Addis Ababa University

College of Technology and Built Environment

Introduction to Emerging Technologies: EmTe 1012

Artificial Intelligence (AI) in Healthcare

Nahom Asnake
UGR/1784/17

Submitted to: Mr. Habtamu Abera

Submission date: May 14, 2025
Table of Contents
Introduction..................................................................................................................................................2
Technical Terms and Definitions................................................................................................................4
Historical Development of AI in Healthcare ............................................................................................. 6
1950s–1970s: The Origins................................................................................................................6
1970s–1980s: Rule-Based Expert Systems.......................................................................................6
1990s–2000s: Rise of Machine Learning ......................................................................................... 6
2010s: Big Data and Deep Learning.................................................................................................6
2020s: Mainstream Integration and COVID-19 Impact ................................................................... 7
Real-World Applications of AI in Healthcare...........................................................................................7
1. Diagnostics....................................................................................................................................7
2. Treatment Personalization ............................................................................................................ 7
3. Robotics and Surgery....................................................................................................................8
4. Predictive Analytics......................................................................................................................8
5. Administrative Use ....................................................................................................................... 8
6. Mental Health................................................................................................................................9
Advantages and Disadvantages of AI in Healthcare...............................................................................10
Advantages of AI in Healthcare............................................................................................................10
Disadvantages of AI in Healthcare........................................................................................................11
Literature Review / Existing Systems.......................................................................................................13
Insights into AI in Healthcare.........................................................................................................13
Contributions from Major Companies............................................................................................13
Key Systems and Their Impact.......................................................................................................14
Conclusion...................................................................................................................................................15
References.......................................................................................................................................16
Introduction

Every year, hundreds and thousands of people die from preventable errors and misplaced data.
These tragic yet avoidable and preventable accidents showcase that there is a need for smarter,
faster and trustworthy healthcare systems. Here is where AI comes in, a rapidly growing field
within computer science. Artificial Intelligence (AI) is coming as one of a kind, powerful tool in
creating solutions to these healthcare problems causing suffering as was mentioned earlier.

Artificial intelligence (AI) is technology that enables computers and machines to simulate human
learning, comprehension, problem solving, decision making, creativity and autonomy. It involves
the creation of an intelligent system that can stimulate by observing, learn from data it is given,
recognize patterns within the data and make its own autonomic decisions. It includes technologies
like machine learning, natural language processing, and neural networks.

Healthcare is one of the most important aspects in modern society, dealing with big data, critical
decision making moments, and actual human lives. The use of AI in healthcare promises to
advance patient care, reduce errors, and improve efficiency. From treating diseases to specific
treatments, AI is becoming a major tool for medical professionals. And healthcare, since it
generates vast amounts of data daily(spanning from patient records and imaging to lab results and
genomics), is suited for the integration and use of artificial intelligence.

AI in healthcare refers to the application of artificial intelligence techniques to analyze and

understand complex medical data, ultimately improving diagnosis, treatment and overall patient
care. AI systems are able to quickly analyze and synthesize data to unveil shortcomings that could
remain hidden. Simply put Ai is changing the way healthcare is performed.

However, despite the advancements made on the use of AI into healthcare does not come without
its own challenges and disadvantages. Concerns arise such as the AI becoming more algorithm
biased, lacking accountability and even the risk of depersonalizing care. These concerns of course
must be carefully addressed without haste. Others mention the requirement for investment,
infrastructure as well as education, which are not easily and readily available.
Even so the benefits that AI brings to the table weighs much more than the negative aspects of it
since it offers quick , cost effective solutions to some of the most challenging diseases to
pandemic out there. Which is why governments nowadays are advocating the implementation of
AI assisted health research and care around the globe, which implies the beginning of a new era
in how we receive and give care.

