Artificial Intelligence in Health Care

This study explores the challenges and factors affecting the implementation of artificial intelligence in healthcare. It discusses the benefits and limitations of AI in medical care, and provides an empirical analysis of AI implementation in healthcare institutions. The major challenges identified include ethical and legal complications, medical conflicts and bias, technological complications, and human rejection. The study also covers specific cases where AI is used in healthcare, such as the Da Vinci surgical system and the Babylon app. The research methods used include empirical analysis, data collection, and data analysis.

Artificial Intelligence in Health Care(ITC571)
Student University
Student name
Adm Number
Affiliation

ABSTRACT
The basis of this project was to identify the factors that are integral to implementation of
artificial intelligence in healthcare and the challenges that are slowing down the
implementation in the healthcare domain. The overall approach of the study was to determine
the challenges affecting the implementation of artificial intelligence. The study also
considered factors that are necessary in the framework for smooth adaptation of artificial
intelligence in the health care sector. An empirical analysis of the artificial intelligence
implementation was used. This analysis took into consideration the advantages and
limitations of the earlier process of implementation. Data was collected in healthcare
institutions. The data collection covered health institutions where artificial intelligence was
practiced and those that did not implement artificial intelligence. Secondary sources of data
were also used to generate the literature review. During data analysis, artificial intelligence
was implemented to compliment human intelligence in diagnoses, imaging and management
of patients. The major challenges determined during the implementation of artificial
intelligence were:
Ethical and legal complications which were but not limited to legal implications due to
failed technology and robots’ safety features.
Medical conflicts and bias., that included feeding the computed algorithms with medical
data without considerations of bias which would result in systematic errors in diagnosis
and/ imaging.
Technological complications such as autonomous weapons of mass distraction, malicious
use and systemic bugs from algorithms codes that would undermine the functionality of
the “artificial intelligence system”.
Human rejection due to the fear that artificial intelligence would render humans jobless
by substituting for doctors and radiologists.

Contents
ABSTRACT.........................................................................................................................................2
1.0 INTRODUCTION...................................................................................................................3
2.0 PROBLEM OVERVIEW........................................................................................................4
3.0 APPLICATION OF TECHNOLOGY...................................................................................5
3.1. About technology.................................................................................................................6
3.2. Research issues.....................................................................................................................6
I. Research objectives:..............................................................................................................6
II. Research questions.................................................................................................................7
III. Research methods..............................................................................................................7
IV. Data collection...................................................................................................................7
V. Ethical considerations............................................................................................................8
VI. Data analysis......................................................................................................................8
4.0. LITERATURE REVIEW............................................................................................................8
4.0 EXPERIMENTAL ANALYSIS............................................................................................14
4.1 Results......................................................................................................................................14
4.2 Factors......................................................................................................................................14
4.3 Challenges of implementation.................................................................................................15
5.0 CONCLUSIONS..........................................................................................................................16
References..........................................................................................................................................18

1.0 INTRODUCTION
Technology has already been used to improve many aspects of medical care. Application
of modern day machines and computerized equipment has been used to increase efficacy and
work output in areas such as medical records, diagnostics such as imaging and drug
development. The central aim of creating artificial intelligence was to create a system that
would pass the Turing test. The system of artificial intelligence to be created was to display
features of human intelligence with cognitive memory, empathy, judgement and the ability to
make human-like decisions. In 2010, substantial strides forward were made with the
invention and success of deep learning. Deep learning is comprised of a machine learning in
which several layers of nodes are present between output layers as well as input layers.
Therefore, simulating these output and input neuron layers is known as artificial neural
network. Artificial neural network play a vital role in underpinning many recent advances in;
(1) self-driving vehicles, (2) image classification, (3) text translation and (4) speech
recognition (Hamid, 2016).
Review of previous study state that, these types of events illustrate a real challenge for
both legal as well as ethical frameworks for sharing confidential data. 2017, has marked as a
change for artificial intelligence in healthcare sector. Also, this change bought numerous
changes in data scientists and clinicians as they are supported by clinical informatics and
started yielding positive results (Johnson, 2018).
This change made clinicians to understand about informatics in huge datasets. The
insights drawn from informatics have become an important pillar for various clinical
practices. However, there is no doubt; artificial intelligence remained overhyped in healthcare
sector at the risk of commercial explosion (Greenberg, 2017).
MYCIN received an acceptability rating of 65% by the evaluators; the corresponding
ratings for acceptability of the regimen prescribed by the five faculty specialists ranged from
42.5% to 62.5%. So, better than the human experts considered individually. However,
MYCIN failed simple because expert professionals are able to explain their reasoning, which
is essential if they are to be used for diagnosis while neural networks cannot. Deep learning
and similar approaches might be useful in two areas though: interpretation of images, e.g.
what's on this X-ray or ultrasound scan, or what type of rash is this; and undiscovered

