See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/357635116

Artificial Intelligence in Healthcare

Article · January 2022

CITATION READS
1 2,205

1 author:

James Angle
Trinity Health
15 PUBLICATIONS 5 CITATIONS

SEE PROFILE

All content following this page was uploaded by James Angle on 06 January 2022.

The user has requested enhancement of the downloaded file.

Artificial Intelligence
in Healthcare
The permanent and official location for the Health Information Management Working Group research
is: https://cloudsecurityalliance.org/research/working-groups/health-information-management/

© 2021 Cloud Security Alliance – All Rights Reserved. You may download, store, display on your
computer, view, print, and link to the Cloud Security Alliance at https://cloudsecurityalliance.org
subject to the following: (a) the draft may be used solely for your personal, informational, non-
commercial use; (b) the draft may not be modified or altered in any way; (c) the draft may not be
redistributed; and (d) the trademark, copyright or other notices may not be removed. You may quote
portions of the draft as permitted by the Fair Use provisions of the United States Copyright Act,
provided that you attribute the portions to the Cloud Security Alliance.

© Copyright 2021, Cloud Security Alliance. All rights reserved. 2

Acknowledgments
Lead Author:
Dr. James Angle

Contributors:
Michael Roza

Reviewers:
Saurabh Saraswat
Ashish Vashishtha

CSA Global Staff:

Vince Campitelli
Alex Kaluza
Claire Lehnert (Design)

The CSA’s Health Information Management Working Group aims to directly influence how health
information service providers deliver secure cloud solutions (services, transport, applications, and
storage) to their clients and to foster cloud awareness within all aspects of healthcare and related
industries.

© Copyright 2021, Cloud Security Alliance. All rights reserved. 3

Table of Contents
Table of Contents.................................................................................................................................4
Abstract...............................................................................................................................................5
1. Introduction.....................................................................................................................................5
2. Robotics...........................................................................................................................................6
3. Robotic Process Automation............................................................................................................7
4. Diagnosis and Treatment Applications.............................................................................................8
5. Big Data and Predictive Analytics...................................................................................................10
6. AI and Telehealth............................................................................................................................ 12
7. Bias in AI......................................................................................................................................... 12
8. Use of AI in the fight against COVID-19.......................................................................................... 15
9. Ethical and Legal Challenges........................................................................................................... 15
9.1 Ethical Challenges................................................................................................................... 15
9.2 Legal Challenges.................................................................................................................... 17
10. AI and Cloud Computing.............................................................................................................. 19
11. Conclusion....................................................................................................................................20
References.........................................................................................................................................20

© Copyright 2021, Cloud Security Alliance. All rights reserved. 4

Abstract
The emergence of artificial intelligence (AI) as a tool for better health care offers opportunities to
improve patient and clinical outcomes and reduce costs. As with any new technology entering the
healthcare arena, there are several challenges faced by AI. Some of the common setbacks for AI in
healthcare include a lack of data exchange, regulatory compliance requirements, and patient and
provider adoption. There are several applications in healthcare where AI can deliver an incredible
value, but healthcare delivery organizations (HDO) must evaluate each to determine if and how they
can be adopted. To be truly effective, AI requires multiple data sets and resides in the cloud, which
causes concern about the privacy and security of patient data.

1. Introduction
Advancements in artificial intelligence (AI) and machine learning (ML) offer opportunities to improve
clinical outcomes, reduce cost, and improve population health. AI-related technologies can change
patient care and administrative processes within healthcare delivery organizations (HDO). The ever-
increasing volume and complexity of healthcare data provide an ideal environment for the application
of AI and ML. It is important to remember that AI is not a single technology but a collection of
technologies that perform many tasks. The following are some, but by no means all, of the uses for
AI in healthcare:

• Physical robots that perform not only predefined tasks like lifting, repositioning, or
assembling objects in places, like delivering supplies in hospitals, but also medical-related
tasks.
• Robotic process automation that performs structured digital administrative tasks. Diagnosis
and treatment applications and rule-based systems showed promise for accurately
diagnosing and treating disease.
• Administrative applications that can provide substantial efficiencies, which are needed
in healthcare as the average nurse in the United States spends 25% of their work time on
regulatory and administrative activities1.

Another important area is big data and predictive analytics. AI capabilities in big data are making
way for advanced technological integration with a view to improve and simplify care for all.2 While
it is still in the early stages of development, AI shows promise in its application in telehealth. Cloud
computing is a vital part of the AI ecosystem that provides computing power and a large storage
capacity. Each of these uses has the potential to change and improve the way healthcare is delivered.

1
Davenport, T., & Kalakota, R. (2019). The potential for artificial intelligence in healthcare. Future
Healthcare Journal, 6(2), 94–98. https://doi.org/10.7861/futurehosp.6-2-94
2
Williams, L. (2019, September 5). How AI and Big Data is Transforming the Healthcare Industry.
Techiexpert.
https://www.techiexpert.com/how-ai-and-big-data-is-transforming-the-healthcare-industry/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 5

2. Robotics
In today’s healthcare environment, robotics are used for a multitude of different tasks throughout
HDOs—everything from cleaning rooms to assisting in surgeries. All areas of healthcare are
experiencing a personnel shortage and robots in the medical field are transforming how surgeries are
performed and streamlining supply delivery and disinfection, freeing up people. Robots are used not
only in the operating room but in other clinical settings in support of healthcare workers. When the
COVID-19 pandemic started, robots were deployed for a much wider range of tasks in order to reduce
exposure to pathogens. During this time, it has become clear that the operational efficiencies and
risk reduction provided by health robotics offer value in many areas.

Some examples are:

• Robots can clean and prep patient rooms, helping to limit person-to-person contact in
infectious disease wards.
• Robots with AI-enabled medicine identifier software reduce the time it takes to identify,
match, and distribute medicine to patients in hospitals.
• In the supply chain, robots can undertake not only monotonous and repetitive tasks but also
those that are potentially dangerous for humans.
• In research, robots can be used in dangerous studies, such as the one that was designed
for testing chemical protection clothing for the U.S. military. There are several other ways
robots can be useful in emergency situations that are too dangerous for humans.