This paper will focus on the application of AI in modern healthcare centres with modern AI
technologies in use for these advancements. This assignment will be all about Artificial
Intelligence in the field of healthcare by first explaining the important technical terms followed
by its historical background then proceeding to explore the applications of this technology to the
real world. We will then see the advantages and disadvantages it brings with itself, with a review
of an already existing AI system and many more important points. Through this structure this
paper aims to demonstrate the use of Ai though it might bring challenges and concerns, it can be
worked upon creating solutions and then of course showcase its ability to improve the health
centre in general with more desirable and promising outcomes than tragic ones.
Technical Terms and Definitions

The important thing here is to first understand the underlying technologies behind what we are
calling Artificial Intelligence (AI) to grasp the applications of it in healthcare. In this section we
will see the definitions of different technical terms that will be used throughout this paper.

Machine Learning (ML)

Machine Learning is a subset of AI that allows systems to learn and improve from experience
without being explicitly programmed for every scenario. ML algorithms use statistical
techniques to identify patterns in data, which can then be used for predictive analysis. In a
healthcare context, machine learning helps detect anomalies in X-rays or predict patient
readmission rates based on past records.

Deep Learning

Deep Learning is an advanced branch of machine learning that uses layered neural networks
(inspired by the human brain) to analyze complex datasets. It excels in tasks involving image and
speech recognition. For instance, deep learning models are widely used in analyzing MRI scans
to detect tumors with higher accuracy than traditional methods.

Natural Language Processing (NLP)

Natural Language Processing enables machines to understand, interpret, and respond to human
language. In healthcare, NLP is applied in transcribing doctor patient conversations, extracting
data from electronic health records, and analyzing patient feedback. It bridges the gap between
human communication and machine understanding.

Neural Networks

Neural Networks are the backbone of many AI systems. Modeled after the human brain’s
structure, they consist of interconnected layers of nodes (“neurons”) that process data inputs and
produce outputs. In healthcare, neural networks are often used in diagnostic tools that analyze
visual inputs like CT scans or ECGs.
Computer Vision

Computer Vision is a field of AI that trains machines to interpret and make decisions based on
visual data. In medicine, computer vision is used for analyzing medical images such as
radiographs, pathology slides, and dermatology scans. It significantly reduces diagnostic time
and improves accuracy, especially in detecting early signs of disease.

Big Data in Healthcare

Big Data refers to massive volumes of complex health related information generated from patient
records, wearable devices, clinical trials, and even social media. AI thrives on big data, using it
to uncover insights, forecast trends, and inform clinical decisions. The effective use of big data in
healthcare can lead to better patient outcomes and more efficient resource allocation.

Clinical Decision Support Systems (CDSS)

CDSS are AI powered tools that assist healthcare providers in making informed clinical
decisions. These systems analyze patient data and medical knowledge to suggest diagnoses, flag
potential drug interactions, and recommend evidence based treatments. CDSS enhance clinical
workflows and reduce the likelihood of human error.
Historical Development of AI in Healthcare
AI in healthcare has a long and evolving journey, with each phase introducing key innovations
that brought the field closer to what we know today. This history reflects how ideas that once
seemed futuristic are now part of everyday medical care.

1950s–1970s: The Origins

The foundation for AI was laid in the 1950s, when Alan Turing proposed the idea of a machine
that could mimic human intelligence. During this time, early attempts were made to apply these
ideas to medicine, such as using computers for basic diagnostic reasoning. However, computing
power was too limited for any large scale use.

1970s–1980s: Rule Based Expert Systems

This period saw the development of expert systems like MYCIN, one of the earliest AI tools
designed to assist doctors in diagnosing blood infections. MYCIN used a set of IF-THEN rules to
simulate human decision-making. Although groundbreaking, these systems were difficult to
update and couldn’t easily adapt to new knowledge.

1990s–2000s: Rise of Machine Learning

With improved computing power and access to more digital data, AI evolved beyond fixed rules.
Machine learning algorithms began to emerge, enabling systems to learn patterns from data
instead of relying solely on pre-programmed rules. This was a crucial shift in the use of AI for
analyzing patient records, diagnostic images, and laboratory results.

2010s: Big Data and Deep Learning

The explosion of electronic health records and the growth of medical imaging datasets provided
the fuel for a new wave of AI breakthroughs. Deep learning models (especially neural
networks)began to outperform traditional algorithms in fields like radiology and pathology.

