CHAPTER ONE

INTRODUCTION

1.1. BACKGROUND OF AI IN FINANCE

Artificial Intelligence (AI) is transforming industries globally, with the financial sector being
one of the major beneficiaries of its advancements. AI refers to the ability of machines to
perform tasks that typically require human intelligence, including learning, reasoning,
problem-solving, and pattern recognition (Goodfellow et al., 2019). In finance, AI helps
process large amounts of data, automates repetitive tasks, and provides more accurate insights
for decision-making.

The application of AI in finance dates back to the early days of algorithmic trading, however,
recent advancements in machine learning, a subset of AI, have allowed financial institutions
to deploy more sophisticated models that can analyze vast datasets, predict trends, detect
anomalies, and automate services, such as robo-advisors and chatbots (Chen et al., 2020). AI
now powers fraud detection systems, credit scoring mechanisms, risk assessments, and high-
frequency trading platforms, optimizing financial operations like never before.

1.2 IMPORTANCE OF AI IN THE FINANCIAL SECTOR

AI plays a vital role in modern finance due to the increased complexity of financial products
and the sheer volume of transactions. It offers several critical benefits, such as:

 Enhanced Decision-Making: AI improves decision-making by providing predictive

analytics and insights derived from vast amounts of data (Owen et al., 2020).
Financial institutions use machine learning models to analyze historical data and
forecast future market conditions or creditworthiness.

 Fraud Prevention: AI helps detect fraudulent activities in real-time by monitoring

transactions and identifying suspicious patterns. With AI, financial institutions can
quickly respond to potential threats, minimizing losses (Srinivasan et al., 2021).

 Cost Reduction: Automation through AI reduces operational costs by automating

tasks such as customer service, data entry, and compliance monitoring.

1.3 DEFINITION OF RELATED KEY TERMS

1. Artificial Intelligence (AI): The simulation of human intelligence by machines that

can perform tasks such as learning, reasoning, and problem-solving (Goodfellow et
al., 2019).

2. Machine Learning (ML): A subset of AI that enables computers to learn from data
without being explicitly programmed, improving their performance on specific tasks
over time (Brown et al., 2021).

1
 3. Deep Learning (DL): A type of machine learning that uses neural networks with
many layers (also called deep neural networks) to model complex patterns in large
datasets (Goodfellow et al., 2019).

4. Robo-Advisors: AI-driven platforms that provide automated financial advice or

portfolio management services based on algorithms (Kumar & Rani, 2019).

5. Algorithmic Trading: The use of computer algorithms to execute trades based on

pre-defined criteria, allowing for faster, more accurate transactions in financial
markets (Chen et al., 2020).

6. Natural Language Processing (NLP): A branch of AI that focuses on the interaction

between computers and human language, allowing machines to understand and
process large amounts of natural language data (Brown et al., 2021).

7. Credit Scoring: A statistical analysis performed by financial institutions to determine

the creditworthiness of individuals or businesses, often enhanced by AI algorithms
(Zhang et al., 2019).

8. Fraud Detection: The identification of suspicious or unusual activities in financial

transactions, where AI models are increasingly used to recognize patterns and prevent
fraud (Srinivasan et al., 2021).

1.4 OBJECTIVES OF THE SEMINAR

The goal of this seminar is to analyze the application of AI technologies in the financial
sector, focusing on how they optimize operations and improve decision-making. Specifically,
this seminar aims to:

 Explore AI's contributions to fraud detection, credit scoring, algorithmic trading, and
customer service.

 Discuss the regulatory and ethical challenges faced by financial institutions deploying
AI.

 Analyze case studies of AI implementations in major financial organizations and their

impact on operational efficiency.

2
 CHAPTER TWO

LITERATURE REVIEW

2.1 HISTORICAL EVOLUTION OF AI IN FINANCE

The integration of Artificial Intelligence in finance has progressed significantly over the past
four decades, transforming the way financial institutions operate. During the 1980s, financial
firms first began utilizing AI for automated trading systems, which relied on simple rule-
based algorithms. These algorithms enabled the execution of trades with minimal human
intervention, marking the start of AI's influence on financial markets (Goodfellow et al.,
2019).