One of the first robotic surgery systems was the da Vinci surgical system. While traditional open
surgery relies on a surgeon who operates with handheld instruments through a large incision, today
robotic systems can perform minimally invasive surgery. Da Vinci surgical systems allow for surgery
using a few small incisions, resulting in less recovery time and better patient outcomes.

While da Vinci was the first, today many companies are looking at how to advance robotics in
healthcare. In early 2018, Google’s co-founder Sergey Brin used the robot to suture on some
synthetic tissue. Since then, Johnson & Johnson bought Auris, which develops robotic technologies
focused on lung cancer, and acquired Orthotaxy, a privately held developer of software-enabled
surgical technologies.3

Today’s medical staff can become overwhelmed by the large volume of routine tasks required of
them—tasks that can be performed using robotics. These tasks can include drawing blood, checking
the patient’s vital signs, and taking care of the patient’s hygiene needs. In the future, these tasks
may be performed automatically by a robotic system. As technology advances, robots will perform
more tasks and work more independently. In the near future, robots could take over administrative
or routine tasks while healthcare professionals focus on providing care to the sick and vulnerable.
With proper planning, we can ensure the human touch stays relevant while taking full advantage of
robotics in healthcare.

3
The Medical Futurist. (2019, October 8). Benefits of Robotics in Healthcare: Tasks Medical Robots
Will Undertake. https://medicalfuturist.com/robotics-healthcare/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 6

3. Robotic Process Automation
Strictly speaking, robotic process automation (RPA) is not AI. AI combines cognitive automation,
ML, and natural language processing with reasoning, hypothesis generation, and analysis. RPA
is process driven and AI is data driven. While the use of AI and RPA tools minimize the need for
human intervention, the way in which they automate processes is different. Having said that, they
complement each other well: AI helps RPA automate tasks and handle more complex use cases. RPA
also enables AI insights to be actioned quickly instead of waiting on manual implementations.4

RPA is software robotics, using technology to mimic back-office tasks such as data extraction, filling
out forms, and moving files. It uses application program interfaces (API) and user interfaces (UI) to
integrate and perform routine tasks. This type of rule-based automation frees up humans to perform
more complex tasks.

In an effort to enable their medical staff to concentrate on providing medical care, HDOs are
implementing RPA wherever possible. Automation will fuel new innovative ideas that can streamline
processes and minimize errors. The following are some of the use cases for RPA within healthcare:

• Appointment Scheduling: Patients can book appointments online by simply looking up

the provider and finding an open time. If any changes are required by either the patient or
provider, they can be made online with automatic notification to the other party.
• Account Settlement: anytime a patient enters a healthcare facility for any service, they incur
a fee. With RPA, these fees can be automatically billed to the patient with zero errors.
• Claims Management: Filling out insurance forms can be tedious and time-consuming. With
RPA, the information received during patient check-in can be automatically processed for
this task. Administrative tasks can now be processed in less time with no errors. This can
make this process a more pleasant patient experience.
• Patient Instructions: After a patient encounter, regardless of what the encounter is for, the
patient will receive instructions. RPA can print the instructions and notify the patient about
any appointments, lab tests, or imaging, and the provider’s instructions.
• Audit Procedures: Regardless of the type of audit, HDOs are constantly being audited. RPA
can produce the data required by having an effective file and data retrieval system. All of the
tasks for the audit can be automated, reducing the burden on the administrative staff.5

There are numerous benefits to RPA in a healthcare environment. These automated processes can
simplify many complex procedures that are repetitive and require many man hours to complete.

4
IBM Cloud Education. (2020, October 22). What is Robotic Process Automation. IBM. https://www.
ibm.com/cloud/learn/rpa
5
Mehta, S. (2021, May 31). Robotic Process Automation (RPA) in Healthcare Industry.
Zenesys. https://www.zenesys.com/blog/robotic-process-automation-rpa-in-healthcare-
industry?source=techstories.org

© Copyright 2021, Cloud Security Alliance. All rights reserved. 7

 No Errors Personalized Higher Reduced
Experience Satisfaction Workload

Figure 1. Benefits of RPA6

Accuracy and compliance are critical in the healthcare industry. HDOs use RPA to optimize
information management and insurance claim processing. These are just a few examples of how RPA
improves accuracy across an organization and increases profitability. Allowing RPA to handle tedious
and routine business processes will enhance clinical workflows and free staff, allowing more time to
treat patients.

4. Diagnosis and Treatment Applications

A medical diagnosis is the process of determining the cause of an illness or condition by examining
information acquired from various sources. Diagnosing diseases correctly is the very foundation
of good healthcare and a critical step in treating a patient. However, diagnosing an illness is a
complicated process requiring resources in time and effort. Given the complexity of the process, it
can be prone to errors.7 According to the Agency for Healthcare Research and Quality, 10% of patient
deaths are a direct consequence of misdiagnosis. By using AI and ML, healthcare service providers
can improve the precision of each diagnosis. Medical diagnostics using AI and ML are rapidly
expanding where AI and MI are helping detect life-threatening conditions in their earliest stages.

Rapid developments in AI, ML, and data mining allow technology and healthcare innovators to
create intelligent systems to optimize and improve the diagnosis process. In diagnosing, ML and
data mining algorithms are particularly useful. They can quickly capture unforeseen patterns within
complex and large datasets.8 The field of medical diagnosis is full of possibilities and advantages
including cost-cutting, early diagnosis, and potentially saving human lives. The following are some of
the use cases:

1. In oncology, ML helps oncologists detect the disease at its earliest stages. In addition
to pinpointing the tumor, ML can accurately determine if it’s malignant or benign in
6
Mehta, S. (2021, May 31). Robotic Process Automation (RPA) in Healthcare Industry.
Zenesys. https://www.zenesys.com/blog/robotic-process-automation-rpa-in-healthcare-
industry?source=techstories.org
7
Gagliano M., Pham J., Tang B., Kashif H., & Ban J. (October 2017). Applications of Machine Learning
in Medical Diagnosis. ResearchGate. https://www.researchgate.net/publication/321151498
8
IBM Cloud Education. (2020, October 22). What is Robotic Process Automation. IBM. https://www.
ibm.com/cloud/learn/rpa