Major milestones include:

● IBM Watson Health, which attempted to bring AI into cancer treatment decision-making.

● Google’s DeepMind, which achieved expert-level performance in diagnosing eye disease

and breast cancer through medical image analysis.

2020s: Mainstream Integration and COVID-19 Impact

AI is no longer experimental, it is now embedded in many healthcare settings. From AI-powered
triage systems to predictive models used during the COVID-19 pandemic to forecast infection
surges and resource needs, AI has proven itself in real-world crises.

Health systems are also using AI to optimize operations, reduce unnecessary hospital stays, and
improve care coordination. The technology is increasingly being integrated into daily hospital
workflows, wearables, mobile apps, and robotic systems.

Real-World Applications of AI in Healthcare

AI is being used across nearly every part of the healthcare industry—from diagnosing diseases
and planning surgeries to managing hospital operations and supporting mental health.

1. Diagnostics

AI helps analyze medical images and detect conditions with high accuracy.

● Radiology: AI systems read X-rays, MRIs, and CT scans to detect issues like tumors,
fractures, or infections. Tools like Google’s DeepMind have matched or surpassed human
radiologists in some tasks.

● Pathology: AI can spot cancer cells in biopsy samples faster and more reliably than
traditional methods.

● Real-World Examples:

○ IBM Watson Health assists in oncology by evaluating treatment options based

on patient records.

○ Google’s DeepMind has been used to predict acute kidney injury and diagnose
eye diseases.

2. Treatment Personalization

AI is used to customize treatments for individual patients.

● Genomics: AI analyzes genetic data to recommend specific drugs that are more likely to
be effective based on a person's DNA.
 ● Oncology: In cancer treatment, AI suggests therapies based on tumor characteristics and
the patient’s medical history.

● Example: Tempus, a health tech company, uses AI and big data to support oncologists
with treatment plans tailored to genetic profiles.

3. Robotics and Surgery

AI-powered robotic systems assist doctors in performing complex surgeries.

● Da Vinci Surgical System: Enhances precision and reduces invasiveness in surgeries.

● Autonomous Tools: Some AI-based tools are beginning to assist in stitching, cutting,
and other delicate procedures with minimal human guidance.

● Benefits: Less blood loss, faster recovery times, and improved surgical accuracy.

4. Predictive Analytics

AI predicts future health events using patient and population-level data.

● Disease Prediction: AI models can forecast the risk of developing diabetes, heart
disease, or even predict cancer years in advance.

● Pandemic Forecasting: During COVID-19, AI predicted hospital bed needs and

infection rates to guide public health decisions.

● Hospital Readmission Prevention: AI identifies patients at high risk of readmission,

helping to plan better follow-up care.

5. Administrative Use

AI streamlines the business side of healthcare.

● Medical Recordkeeping: AI-powered Natural Language Processing (NLP) tools

transcribe and organize doctor notes, lab results, and reports.
 ● Scheduling & Billing: AI systems automate appointment booking, billing, and insurance
claim processing.

● Outcome: Reduced paperwork, improved accuracy, and better time management for
doctors and staff.

6. Mental Health

AI is making mental healthcare more accessible and personalized.

● AI Chatbots: Tools like Woebot provide mental health support through natural
conversations, helping people manage anxiety and depression.

● Emotional Analysis: AI analyzes tone, word choice, and facial expressions to assess
emotional well-being.

● Early Detection: AI tools can flag potential mental health issues by monitoring online
behavior or wearable data.
Advantages and Disadvantages of AI in Healthcare
Advantages of AI in Healthcare

1. Improved Diagnostic Accuracy

AI algorithms can analyze complex medical data with a level of precision that often exceeds
human capability. They are especially effective in radiology, pathology, and dermatology, where
AI tools can detect tumors, fractures, and skin conditions with remarkable accuracy, even in their
early stages.

2. Faster Medical Decisions

AI systems can process and analyze data far faster than humans, allowing healthcare
professionals to make informed decisions quickly. This is particularly vital in emergency
settings, where time-sensitive decisions can mean the difference between life and death.