In the 2000s, advancements in machine learning, a subset of AI, began revolutionizing

finance by enabling systems to learn from historical financial data. These early AI systems
evolved into more sophisticated tools capable of identifying complex patterns, assessing
risks, and making informed decisions without human oversight. The financial crisis of 2008
highlighted the need for more robust risk management systems, and AI played a key role in
enhancing predictive analytics, identifying potential risks in markets, and supporting
informed decision-making (Chen et al., 2020).

Today, AI technologies such as deep learning, natural language processing (NLP), and
robotic process automation (RPA) are widely used in banking, insurance, and investment
management. These technologies have opened new avenues for personalized financial
services, enhanced fraud detection, and automated customer service (Brown et al., 2021).

2.2 AI TECHNOLOGIES APPLIED IN FINANCE

AI technologies applied in finance span multiple fields, providing a diverse set of tools to
enhance financial processes.

 Machine Learning (ML): The core of many AI applications in finance, ML uses

algorithms to sift through massive datasets and learn patterns. Its use ranges from
credit scoring and loan approvals to risk management. In investment management,
ML algorithms analyze historical stock prices and market trends, providing predictive
insights that guide portfolio adjustments (Brown et al., 2021).

 Natural Language Processing (NLP): NLP has become particularly useful in

customer service, as it enables the creation of intelligent chatbots and virtual
assistants. Financial institutions employ NLP to process and understand human
language, allowing them to automate customer queries, investment advice, and
sentiment analysis from social media or news articles (Owen et al., 2020).

 Deep Learning: Deep learning models, which are inspired by the human brain’s
structure, are increasingly used in high-frequency trading (HFT) and fraud detection.
These models can process large volumes of data, recognize intricate patterns, and
continuously improve their decision-making over time. For example, in credit card

3
 fraud detection, deep learning models have outperformed traditional rule-based
systems by recognizing suspicious transactions in real-time (Chen et al., 2020).

 Robotic Process Automation (RPA): RPA automates repetitive, rule-based tasks

such as compliance reporting, data entry, and document processing. Financial
institutions are adopting RPA to streamline back-office operations, allowing
employees to focus on more complex, value-adding activities (Brown et al., 2021).

2.3 INNOVATIONS IN FFINANCE SERVICES ENABLED BY AI

AI has enabled several innovations that are reshaping financial services:

 Robo-Advisors: Robo-advisors are AI-powered platforms that offer automated,

algorithm-driven financial planning services with little to no human intervention.
They use AI to evaluate users' financial goals, risk tolerance, and market conditions,
providing customized investment advice. Platforms like Betterment and Wealthfront
have democratized financial advice, making it more accessible and affordable (Kumar
& Rani, 2019).

 Fraud Detection and Prevention: AI has been instrumental in improving fraud

detection systems, particularly in payment processing platforms like PayPal. AI-
driven fraud detection algorithms analyze transaction data to identify irregularities or
fraudulent patterns, enabling real-time fraud prevention (Srinivasan et al., 2021).
Machine learning models are able to continuously learn from new data, becoming
more effective over time in recognizing new forms of fraud.

 Algorithmic Trading: Algorithmic trading has become one of AI's most impactful
contributions to finance. By processing vast datasets in milliseconds, AI algorithms
optimize trade execution strategies and exploit market inefficiencies at a speed that is
unattainable for human traders (Chen et al., 2020). Hedge funds and investment banks
increasingly rely on algorithmic trading to increase profit margins, improve liquidity,
and reduce risks associated with manual trading errors.

 Credit Scoring: Traditional credit scoring models often rely on a limited range of
data, leading to biased outcomes. AI-driven credit scoring expands this by analyzing a
broader range of data, such as social media behavior, real-time transaction patterns,
and alternative financial histories. This allows financial institutions to evaluate
creditworthiness more accurately and offer credit to underserved populations (Owen
et al., 2020).