© Copyright 2021, Cloud Security Alliance. All rights reserved. 8

 milliseconds. Although computer-based predictions aren’t error-free, the accuracy of
classification is impressive at 88%.9 ML can aid oncological diagnosis and treatment by:
a. Improving the precision of blood and culture analysis.
b. Mapping the diseased cells and flagging areas of interest.
c. Creating tumor staging paradigms.
d. Improving healthcare professionals’ productivity.
2. In dermatology, artificial intelligence is used to improve clinical decision-making and ensure
the accuracy of skin disease diagnoses. ML can aid dermatological diagnosis and treatment by:
a. Using algorithms that separate melanomas from benign skin lesions.
b. Employing tools that track the development and changes in skin moles.
c. Utilizing algorithms that pinpoint biological markers for acne, nail fungus, and
seborrheic dermatitis.
3. AI has helped geneticists progress significantly in the transcription of human genes.
ML and AI technologies are key in preventive genetics. Scientists rely on algorithms to
determine how drugs, chemicals, and environmental factors influence the human genome.
Additionally, geneticists are hopeful that they will be able to improve the efficiency of
gene editing, changing DNA fragments to protect a fetus from the impact of a mutation or
reverse its effect.
4. Mental health disorders are one of the costliest conditions to manage in the United States.
According to The National Institute of Mental Health, 1 in 5 adult Americans is affected by a
mental health disorder. The impact of leaving these conditions untreated or misdiagnosed
can be disastrous. AI can impact mental health research and the efficiency of medical
diagnosis through ML. The top applications of innovative technologies in the field are:
a. Cognitive behavior therapy fueled by chatbots and virtual therapists.
b. Mental health disease prevention by using tools that help high-risk groups avoid
social isolation.
c. Identifying groups with a high risk of suicide.
d. Early detection of mental health disorders.
5. Neuroscience and neurology benefit from implementing artificial intelligence in gathering,
processing, and interpreting research data, from analyzing scans to providing insights about
the human brain and detecting behavior patterns.
6. Technological algorithms support stroke prediction and recovery monitoring, helping
differentiate between resting and stroke-related paralysis, predicting the recovery curve of
ischemic stroke patients for over 90 days, and monitoring early-onset stroke patients for 48
hours after hospitalization.
7. AI supports research on degenerative conditions which is useful for identifying complex
disorders and aiding in optimized interventions.
8. AI can also assist in the diagnosis of Ophthalmology conditions. Some of the latest
innovations that these healthcare centers have adopted are:
a. AI-driven vision screening programs.
b. Identifying diabetic retinopathy and providing physicians with treatment insights.
c. Early-stage diagnosis of macular degeneration.

9
Zhang, Q., Xiao, Y., Suo, J., Shi, J., Yu, J., Guo, Y., Wang, Y., & Zheng, H. (2017). Sonoelastomics
for Breast Tumor Classification: A Radiomics Approach with Clustering-Based Feature Selection on
Sonoelastography. Ultrasound in Medicine & Biology, 43(5), 1058–1069. https://doi.org/10.1016/j.
ultrasmedbio.2016.12.016

© Copyright 2021, Cloud Security Alliance. All rights reserved. 9

 d. High-precision glaucoma and cataract screening.
9. Type I and type II diabetes are widespread conditions, and researchers have a tremendous
amount of data at their disposal. Scientists need to focus on consolidating these insights and
putting them into a single framework. ML can aid diagnosis and treatment of diabetes by:
a. Using vector machine modeling and building neural networks for pre-diabetes screening.
b. Creating tools for managing personalized insulin delivery.
c. Predicting treatable complications in diabetes patients.
d. Identifying genetic and other biomarkers for diabetes.
10. AI helps ICU doctors with judgment calls, such as when to extubate a patient— premature
extubation or prolonged intubation has a strong link to raising ICU mortality rates.
Additionally, MI can identify high-risk patients to aid in monitoring for deterioration signs.
11. AI supports early disease detection in rare diseases such as breast cancer by interpreting
mammogram data efficiently and Alzheimer’s disease using various automated systems and
tools.

AI allows healthcare professionals to increase the scale of medical diagnoses using ML and shift from
analyzing individual cases to monitoring communities and predicting disease outbreaks.10

5. Big Data and Predictive Analytics

Healthcare has become a data-driven environment with a multitude of data sources. Data is available
from medical records, insurance companies, medical devices, and wearable Internet of Medical
Things (IoMT), just to name a few data sources. With advances in AI, MI, and predictive analytics
this provides HDOs the ability to transform this data into meaningful and actionable information.
With predictive analytics, HDOs can make better decisions about patient care requirements and
optimize resources to provide better services and reduce cost. Healthcare data, including both static
and dynamic data sources, is largely unstructured. Using big data analytics and AI, this data can be
turned into actionable data that provides insight vital to patient care. This technology also provides
a great potential for improving population health management by analyzing disease patterns and
tracking disease outbreaks.11

Applications of big data analytics can improve patient-based service, detect spreading diseases
such as COVID-19, generate new insights into disease mechanisms, and provide better treatment
methods. Data mining techniques can aid in identifying and revealing disease monitoring and health-
based trends. Big data analytics can provide advantages to HDOs by digitizing, combining, and
effectively using big data to realize significant benefits in the delivery of healthcare. The benefits
realized can include early disease detection, allowing for treatment options and potentially better
outcomes and effectively managing population-specific health issues.12

10
Usachev, D. (2021). 10 Use Cases of Machine Learning for Medical Diagnosis. Fayrix. https://fayrix.
com/blog/ten-use-cases-of-machine-learning-for-medical-diagnosis
11
Mehta, N., Pandit, A., & Shukla, S. (2019). Transforming healthcare with big data analytics and
artificial intelligence: A systematic mapping study. Journal of Biomedical Informatics, 100, 103311.
https://doi.org/10.1016/j.jbi.2019.103311
12
Cloud Security Alliance. (2020, June 16). Telehealth Data in the Cloud. https://cloudsecurityalliance.
org/artifacts/telehealth-data-in-the-cloud/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 10

HDOs are struggling with the ever-increasing volume and types of data. To successfully manage and
make use of this data, HDOs must overcome talent, infrastructure, and security challenges. While
there is support at the leadership level for the use of AI, implementation is challenging. However,
there are multiple problems with AI implementation, ranging from integrating insights into existing
clinical workflows to consolidating fragmented data to achieving clinically reliable, clean data.