3. Automation of Administrative Work

Routine tasks such as appointment scheduling, billing, data entry, and insurance claims can be
automated using AI. This reduces clerical workload, minimizes human error, and allows
healthcare staff to focus more on patient care.

4. Personalized Treatment Plans

AI can tailor treatment plans based on a patient’s genetic profile, lifestyle, and medical history.
These individualized approaches improve treatment effectiveness and reduce the risk of adverse
reactions, especially in oncology and chronic disease management.

5. AI-Assisted Surgeries and Robotics

Robotic surgery systems enhanced by AI offer greater precision, stability, and minimal
invasiveness. These systems can assist surgeons during complex procedures, reduce human
fatigue, and lead to shorter recovery times for patients.

6. Real-Time Health Monitoring

Wearable AI-powered devices can continuously track patients’ vital signs like heart rate, oxygen
levels, and glucose levels. These devices can alert both patients and providers to irregularities
before a condition becomes critical, enhancing preventive care.
7. Cost Efficiency in Healthcare Delivery
By reducing diagnostic errors, preventing hospital readmissions, and streamlining operations, AI
significantly lowers healthcare costs. Hospitals that adopt AI tools report improvements in
resource allocation and a measurable drop in operational expenditures.

Disadvantages of AI in Healthcare

1. Data Privacy and Confidentiality Risks

AI systems require access to large amounts of sensitive health information. If not properly
secured, this data can be vulnerable to breaches, leading to serious privacy violations and
potential misuse of personal medical records.

2. Algorithmic Bias and Inequality

If AI systems are trained on biased datasets, they may deliver unequal outcomes. For example, a
diagnostic tool trained primarily on data from one demographic might be less accurate for others,
worsening disparities in healthcare access and treatment.

3. Lack of Human Judgment

AI systems lack empathy and human intuition. While they can make decisions based on
patterns, they may not account for emotional, cultural, or ethical factors that are crucial in
medical practice, particularly in end-of-life care or mental health.

4. Dependence on Data Quality

Poor-quality or incomplete data can result in flawed AI performance. If training data includes
errors, missing information, or inconsistencies, the system’s predictions and suggestions may be
incorrect or even dangerous.
5. Regulatory and Legal Challenges

There are still no universal standards or legal frameworks governing the use of AI in healthcare.
This leads to uncertainty about liability, who is responsible if an AI system contributes to a
misdiagnosis or treatment error?

6. High Cost of Implementation

Setting up AI infrastructure, including software, hardware, and staff training, is expensive.

Smaller hospitals or those in developing regions may struggle to afford such investments, leading
to unequal access to AI-driven care.

7. Potential Job Displacement

As AI systems take over administrative and diagnostic roles, some healthcare professionals
(especially in clerical or technical positions) worry about job security. While AI is intended to
assist, not replace, this fear may hinder acceptance and collaboration.
Literature Review / Existing Systems
Insights into AI in Healthcare

1. The Potential for Artificial Intelligence in Healthcare

Davenport and Kalakota (2019) discuss the transformative role of AI in healthcare,

highlighting its applications in diagnosis, treatment recommendations, patient
engagement, and administrative tasks. They emphasize that while AI can perform certain
healthcare tasks as well as or better than humans, widespread automation faces
challenges such as implementation barriers and ethical considerations.

2. AI in Radiology: Enhancing Diagnostic Accuracy

A study published in the Washington Post (2025) explores the integration of AI in

radiology, noting that over 340 AI tools have been approved by the FDA, primarily for
detection tasks. These tools assist in identifying conditions like brain tumors and breast
cancer, improving diagnostic speed and accuracy. However, the study also points out
limitations, including false positives and the necessity for human oversight.

3. DeepMind's AI in Ophthalmology

Research highlighted by Wired (2023) details how DeepMind, in collaboration with

Moorfields Eye Hospital, developed an AI system capable of analyzing optical coherence
tomography (OCT) scans to identify over 50 retinal diseases with expert-level accuracy.
This advancement aims to reduce the burden on healthcare facilities by prioritizing
patients in need and preventing blindness through early intervention.