2.4 CHALLENGES IN AI IMPLEMENTATION

While AI is transforming the financial sector, several challenges persist:

 Data Privacy and Security: Financial institutions handle vast amounts of sensitive
data, which makes them prime targets for cyberattacks. AI systems that process this
data must comply with strict data protection regulations such as the General Data

4
 Protection Regulation (GDPR) in Europe (Zhang et al., 2019). Moreover, ensuring the
security of AI systems themselves is paramount to avoid the risk of malicious actors
manipulating these systems for fraudulent purposes.

 Bias in Algorithms: One of the significant ethical challenges in AI is algorithmic

bias. AI models trained on biased data may inadvertently perpetuate unfair practices
in areas such as loan approvals, credit scoring, or investment decisions. For example,
if historical data is biased against specific demographic groups, AI systems may
replicate or even exacerbate those biases (Owen et al., 2020).

 Regulatory Compliance: Financial institutions operate in a highly regulated

environment, and the rapid integration of AI into financial systems poses challenges
in maintaining regulatory compliance. Regulators are grappling with how to ensure
that AI systems are transparent and accountable, especially in high-stakes areas like
credit scoring and algorithmic trading (Brown et al., 2021).

 Cost and Complexity of Implementation: Implementing AI solutions requires

significant upfront investment in technology and expertise. Many smaller financial
institutions face barriers to AI adoption due to the cost and complexity of integrating
AI systems into their existing infrastructure (Kumar & Rani, 2019).

5
 CHAPTER THREE

DISCUSSION

3.1 APPLICATION OF AI IN FINANCIAL SERVICES

3.1.1 AI IN BANKING SECTOR

We are already seeing several areas in banking services that have been taking advantage of
this disruptive technology. The following are some use cases where AI has been most
impactful within the BFSI industry.

 Chatbots: AI-powered chatbots incorporated with Natural Language Processing

(NLP), engage and interact with customers 24/7 and enhance online conversations. In
addition to typical responses to customers’ questions to help them work through their
account details, chatbots can now help in opening new accounts and directing
complaints to appropriate customer service units amongst others.

 Fraud Detection & Prevention: Until very recently, banks have relied on traditional,
rule-based Anti-Money Laundering (AML) transaction monitoring and name
screening systems which generate a high number of false positives. With the alarming
increase in fraud-related crimes and ever-changing fraud patterns, enhanced AI
components are being added to the existing systems to enable the identification of
previously undetected transactional patterns, data anomalies and suspicious
relationships between individuals and entities. This allows for a more proactive
approach, where AI is used to prevent fraud before it happens as opposed to the
traditional reactive approach to fraud detection.

 Customer Relationship Management: Customer relationship management is an

important factor for banks. Banks are now providing more personalised 24/7 services
to individual customers such as providing facial recognition and voice command
features to log in to financial apps. Banks are also leveraging Artificial Intelligence to
analyse customer behavioural patterns and automatically perform customer
segmentation which allows for targeted marketing and improved customer experience
and interaction.

 Predictive Analytics: The advent of Machine Learning (ML) & AI has opened the
door to accurate forecasting and prediction. Data Analytics and AI are being applied
to revenue forecasting, stock price predictions, risk monitoring and case management.
The exponential increase in the data collected has been pivotal in improving the
performance of the models, resulting in a gradual decline in the level of human
intervention required.

 Credit Risk Management: As regulators continue to focus on risk management

supervision, financial institutions are mandated to develop more reliable models and
solutions. The use of AI in credit risk management is gaining more popularity
especially in the Fintech and the Digital Banking market.

6
 AI is used to determine the creditworthiness of the facility borrower by harnessing data to
predict the probability of default which helps to improve the accuracy of credit decisions.
As a result, the market is moving towards insights-driven lending rather than expert
judgments, which helps maximize rejection of high-risks customers and minimize
rejection of creditworthy customers as well as a reduction in credit losses incurred by /the
financial institutions.