Talent shortages present another roadblock for AI adoption. Hiring and training data science expertise
lies at the heart of implementation challenges. Most HDOs found this to be a major challenge for their
organization. However, with data maturity comes the opportunity for HDOs to take advantage of AI
and ML to increase revenue, improve patient outcomes, and enhance patient experience.13

When implemented, predictive analytics can be used by HDOs to estimate the likelihood of a future
outcome based on patterns in the historical data. This allows HDO staff to receive alerts about
potential events before they happen and therefore make more informed choices. The following are
some ways healthcare organizations are deploying predictive capabilities across the enterprise to
extract actionable, forward-looking insights from their growing data assets:14

1. Patients predictions for improved staffing;

2. Developing precision medicine and new therapies;
3. Real-time alerting;
4. Enhancing patient engagement;
5. Using health data for informed strategic planning;
6. Medical diagnosis;
7. Telemedicine;
8. Medical imaging;
9. Advanced risk and disease management; and
10. Mental health interventions.

While this is not a complete list, it does give a good representation of how big data analytics and AI
can improve patient care and improve outcomes. In addition to this, big data analytics and AI have
proven to be of value in life sciences and medical research. Because of its immense potential for
improving the quality of healthcare, these emerging techniques have attracted increasing attention
in healthcare research and practice.

In today’s digital world, where everyone has a smart device or computer at their fingertips, the
healthcare industry is tapping into this potential with AI-driven health apps. Patients can manage
their condition using health apps facilitated by virtual assistance.15 This influx of data adds to a
growing pool that HDOs can use for predictive analytics to improve patient outcomes.
13
Forrester Consulting. (February 2020). The Cloud, Data, And AI Imperative For Healthcare:
Hospital Readiness And Adoption Paths For Next-Generation Technologies. https://www.ibm.com/
downloads/cas/GDMPZQJE
14
Bresnick, J. (2018a, September 4). 10 High-Value Use Cases for Predictive Analytics in Healthcare.
Health IT Analytics. https://healthitanalytics.com/news/10-high-value-use-cases-for-predictive-
analytics-in-healthcare
15 Williams, L. (2019, September 5). How AI and Big Data is Transforming the Healthcare Industry.
Techiexpert. https://www.techiexpert.com/how-ai-and-big-data-is-transforming-the-healthcare-
industry/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 11

6. AI and Telehealth
Currently, the COVID-19 response relies heavily on social distancing measures to fight the pandemic.
For healthcare systems, telehealth (the use of information and communication technologies
to access and manage healthcare services while video conferencing) has emerged as a critical
technology for safe and efficient communications between healthcare providers and patients.
According to a new World Health Organization (WHO) policy, telemedicine should be one of the
alternative models for clinical services and clinical decision support (within the optimizing service
delivery action).16

As technology continues to advance, AI has become an everyday reality. Telehealth is one of the
newest industries to utilize AI widely, from the distribution of electronic medical cards to personal
consultations. As telehealth expands across the US today, AI is one of the most powerful forces
evolving it. Telehealth services benefited patients and healthcare providers who reduced their risk of
exposure to COVID-19 by decreasing the number of patients they saw in person. However, despite
the convenience of receiving telehealth services, patients have significant concerns about privacy
and security, the physician-patient relationship, and the ability to receive the right diagnosis and
proper treatment.17

7. Bias in AI
The use of AI has fostered improvement in diagnosis and treatments. Diagnosis has moved from
clinical examination-based to evidence-based, supported by the doctor’s experience and skill.In the
last few years, AI has begun to play a supportive role to doctors when analyzing evidence. With
the use of AI in telemedicine, it is easier for doctors to analyze, screen, and diagnose conditions
remotely. AI diagnosis can potentially outperform humans, executing instructions in a mathematical
fashion with a reliance on built-in logic moderated by statistical evidence extracted by ML methods
from large-scale datasets.

Remote monitoring is also being used for distant surveillance of patients with chronic disease, such
as chronic heart failure, chronic obstructive pulmonary disease (COPD), and diabetes mellitus, and
is another area of telehealth where AI has improved patient outcomes. Remote monitoring involves
data acquisition using a sensor, transmission of data to the clinician, integration of data with other
data describing the state of the patient, and determination of the appropriate action in the care of
the patient. AI systems for telemonitoring depend on and expand the scope of other health system
information and communications technology components.18

16
Cloud Security Alliance. (2020, June 16). Telehealth Data in the Cloud. https://cloudsecurityalliance.
org/artifacts/telehealth-data-in-the-cloud/
17
David, C. (2020, December 3). How AI is Transforming Telehealth. Ai4. https://ai4.io/how-ai-is-
transforming-telehealth
18
Kuziemsky C., Maeder A., John O., Gogia S., Basu A., Meher S., & Ito M. (2019). Role of Artificial
Intelligence within the Telehealth Domain: Official 2019 Yearbook Contribution by the members of
IMIA Telehealth Working Group. International Medical Informatics Association Yearbook of Medical
Informatics. https://doi.org/10.1055/s-0039-1677941

© Copyright 2021, Cloud Security Alliance. All rights reserved. 12

In the age of IoMT, ultrafast connectivity means a diverse range of medical devices and equipment
are connected. As a result, telemedicine can make use of real-time data for remote healthcare.
Patient wearables track patient vitals throughout the day and transmit data to the cloud for easy,
ongoing evaluation by their doctor. Some medical apps focus on general wellness and health, while
some apps are used for the diagnosis and treatment of diseases. For example, apps like Runkeeper
and MyFitnessPal track different health metrics for informational purposes, while others use apps
to track specific health metrics sent to providers to use in developing and monitoring the patient’s
treatment plan.19

While the benefits of AI are clear, they are not without risk. The risks inherent in AI used in the
medical arena and relevant to any discussion of AI include: bias, ethics, security, and privacy.