Contributions from Major Companies

● IBM Watson for Oncology

IBM's Watson Health leverages AI to provide oncologists with treatment options based
on comprehensive cancer research and patient data. It assists in diagnosing diseases and
suggesting tailored treatments, enhancing the efficiency and effectiveness of cancer care.

● Google DeepMind's Retinal Diagnosis

DeepMind's collaboration with Moorfields Eye Hospital led to the development of an AI

system that accurately analyzes OCT scans, identifying various retinal diseases. This
 system aids in early detection and treatment, aiming to prevent blindness and reduce the
workload on healthcare professionals.

● Aidoc in Radiology

Aidoc offers AI-powered solutions in radiology, assisting in the detection of acute

abnormalities in medical imaging. By prioritizing critical cases, Aidoc enhances the
workflow efficiency of radiologists and contributes to improved patient outcomes.

Key Systems and Their Impact

● IBM Watson for Oncology

Watson for Oncology processes vast amounts of medical literature and patient data to
assist oncologists in making informed treatment decisions. Its ability to analyze and
interpret data supports personalized cancer care.

● DeepMind's Retinal Diagnosis System

The AI system developed by DeepMind and Moorfields Eye Hospital demonstrates high
accuracy in diagnosing retinal diseases, facilitating timely interventions and reducing
preventable blindness.

● Aidoc's Radiology AI

Aidoc's AI solutions streamline radiology workflows by quickly identifying critical

findings in imaging studies, allowing for prompt diagnosis and treatment.
Conclusion
Artificial Intelligence is already changing healthcare by combining computational power with
clinical insight to give faster, more accurate, and more personalized care. What began as simple
rule based expert systems has advanced (through machine learning, deep neural networks, and
natural language processing) into tools that can detect disease, tailor treatment, automate
workflows, and even predict public health trends. Yet these advances are not without challenges:
data privacy, algorithmic bias, regulatory gaps, and the need to preserve the human touch remain
paramount concerns.

At its heart, this paper has shown that AI in healthcare:

● Enhances Diagnostics and Treatment

From radiology to genomics, AI accelerates disease detection and helps craft
individualized treatment protocols.
● Streamlines Operations
Administrative automation and predictive analytics reduce cost, minimize errors, and
free clinicians to focus on patient care.
● Expands Access
Chatbots, remote monitoring, and wearable-driven insights extend care into underserved
and rural communities.
● Sparks Research and Innovation
AI driven drug discovery and real-time outbreak modeling demonstrate the technology’s
power to tackle urgent global health challenges.

Looking forward, the true promise of AI lies in responsibly harnessing its capabilities. Ethical
frameworks and transparent algorithms must guide development to prevent bias and protect patient
privacy. Robust regulation should ensure accountability, while ongoing clinician education will
foster trust and collaboration between human and machine. When AI is built on principles of
inclusivity and equity, it can help close health disparities rather than widen them.

In an era of rising healthcare costs, aging populations, and emerging pandemics, AI offers an
unknown opportunity to improve outcomes, lower expenses, and save lives. By maintaining a
steadfast commitment to ethical design, cross-disciplinary partnerships, and patient-centered care,
healthcare systems can fully realize AI’s potential. The future of medicine will not be AI versus
clinicians rather it will be AI and clinicians, working side by side to deliver smarter, more
personalized care to every individual.
References
● Ministry of Science and Higher Education. (2019). Introduction to Emerging
Technologies Course Module (EMTE1011/1012). Bahir Dar University; Adama Science
& Technology University; Addis Ababa Science & Technology University.

● CHG-MERIDIAN. Advantages and Disadvantages of Artificial Intelligence in

Healthcare. Retrieved from: [Link]
centre/blog/[Link]

● Zenbit. AI in Healthcare: Advantages and Disadvantages. Retrieved from:

[Link]

● XSOLIS. (2023). The Evolution of AI in Healthcare. Retrieved from:

[Link]

● Keragon. (2023). History of AI in Healthcare. Retrieved from:

[Link]

● Davenport, T., & Kalakota, R. (2019). The potential for artificial intelligence in
healthcare. Future Healthcare Journal, 6(2), 94–98.