3.1.2 AI IN FRAUD DETECTION

Fraud is one of the most significant challenges faced by financial institutions. With the rising
complexity of financial transactions, traditional fraud detection methods are often ineffective
in detecting sophisticated fraud schemes. AI has proven to be an effective tool for fraud
prevention due to its ability to learn patterns from vast datasets and detect anomalies in real
time.

AI-based fraud detection systems use machine learning algorithms to analyze historical
transaction data and recognize patterns that could indicate fraudulent activities. These
systems continuously learn and evolve as they process more data, enabling them to adapt to
new types of fraud. For example, PayPal implemented an AI-powered fraud detection system
that reduced fraud rates by analyzing billions of transactions and identifying suspicious
behaviors (Srinivasan et al., 2021). By leveraging AI, PayPal was able to improve the
accuracy of its fraud detection process and reduce false positives, which are common in rule-
based fraud detection systems.

Another case is JP Morgan Chase, which utilizes AI-driven models to detect fraud in wire
transfers. The bank implemented an AI system that can scan over 150 variables per
transaction, identifying subtle anomalies that might indicate fraud. This proactive approach
has led to a significant decrease in fraudulent wire transfers, providing greater security for
both the bank and its customers (Brown et al., 2021).

3.1.3 AI IN CREDIT SCORING

Traditional credit scoring models are primarily based on factors like income, credit history,
and debt-to-income ratio. However, these models can be limited in scope and may not
provide an accurate picture of a borrower’s creditworthiness, especially in the case of
individuals with limited credit history (Zhang et al., 2019). AI has revolutionized the credit
scoring process by introducing new methods that incorporate alternative data sources such as
social media behavior, mobile phone usage, and online transaction patterns.

For example, Zest AI, a fintech company, uses machine learning algorithms to assess credit
risk by analyzing over 10,000 data points, including financial and non-financial information.
This approach allows Zest AI to make more accurate credit risk assessments, particularly for
those who are typically underserved by traditional models. As a result, Zest AI has helped
reduce defaults while providing credit access to more individuals (Zhang et al., 2019).

Additionally, LenddoEFL, a digital credit scoring platform, uses AI to analyze alternative

data such as smartphone usage and social network connections to create credit profiles for

7
individuals in developing markets. This has enabled lenders to expand credit access to people
who lack formal credit histories, empowering financial inclusion (Kumar & Rani, 2019).

3.1.4 AI IN ALGORITHMIC TRADING

Algorithmic trading, which involves using pre-programmed algorithms to execute trades, has
become more prevalent with advancements in AI and machine learning technologies. AI
models can analyze vast amounts of market data in real-time, identifying patterns and making
split-second decisions to execute trades at optimal prices. This has given institutional
investors a competitive edge in financial markets by improving speed and efficiency.

One notable case study is Renaissance Technologies, a hedge fund that uses AI and machine
learning to make highly sophisticated trades. The firm’s flagship Medallion Fund uses
complex AI models to analyze historical market data and predict future trends. This approach
has resulted in consistently high returns, even during periods of market volatility (Chen et al.,
2020).

Similarly, Goldman Sachs has implemented AI-driven trading algorithms that optimize the
execution of large trades, minimizing price impact and reducing transaction costs. AI models
allow the bank to adapt to changing market conditions and make real-time decisions that
would be impossible for human traders to replicate (Owen et al., 2020).

3.1.5 AI IN WEALTH MANAGEMENT

AI is also transforming wealth management by providing personalized investment advice

through robo-advisors. Robo-advisors use AI algorithms to assess an individual's financial
goals, risk tolerance, and time horizon, then recommend an optimal investment portfolio.
These AI-driven platforms have democratized access to wealth management services, making
it possible for individuals with smaller investment portfolios to receive high-quality financial
advice.

For instance, Betterment, a popular robo-advisor, uses AI to manage over $30 billion in
assets. The platform offers automated portfolio management, tax-loss harvesting, and
financial planning services at a fraction of the cost of traditional financial advisors. This has
allowed Betterment to attract a diverse range of investors, from millennials looking to save
for retirement to high-net-worth individuals seeking tax-efficient investment strategies
(Kumar & Rani, 2019).