Bias in medical systems can contribute to harmful patient outcomes. So, what is bias and why do
we want to manage it? When talking about individual bias it is a preconceived tendency, inclination,
feeling, or opinion. Statistically, it is a systematic distortion of a statistic as a result of a sampling
procedure. It results in a degree to which the result deviates from the truth. AI bias is an anomaly
in the output of AI algorithms. Bias can contribute to harmful patient outcomes as it can result
in differential treatment. The presumption is that bias is present throughout AI systems and the
challenge is identifying, measuring, and managing it.

There are two reasons for AI bias:

1. Cognitive bias: Feelings toward a person or group based on group membership. Cognitive
bias is a term that describes the tendency for people’s feelings and experiences to affect
their judgment. When it comes to AI decisions, a prominent cognitive bias is the availability
heuristic, which states that human beings tend to rely on information that supports their
current beliefs.20
a. Cognitive bias can get into ML algorithms by either:
i. Designers unknowingly introducing them into the algorithm.
ii. Use of a training data set that includes biases.
2. Lack of complete data: When data is incomplete, it may not be representative and
include bias.21

19
Peck A.A. (2021, February 23). The Future is Digital Healthcare. The National Law Review, XI (54).
https://www.natlawreview.com/article/future-digital-healthcare
20
Fallmann, D. (2021, August 19). Human Cognitive Bias And Its Role In AI. Mindbreeze InSpire.
https://inspire.mindbreeze.com/blog/human-cognitive-bias-and-its-role-in-ai
21
Dilmegani, C. (2021, October 8). Bias in AI: What it is, Types & Examples of Bias & Tools to fix it. AI
Multiple. https://research.aimultiple.com/ai-bias/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 13

 Real world patterns of health
Discriminatory Data
inequality and discrimination

Sampling biases and lack of

Unequal access and resource allocation
representative datasets

Discriminatory Patterns of bias and discrimination

healthcare processes baked into data distributions

Biased clinical
decision making World Data

Biased AI design and

Application injustices Use Design
deployment practices

Disregarding and deepening Power imbalances in agenda

digital divides setting and problem formulation

Exacerbating global health inequality and Biased and exclusionary design,

rich-poor treatment gaps model building and testing practices

Hazardous and discriminatory repurposing Biased deployment, explanation

of biased AI systems and system monitoring practices

Figure 2. Bias in AI22

The question, then, is how we remove unwanted bias from AI algorithms. It is important to
understand that removing unwanted bias does not guarantee fairness. Knowing this, there are some
steps you can take to reduce bias:

1. Ensure you understand the algorithm and data being assessed. Injecting diversity into your
design team can support the role of cultural sensitivity in the decision-making process.
2. Test all data for bias. There are methods for measuring and mitigating unwanted bias.
3. As you identify bias, look for how you can improve the process. Through training, design,
and cultural changes, you can improve the process to reduce bias.
4. Maintaining a diverse AI team can help you mitigate unwanted AI biases. Improving diversity
in the AI community throughout the lifecycle of an algorithm is critical and will enable the
identification of biases

The proliferation of modeling and predictive approaches based on data-driven and MI techniques has
helped to expose various social biases baked into real-world systems, and there is increasing evidence
that the general public has concerns about the societal risks of AI. Identifying and addressing biases
early in the problem formulation process is an important step to improving the process.

22
Source: British Medical Journal

© Copyright 2021, Cloud Security Alliance. All rights reserved. 14

8. Use of AI in the Fight Against
COVID-19
The COVID-19 pandemic has provided us with a unique opportunity to prove that technologies like AI
and ML could be leveraged for the benefit of all.

From the initial reports of the pandemic in December 2019 and the early predictions of its spread and
impact–by a team of scientists who presented their results at the AAAS Annual Meeting in February
2020–to the deployment of AI in the development of vaccines, AI has played a central role in the
fight against COVID-19. ML is helping researchers and practitioners to not only analyze large volumes
of data to forecast the spread of COVID-19, but also to utilize this data and generate models to
formulate an early warning system for future pandemics and to identify vulnerable populations.

AI-based technologies are at the forefront of many efforts by governments and private institutions to
track the virus and create valuable insights based on the past 18 months of recorded data.

AI-based algorithms are being used in various ways, including:

• Optimizing healthcare resources:

• Prioritizing hospital resource allocations—medical staff, hospital beds, ventilators, etc.
• Setting up medical centers in epicenters.
• In vaccine development, enhancements to booster shots based on available datasets.
• In vaccine distribution, models for faster and better reach to high impact areas.
• Supporting contact tracing to help prevent further spread in the community, by analyzing
cell phone data to trace those who have been in contact with or proximate to individuals
diagnosed with COVID-19.

9. Ethical and Legal Challenges

AI is digital technology that will have a significant impact on healthcare in the near future, and AI
applications have the potential to transform the patient-clinician relationship. The complexity and
capabilities of AI make it unique and controversial and the use of AI raises fundamental questions
about what role AI plays in healthcare, what AI systems should do, and what risks are involved. Key
issues involve the ethical and legal challenges posed by AI in healthcare.

9.1 Ethical Challenges

Ethics is a complex and complicated concept. This discussion aims not to define the concept of
ethics but rather to identify some of the ethical issues related to AI. By identifying some of the
issues, we can start to understand and address the issues. Ethical issues related to AI include:

