainability Approach, Human
Interaction and Scoring System. Sensors 2022, 22, 8068. [CrossRef] [PubMed]
120. Borchert, R.; Azevedo, T.; Badhwar, A.; Bernal, J.; Betts, M.; Bruffaerts, R.; Burkhart, M.C.; Dewachter, I.; Gellersen, H.; Low, A.
Artificial intelligence for diagnosis and prognosis in neuroimaging for dementia; a systematic review. medRxiv 2021. [CrossRef]
121. Bernal, J.; Mazo, C. Transparency of Artificial Intelligence in Healthcare: Insights from Professionals in Computing and Healthcare
Worldwide. Appl. Sci. 2022, 12, 10228. [CrossRef]
122. Burdon, M.; Andrejevic, M. Big data in the sensor society. In Big Data Is Not a Monolith; The MIT Press: Cambridge, MA, USA,
2016; pp. 61–76.
123. GATEKEEPER. Available online: https://www.gatekeeper-project.eu/ (accessed on 22 March 2024).

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.