3.1.6 AI FOR REGULATORY COMPLIANCE

AI has also emerged as a key tool for regulatory compliance in finance, commonly referred to
as “RegTech.” Financial institutions must comply with a range of regulations related to anti-
money laundering (AML), know-your-customer (KYC) requirements, and data privacy. AI
solutions can streamline the compliance process by automating the collection, verification,
and monitoring of customer data. These systems can identify suspicious activities, automate
reporting, and reduce the risk of non-compliance.

8
HSBC, for example, has implemented AI-powered RegTech solutions to automate its AML
compliance processes. The AI system scans vast amounts of transaction data, identifying
suspicious patterns that may indicate money laundering. This has resulted in faster and more
accurate compliance checks, reducing the burden on human compliance officers and
improving regulatory outcomes (Brown et al., 2021).

CHAPTER FOUR

CONCLUSION

4.1 SUMMARY OF KEY FINDINGS

Artificial intelligence (AI) is fundamentally reshaping the landscape of healthcare by

revolutionizing the diagnosis and treatment of diseases. As explored in this seminar, AI
applications are enabling earlier and more accurate diagnosis, personalized treatment plans,
and enhanced patient management. From radiology to oncology and cardiology, AI's ability
to analyze vast datasets and identify complex patterns that may go unnoticed by human
professionals has significantly improved the precision and efficiency of healthcare delivery.
AI technologies such as machine learning, natural language processing (NLP), and robotics
are driving innovations that are transforming medical practices and improving patient
outcomes across multiple domains.

In diagnostics, AI-powered tools are already outperforming humans in certain areas like
image recognition, where they can detect abnormalities in medical scans with higher
accuracy. In treatment, AI is at the forefront of precision medicine, where individual genetic

9
profiles are analyzed to create targeted therapies, particularly in cancer treatment.
Additionally, AI's role in patient monitoring, through wearables and remote health systems,
allows for continuous health surveillance, reducing hospital readmissions and ensuring timely
medical interventions.

4.2 BENEFITS OF AI IN HEALTHCARE

The adoption of AI in healthcare brings numerous benefits that are not limited to just clinical
settings but extend to administrative, research, and operational efficiency.

1. Enhanced Diagnostic Accuracy: AI systems can analyze medical data more

thoroughly and rapidly than humans, identifying early-stage diseases that might be
missed in routine diagnostics. This ability is particularly crucial in fields like
oncology, where early detection of cancer can significantly improve survival rates
(McKinney et al., 2020).
2. Personalized Treatment: AI facilitates precision medicine by tailoring treatment
plans to individual patients based on their genetic makeup, lifestyle, and medical
history. This reduces the trial-and-error approach in prescribing medications and
minimizes side effects, especially in complex treatments such as chemotherapy
(Wang et al., 2019).
3. Improved Efficiency and Cost Reduction: AI streamlines healthcare workflows by
automating time-consuming tasks such as data entry, medical image analysis, and
even some administrative processes like scheduling. This reduces operational costs,
frees up medical professionals for more critical tasks, and ultimately improves patient
care.
4. Continuous Patient Monitoring: AI-powered wearable devices and remote
monitoring systems provide real-time tracking of patient health, allowing for early
detection of potential issues and reducing the need for hospitalization. This
technology has proven to be especially beneficial in managing chronic diseases,
where early interventions can prevent complications (Zhang et al., 2020).
5. Support for Healthcare Professionals: AI acts as a powerful tool in assisting
healthcare professionals in decision-making, particularly in complex or high-stakes
scenarios. From recommending treatment options based on vast datasets to
automating mundane tasks, AI allows healthcare professionals to focus on patient care
and innovation (Topol, 2019).

4.3 LIMITATIONS AND ETHICAL CONSIDERATIONS

While AI in healthcare offers transformative potential, it also comes with several limitations
and ethical concerns that need to be addressed to ensure its responsible and equitable use.