© Copyright 2021, Cloud Security Alliance. All rights reserved. 15

 1. Informed Consent of Use: Is it the clinician’s responsibility to educate the patient around the
complexities of AI, including the form(s) of ML used by the system, the kind of data inputs,
and the possibility of biases or other shortcomings in the data that is being used? Does the
clinician even have a responsibility to notify the patient that AI is being used?23
2. Transparency and Safety
a. Recent AI research has focused on ML using past examples to develop an internal
model of a problem and using that model to make inferences about new patients.
ML systems can make diagnoses based on data for managing patients and improve
patient outcomes. In clinical decision support systems, rules are authored by experts

and have a defined provenance, which is ideally evidence-based. ML, on the other
hand, is dependent on the data on which it has been trained. When patients are not
adequately represented by the data that the ML system was trained on, the accuracy
of the system is compromised.24
b. In order to ensure safety and gain patient confidence, there must be a fair measure of
transparency. AI developers need to be transparent regarding the data used and any
software issues, such as bias.
3. Data Privacy: A study from the University of California–Berkeley says that advances in AI
have rendered the Health Insurance Portability and Accountability Act of 1996 (HIPAA)
obsolete, since HIPAA does not cover tech companies—HIPAA only protects patient health
data when it comes from organizations that provide healthcare services. HIPAA also fails to
regulate genetics testing companies, which are selling customer data to pharma and biotech
firms. While genetic testing does not use AI, this example demonstrates the potential for
harm to privacy rights when developing technological solutions. Genetics testing companies
don’t just tell you about your ancestry. They can also provide valuable information about
your genetic predisposition to certain health risks. In 2017, 23andMe received regulatory
approval to analyze their customers’ genetic information for risk of ten diseases, including
celiac, late-onset Alzheimer’s, and Parkinson’s. One of the uses of this data could be
insurance companies who might use this predictive genetic testing to bias selection
processes and charge higher premiums. The issue here is how we ensure we use the data in
a way that alleviates the privacy concerns.25

23
Gerke S., Minssen T., & Glenn Cohen I. (2020, May 26). Ethical and Legal Challenges of Artificial
Intelligence-Driven Healthcare. Artificial Intelligence in Healthcare, 1st edition, Adam Bohr, Kaveh
Memarzadeh (eds.). https://ssrn.com/abstract=3570129
24
Challen, R. (2019, July 1). Emerging Safety Issues in Artificial Intelligence | PSNet. EPSRC Centre
for Predictive Modelling in Healthcare and Department of Mathematics, Living Systems Institute
University of Exeter. https://psnet.ahrq.gov/perspective/emerging-safety-issues-artificial-
intelligence
25
Kulkarni A. (2021, February 10). AI in Healthcare: Data Privacy and Ethics Concerns. Lexalytics.
https://www.lexalytics.com/lexablog/ai-healthcare-data-privacy-ethics-issues

© Copyright 2021, Cloud Security Alliance. All rights reserved. 16

9.2 Legal Challenges
Many of the ethical issues discussed have a direct
correlation to the legal issues associated with the
use of AI. The following are challenges directly
associated with the legal issues:26

1. Safety: It is very important that black-

box medicine, the use of opaque
computational models to make decisions
related to health care, operate safely.
Techniques such as ML do not easily
lend themselves to human concepts of
explanation and significance. ML outputs 26

are typically probabilistic and sometimes

inscrutable.27 The question is how HDOs can be sure the black-box software they are using
is safe. In the United States, the Food and Drug Administration (FDA) regulates AI in the
healthcare system and has asserted this ability and intent.28

2. Liability: If an AI system gives an incorrect recommendation and a clinician adopts those

recommendations and harms the patient, who is responsible? Is the clinician responsible
for acting on the recommendations or is the developer responsible? Under theories of strict
liability, a developer may be held liable for defects in their AI. At least for the near term, the
AI itself will probably not be liable for its acts or omissions. As a result, those involved in the
process (the developer and provider) will likely have exposure to liability associated with the
AI. Whether the liability is professional liability or product liability will likely depend on the
functions the AI is performing.29 How we set liability depends on if we are trying to deter and
compensate or simply compensate. The compensation approach only could mirror vaccine
compensation in the United States, where developers pay into a fund and the system
collectivized the risk by paying out to those harmed.30

3. Data Protection and Privacy: AI requires enormous amounts of data to analyze in order to
spot patterns, learn and make conclusions, and produce insights that are entirely new. While
26
Retrieved from www.traverselegal.com
27
Ordish J. (2018, August). Legal Liability for Machine Learning in Healthcare. PHG Foundation.
https://www.phgfoundation.org/briefing/legal-liability-machine-learning-in-healthcare
28
Nicholson Price II W. (2017). Artificial Intelligence in Health Care: Applications and Legal Issues.
American Bar Association. https://www.americanbar.org/groups/science_technology/publications/
scitech_lawyer/2017/fall/artificial-intelligence-health-care/
29
Davis Jr., W. K., Francois, A., & Camin Murray, C. (2021, February 16). Top Ten Legal Considerations
for Use and/or Development of Artificial Intelligence in Health Care. The National Law Review, XI
(47). https://www.natlawreview.com/article/top-ten-legal-considerations-use-andor-development-
artificial-intelligence-health
30
Gerke S., Minssen T., & Glenn Cohen I. (2020, May 26). Ethical and Legal Challenges of Artificial
Intelligence-Driven Healthcare. Artificial Intelligence in Healthcare, 1st edition, Adam Bohr, Kaveh
Memarzadeh (eds.). https://ssrn.com/abstract=3570129

© Copyright 2021, Cloud Security Alliance. All rights reserved. 17

 there is cutting-edge technology that can transform the healthcare industry, the technology
needed is not always cutting-edge— sometimes it is the simple ability to access patient
records, make appointments, and locate equipment. Cutting-edge and routine technologies
have a need for data, and the use of this data comes with the obligation to protect the data.
AI poses unique challenges and risks with respect to privacy breaches and cybersecurity
threats, which have an obvious negative impact on patients and HDOs. Ensuring privacy for
all data will require data privacy laws and regulation be updated to include data used in AI
and ML systems. Privacy laws need to be consistent and flexible to account for innovations
in AI and ML. Current regulations have not kept up with changes in technology. HIPAA calls
for deidentification of data; however, technology today can link de-identified data resulting
in identification.

4. Data Security: Security presents AI with a new set of challenges, compounded by the fact
that most algorithms require access to massive datasets. Moving large amounts of data
between systems is new to most HDOs, which are becoming ever more sensitive to the
possibility of data breaches. Storing large datasets for AI makes a very attractive target for
hackers.31 Security will always be paramount; however, stakeholders must get more familiar
with the challenges and opportunities of the data sharing that is vital for AI to flourish in a
healthcare IT ecosystem. In order to protect healthcare AI and ML devices, there are a few
simple steps the HDO can take: 1) Ensure access control is in place and functioning with
multi-factor authentication. 2) Incorporate anomaly detection in the endpoint security to
identify unusual activity. 3) In addition to the previous steps, the HDO should create and
deploy a hardened, secure device.