1. Data Privacy and Security: AI relies on vast amounts of sensitive patient data to
function effectively. The collection, storage, and sharing of this data pose significant
privacy and security risks, especially if healthcare systems are breached. It is

10
 imperative that stringent cybersecurity measures be implemented to protect this
sensitive information (Char et al., 2018).
2. Algorithmic Bias: AI systems are only as good as the data they are trained on. If the
data used to train AI algorithms is biased, it may result in unequal treatment across
different demographics. For example, underrepresentation of certain ethnic or gender
groups in training data could lead to inaccurate diagnoses or treatment
recommendations for those populations. Addressing this bias is critical to ensuring
that AI enhances, rather than exacerbates, healthcare disparities (Jiang et al., 2020).
3. Transparency and Accountability: Many AI models, particularly those based on
deep learning, function as "black boxes," where the decision-making process is not
easily explainable or interpretable by humans. In healthcare, where lives are at stake,
it is crucial to have transparent and interpretable AI systems to ensure accountability.
The concept of "explainable AI" (XAI) is gaining traction, aiming to make AI systems
more understandable to healthcare providers and patients alike (Esteva et al., 2019).
4. Regulatory and Legal Challenges: The rapid pace of AI development outstrips the
ability of regulatory bodies to create comprehensive frameworks to govern its use.
Clear regulations must be established to ensure patient safety, ethical use of AI
technologies, and legal recourse in cases of AI-related errors (Char et al., 2018).

4.4 ETHICAL AND SOCIETAL IMPLICATIONS

The widespread adoption of AI in healthcare also raises broader ethical and societal
questions. As AI continues to automate more healthcare tasks, there are concerns about job
displacement, particularly for administrative and diagnostic roles. While AI will likely
augment human capabilities rather than replace them entirely, it is essential to consider how
the healthcare workforce will adapt to these changes.

Additionally, AI's role in healthcare decision-making presents philosophical questions about

the role of machines in patient care. Can AI be trusted to make decisions that directly impact
human lives? While AI systems can provide invaluable assistance, there is still a need for
human oversight, particularly in situations requiring empathy, ethical judgment, or complex
decision-making that involves balancing medical, emotional, and ethical considerations.

4.5 FUTURE DIRECTIONS AND INNOVATIONS

Looking to the future, the integration of AI in healthcare is likely to accelerate, driven by

continuous technological advances and increasing volumes of medical data. Several key
trends are expected to shape the future of AI in healthcare:

1. Integration of AI with Genomic Data: As the cost of genomic sequencing continues

to decrease, AI systems will increasingly incorporate genetic data into healthcare
decision-making. This will open new avenues for personalized medicine, where
treatments are specifically designed for an individual’s genetic profile.

11
 2. AI-Powered Drug Discovery: AI is revolutionizing the field of drug discovery by
accelerating the process of identifying potential new drugs. By simulating drug
interactions and predicting the success of clinical trials, AI could drastically reduce
the time and cost required to bring new treatments to market (Topol, 2020).
3. Remote Healthcare Delivery: The COVID-19 pandemic has accelerated the
adoption of telemedicine, and AI is playing a crucial role in this transformation. AI-
powered virtual assistants and chatbots are being developed to triage patients, provide
real-time medical advice, and streamline remote patient monitoring. This trend is
likely to continue, improving access to healthcare, particularly in underserved or rural
areas (Zhang et al., 2020).
4. AI and Robotics in Surgery: AI-driven robots are already assisting surgeons in
performing minimally invasive procedures. Future innovations in robotics and AI are
expected to make surgeries even more precise and less invasive, reducing recovery
times and improving patient outcomes.

4.6 FINAL THOUGHTS

AI's potential to revolutionize healthcare is undeniable. By improving diagnostic accuracy,

optimizing treatment plans, enhancing patient monitoring, and reducing administrative
burdens, AI is paving the way for a more efficient, patient-centered healthcare system.
However, to fully realize these benefits, the healthcare industry must address the ethical,
legal, and technical challenges that accompany AI’s widespread adoption.

The future of healthcare will likely be a partnership between human expertise and machine
intelligence, where AI augments, rather than replaces, the role of healthcare professionals.
With careful oversight, robust regulatory frameworks, and ethical considerations at the
forefront, AI has the potential to create a more equitable, efficient, and personalized
healthcare system that improves outcomes for patients worldwide.

12
 REFERENCES

Alverez-Valle J, Moore GJ. Project InnerEye open-source deep learning toolkit:

Democratizing medical imaging AI. Microsoft, 2020.

Bellemo V, Lim ZW, Lim G, et al.. Artificial intelligence using deep learning to screen
for referable and vision-threatening diabetic retinopathy in Africa: a clinical
validation study. Lancet Digit Health 2019

Char, D. S., Shah, N. H., & Magnus, D. (2018). Implementing machine learning in health
care—Addressing ethical challenges. The New England Journal of Medicine,
378(11), 981-983.

Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S.
(2019). Dermatologist-level classification of skin cancer with deep neural
networks. Nature, 542(7639), 115-118.

Gulshan V, Peng L, Coram M, et al.. Development and validation of a deep learning

algorithm for detection of diabetic retinopathy in retinal fundus photographs.
JAMA 2016;316:2402–10.

13
Haque A, Milstein A, Fei-Fei L. Illuminating the dark spaces of healthcare with ambient
intelligence. Nature 2020.

Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., ... & Wang, Y. (2020). Artificial
intelligence in healthcare: Past, present, and future. Stroke and Vascular
Neurology, 2(4), 230-243.

McBee, M. P., Awan, O. A., Colucci, A. T., Ghobadi, C. W., Kadom, N., Kansagra, A.
P., & Tridandapani, S. (2018). Deep learning in radiology. Academic Radiology,
25(11), 1472-1480.

Muehlematter UJ, Daniore P, Vokinger KN. Approval of artificial intelligence and

machine learning-based medical devices in the USA and Europe (2015–20): a
comparative analysis. Lancet Digital Health 2021;3:e195–203.

Nachev P, Herron D, McNally N, Rees G, Williams B. Redefining the research hospital.

NPJ Digit Med 2019.

Nelson, B., Lee, D. S., & Anders, M. E. (2021). Advancements and challenges in AI for
healthcare. Journal of Medical Systems, 45(4), 123-130.

Oktay O, Nanavati J, Schwaighofer A, et al. Evaluation of deep learning to augment

image-guided radiotherapy for head and neck and prostate cancers. JAMA Netw
Open 2020;3:e2027426.

Russell, S., & Norvig, P. (2020). Artificial Intelligence: A Modern Approach. Pearson.

Senior AW, Evans R, Jumper J, et al.. Improved protein structure prediction using
potentials from deep learning. Nature 2020;577:706–10.

Simonite T. The US government will pay doctors to use these AI algorithms. Wired,
2020.

Strodthoff N, Strodthoff C. Detecting and interpreting myocardial infarction using fully

convolutional neural networks. Physiological Measurement 2019;40:015001.

The AlphaFold team . AlphaFold: a solution to a 50-year-old grand challenge in biology.

DeepMind, 2020.

Topol, E. J. (2019). Deep Medicine: How Artificial Intelligence Can Make Healthcare
Human Again. Basic Books.

University of Leeds . NPIC - Northern Pathology Imaging Co-operative. University of

Leeds, 2020.

14
Wang A, Nguyen D, Sridhar AR, et al.. Using smart speakers to contactlessly monitor
heart rhythms. Commun Biol 2021;4:319.

Wang, F., Preininger, A., et al. (2019). Precision medicine: The role of AI in targeting
treatment. Nature Medicine, 25(1), 30-36.

Yang, G. Z., et al. (2021). Medical robotics in surgery and rehabilitation. Annual Review
of Biomedical Engineering, 23(1), 345-368.

Zhang, H., Liu, Y., & Lu, W. (2020). AI-based coronary artery analysis: Toward
automated cardiovascular diagnosis. IEEE Transactions on Medical Imaging,
39(12), 4124-4132..

15