5. Intellectual Property Laws: These laws present another set of challenges for the
development of black-box medicine. Turning AI and big data into safe and effective products,
services, and processes is expensive. HDOs invest large sums of money in developing
black-box algorithms. How can they protect that investment? Ideally, intellectual property
laws would protect the developers of AI and ML, but intellectual property is relatively poor
for black-box medicine. Patents are a good choice to protect technological innovation,
but patents do not provide strong protections for black-box medicine.32 The question of
how HDOs protect their investment in AI is still a significant issue that needs to be better
defined.

As discussed above, there are many ethical and legal challenges to implementing AI and ML
technologies. The primary ethical challenges are:

1. Informed consent;
2. Transparency and safety; and
3. Data privacy.

31
​ resnick, J. (2018b, September 17). Arguing the Pros and Cons of Artificial Intelligence in
B
Healthcare. Health IT Analytics. https://healthitanalytics.com/news/arguing-the-pros-and-cons-of-
artificial-intelligence-in-healthcare
32
Nicholson Price II W. (2017). Artificial Intelligence in Health Care: Applications and Legal Issues.
American Bar Association. https://www.americanbar.org/groups/science_technology/publications/
scitech_lawyer/2017/fall/artificial-intelligence-health-care/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 18

The primary legal challenges are:

1. Safety;
2. Liability;
3. Data protection and privacy;
4. Cybersecurity; and
5. Intellectual property.

It is crucial that all stakeholders work together on these challenges to ensure that AI will be
successfully implemented in a way that is ethical and legal.33

10. AI and Cloud Computing

The age of AI technology is already upon us, taking over tasks that were once performed by
healthcare providers. AI is working in conjunction with cloud computing to bring about major
transformations in healthcare, which has become data-driven and is looking to AI for the future of
data management and analysis.34 AI and cloud computing make the perfect match for data-driven
innovation in healthcare. AI’s cognitive capabilities and ML thrive on large volumes of data, which are
scalable and quickly accessible in a cloud environment.

HDOs are using AI and ML coupled with cloud computing to improve patient outcomes as well as
reduce costs.35 The use of cloud technology with AI can help extract meaningful insights from big
data sets that are now available to HDOs. The analysis of large data sets required for effective AI
and ML are now possible using the cloud. This ability will push AI advancements as they incorporate
medical devices, telehealth, IoMT, devices, and big data. This will enable HDOs to better manage
chronic illnesses and identify potential health issues based on data analysis using AI and ML.

As with any cloud implementation, there are privacy and security concerns that need to be
addressed. HDOs cannot lose sight of the shared responsibility model used on cloud computing.
This requires they assess the risk and ensure controls are in place to mitigate these risks.

33
Gerke S., Minssen T., & Glenn Cohen I. (2020, May 26). Ethical and Legal Challenges of Artificial
Intelligence-Driven Healthcare. Artificial Intelligence in Healthcare, 1st edition, Adam Bohr, Kaveh
Memarzadeh (eds.). https://ssrn.com/abstract=3570129
34
Data Pillar. (2019, October 4). Why Cloud is Important for AI Initiative. https://datapillar.ai/why-
cloud-is-important-for-ai-initiative/
35
Langlois L. (2021, May 25). Artificial Intelligence in Cloud Computing: The Secret to Streamlining
Healthcare., Providertech. https://www.providertech.com/artificial-intelligence-in-cloud-computing/

© Copyright 2021, Cloud Security Alliance. All rights reserved. 19

11. Conclusion
Data are critical for delivering evidence-based health care, which makes it critical for any AI algorithm.
Without data, the process and outcomes are unknown. The availability of data has been greatly
increased with the use of cloud computing and has turned healthcare into a data-rich environment.
AI now has the potential to be integrated into all aspects of healthcare, from management to
delivery and diagnosis. The use of cloud technology coupled with AI enables HDOs to extract more
meaningful insights from the massive amounts of health data. The future of healthcare is certainly
linked to the advancements occurring in AI cloud computing.

While there have been significant developments in AI, there remain some issues that must be
addressed, chief among which are the ethical and legal issues that need to be solved. Governments
will need to introduce regulations that ensure the proper use of AI and the related data sets, which
will help provide patients with some measure of assurance that AI and ML used in healthcare are safe
and the patient’s data is protected. AI and ML in healthcare have a truly bright future.

References
Bresnick, J. (2018a, September 4). 10 High-Value Use Cases for Predictive Analytics in Healthcare.
Health IT Analytics. https://healthitanalytics.com/news/10-high-value-use-cases-for-predictive-
analytics-in-healthcare

Bresnick, J. (2018b, September 17). Arguing the Pros and Cons of Artificial Intelligence in Healthcare.
Health IT Analytics. https://healthitanalytics.com/news/arguing-the-pros-and-cons-of-artificial-
intelligence-in-healthcare

Challen, R. (2019, July 1). Emerging Safety Issues in Artificial Intelligence | PSNet. EPSRC Centre
for Predictive Modelling in Healthcare and Department of Mathematics, Living Systems Institute
University of Exeter. https://psnet.ahrq.gov/perspective/emerging-safety-issues-artificial-
intelligence

Cloud Security Alliance. (2020, June 16). Telehealth Data in the Cloud. https://cloudsecurityalliance.
org/artifacts/telehealth-data-in-the-cloud/

Cloud Security Alliance. (2021, July 21). Healthcare Big Data in the Cloud. https://
cloudsecurityalliance.org/artifacts/healthcare-big-data-in-the-cloud/

Data Pillar. (2019, October 4). Why Cloud is Important for AI Initiative. https://datapillar.ai/why-cloud-
is-important-for-ai-initiative/

Davenport, T., & Kalakota, R. (2019). The potential for artificial intelligence in healthcare. Future
Healthcare Journal, 6(2), 94–98. https://doi.org/10.7861/futurehosp.6-2-94

David, C. (2020, December 3). How AI is Transforming Telehealth. Ai4. https://ai4.io/how-ai-is-

transforming-telehealth

© Copyright 2021, Cloud Security Alliance. All rights reserved. 20

Davis Jr., W. K., Francois, A., & Camin Murray, C. (2021, February 16). Top Ten Legal Considerations
for Use and/or Development of Artificial Intelligence in Health Care. The National Law Review, XI
(47). https://www.natlawreview.com/article/top-ten-legal-considerations-use-andor-development-
artificial-intelligence-health

Dilmegani, C. (2021, October 8). Bias in AI: What it is, Types & Examples of Bias & Tools to fix it. AI
Multiple. https://research.aimultiple.com/ai-bias/

Fallmann, D. (2021, August 19). Human Cognitive Bias And Its Role In AI. Mindbreeze InSpire. https://
inspire.mindbreeze.com/blog/human-cognitive-bias-and-its-role-in-ai

Forrester Consulting. (February 2020). The Cloud, Data, And AI Imperative For Healthcare:
Hospital Readiness And Adoption Paths For Next-Generation Technologies. https://www.ibm.com/
downloads/cas/GDMPZQJE

Gagliano M., Pham J., Tang B., Kashif H., & Ban J. (October 2017). Applications of Machine Learning in
Medical Diagnosis. ResearchGate. https://www.researchgate.net/publication/321151498

Gerke S., Minssen T., & Glenn Cohen I. (2020, May 26). Ethical and Legal Challenges of Artificial
Intelligence-Driven Healthcare. Artificial Intelligence in Healthcare, 1st edition, Adam Bohr, Kaveh
Memarzadeh (eds.). https://ssrn.com/abstract=3570129

Harrus, I., & Wyndham, J. (2021, May). Artificial Intelligence and COVID-19: Applications and
Impact Assessment. American Association for the Advancement of Science. https://www.
aaas.org/sites/default/files/2021-05/AIandCOVID19_2021_FINAL.pdf?adobe_mc=MCMID%3D
35126557705624345680603326284021649826%7CMCORGID%3D242B647254
1199F70A4C98A6%2540AdobeOrg%7CTS%3D1637858072

IBM Cloud Education. (2020, October 22). What is Robotic Process Automation. IBM. https://www.
ibm.com/cloud/learn/rpa

Kulkarni A. (2021, February 10). AI in Healthcare: Data Privacy and Ethics Concerns. Lexalytics.
https://www.lexalytics.com/lexablog/ai-healthcare-data-privacy-ethics-issues

Kuziemsky C., Maeder A., John O., Gogia S., Basu A., Meher S., & Ito M. (2019). Role of Artificial
Intelligence within the Telehealth Domain: Official 2019 Yearbook Contribution by the members of
IMIA Telehealth Working Group. International Medical Informatics Association Yearbook of Medical
Informatics. https://doi.org/10.1055/s-0039-1677941

Langlois L. (2021, May 25). Artificial Intelligence in Cloud Computing: The Secret to Streamlining
Healthcare., Providertech. https://www.providertech.com/artificial-intelligence-in-cloud-computing/

Mehta, N., Pandit, A., & Shukla, S. (2019). Transforming healthcare with big data analytics and
artificial intelligence: A systematic mapping study. Journal of Biomedical Informatics, 100, 103311.
https://doi.org/10.1016/j.jbi.2019.103311

© Copyright 2021, Cloud Security Alliance. All rights reserved. 21

 Mehta, S. (2021, May 31). Robotic Process Automation (RPA) in Healthcare Industry. Zenesys. https://
www.zenesys.com/blog/robotic-process-automation-rpa-in-healthcare-industry?source=techstories.
org

Nicholson Price II W. (2017). Artificial Intelligence in Health Care: Applications and Legal Issues.
American Bar Association. https://www.americanbar.org/groups/science_technology/publications/
scitech_lawyer/2017/fall/artificial-intelligence-health-care/

Ordish J. (2018, August). Legal Liability for Machine Learning in Healthcare. PHG Foundation. https://
www.phgfoundation.org/briefing/legal-liability-machine-learning-in-healthcare

Peck A.A. (2021, February 23). The Future is Digital Healthcare. The National Law Review, XI (54).
https://www.natlawreview.com/article/future-digital-healthcare

Price, S. (2020, June 5). Using Artificial Intelligence to determine COVID-19 severity. Health Europa.
https://www.healtheuropa.eu/using-artificial-intelligence-to-determine-covid-19-severity/100501/

Sivasubramanian, S. (2020, May 28). How AI and machine learning are helping to tackle COVID-19.
World Economic Forum. https://www.weforum.org/agenda/2020/05/how-ai-and-machine-learning-
are-helping-to-fight-covid-19/

Stahl, B. C. (2021). Ethical Issues of AI. SpringerBriefs in Research and Innovation Governance, 35–53.
https://doi.org/10.1007/978-3-030-69978-9_4

The Medical Futurist. (2019, October 8). Benefits of Robotics in Healthcare: Tasks Medical Robots Will
Undertake. https://medicalfuturist.com/robotics-healthcare/

Usachev, D. (2021). 10 Use Cases of Machine Learning for Medical Diagnosis. Fayrix. https://fayrix.
com/blog/ten-use-cases-of-machine-learning-for-medical-diagnosis

Williams, L. (2019, September 5). How AI and Big Data is Transforming the Healthcare Industry.
Techiexpert. https://www.techiexpert.com/how-ai-and-big-data-is-transforming-the-healthcare-
industry/

Zhang, Q., Xiao, Y., Suo, J., Shi, J., Yu, J., Guo, Y., Wang, Y., & Zheng, H. (2017). Sonoelastomics for
Breast Tumor Classification: A Radiomics Approach with Clustering-Based Feature Selection on
Sonoelastography. Ultrasound in Medicine & Biology, 43(5), 1058–1069. https://doi.org/10.1016/j.
ultrasmedbio.2016.12.016

© Copyright 2021, Cloud Security Alliance. All rights reserved. 22

View publication stats