associations, e.g. are patients who were given drug X combined with drug Y for disease P
more likely to get disease Q later on in life. Deep learning in medicine also has a downside
even supposing it works: lots of patient records are required, and anonymous ones can be
linked to the people they describe, so there's a confidentiality problem. This was the MYCIN
project, and in spite of the excellent research results, it never made its way into clinical
practice. One of the problems of expert systems was getting them adopted by end users.
2.0 PROBLEM OVERVIEW.
Artificial intelligence has made tremendous steps in the modern error of robotic
technologies. However, in the health care sector, the adaptation of the artificial intelligence
solution to efficient medical care has been painfully slow. It was therefore essential to
identify factors and challenges that influenced or hindered this systematic adoption of
artificial intelligence in the health sector. The future of medicine is in artificial intelligence.
Several developments and technological innovation are in progress to meet the need for
artificial intelligence in the field of medicine. However, these steps are met by bundles of
uncertainties that do not allow smooth implementation and usage of the robotic artificial
intelligence solutions.
The innovation and inventions of artificial intelligence system in the field of medicine has
not been fully looked into ever since because most of the innovators and researchers believe
that the most of the medical related problems which are problems affecting human beings are
not meant to be checked using the artificial intelligence systems.
This assumption has made the sector to suffer very much since most of the functioning of
the health sector needs human brains in most of the cases. This to some extend has led to the
increase in the number of deaths since wrong diagnostic using human evaluation and
reasoning might take place hence leading to the increased number of deaths.
If the same challenge might be looked into, this problem of wrong diagnosis might be
hardly present. Though some countries are employing the use of robotics in the surgery of
human beings though up to now this is not fully functioning. The above is the most challenge
that faces the field of medicine as far as the employment of technological techniques which
uses human interaction is concerned.

3.0 APPLICATION OF TECHNOLOGY.
In the field of medicine and health care delivery, technology has been used to improve
diagnostic procedures, management and drug development. This is for the simple reason that
technology has been acting as the back born of almost every process in the health care
delivery system. Straight from the patient record capture, management and data retrieval to
complex diagnosis and also currently on some surgeries using robots.
In the diagnostic field of medicine for instance, technology such as imaging using
complex machines has been used to enhance efficiency and speed of diagnosis. Machines
include but not limited to computed tomography scans, magnetic resonance imaging
modalities and ultra sounds, this has helped many patients with various medical connected
problems and illness to get quick and more accurate diagnosis since almost everything is
done by the artificial intelligence machines. This is the most effective part of medical field
where artificial intelligence has played a very practical part and has also been more efficient
in the delivery of service.
In the laboratories, machines are used to increase the efficacy of pathologic diagnoses for
cytology or tissue diagnosis.
The following are specific cases where artificial intelligence is used.
1. Da Vinci surgical system. It provides for robotic-assisted minimally invasive surgery.
A 3D visualizations system assists the surgeons to have a strong magnified view. The
surgeon controls Da Vinci ensuring that complex, tiny and routine surgical activities are of
the highest level of accuracy. By doing so, the whole process of surgery becomes so fast that
even as compared to manual process where the whole process is taken by the surgeons, this is
so fast and its accuracy is very high. The only problem with the robotic system is that its rate
of spoiling or experiencing mechanical problem is do easy. And this can take place while the
process of operation is underway.

2. Babylon in the United Kingdom.
This is an artificial intelligence app that analyses symptoms of its users and offers a
management plan. The AI tool uses patient’s past medical history, genetics and symptoms to
plan treatment. The tool also is connected to other portable devices that record patients’ heart
rate and sleep patterns and advices them on healthy living. Though some illness varies with
regions, hence making the application limited and also its accurate can be compromised when
the illness mutation takes place.
3. Retina scanner from the California Healthcare Foundation.
Using computer algorithms and data about normal and pathological retina, this tool
examines the images of a patient’s retina for signs of diabetes. This AI system allows for
quick diagnoses of subtle changes to the human retina that could be missed by the naked eye
thus reducing retinopathies as a diabetic complication.
However, artificial intelligence programs have met tremendous resistance to penetrate into
clinical practice. This formed the basis for the research.
3.1. About technology.
Artificial intelligence requires a systematic assessment in order to integrate clinical care
in routine. Authors are of opinion that the term AI (artificial intelligence) is often refer to
various machine learning techniques. In other words, artificial intelligence makes an
important difference in reality, therefore it is important to reach ANI (artificial narrow
intelligence) (Krisberg, 2017).
Artificial narrow intelligence (ANI) does more that required for completing the tasks. In
brief, artificial narrow intelligence can defeat humans in complex scenarios. International
Business Machines Corporation Watson’s victory is one best illustration for this. The
researchers also quoted that, when artificial narrow intelligence come in to existence, the
pieces will ultimately fall in to correct place (Johnson, 2018).
3.2. Research issues.
I. Research objectives:
The basis of the project was:

To determine the factors that are necessary to the process of adoption and implementation of
artificial intelligence.
To identify challenges facing the implementation and use of artificial intelligence in the
practice of health care.
II. Research questions.
What was the process of implementation of artificial intelligence?
What were the factors that the health organizations considered for smooth adoption of the
artificial intelligence?
What are the challenges that are facing the implementation of artificial intelligence in
facilities that had not yet implemented AI?
III. Research methods.
Empirical analysis was used as the primary research method. Empiric data was
quantitatively and qualitatively observed and experienced in the setting of the health
scenarios. Considering the implementations that were in place before, the advantages and
limitations that resulted from the implementation of artificial intelligence solutions in the
health facilities was recorded. Secondary sources of information were reviewed and used to
formulate the literature review.
IV. Data collection.
This being the process of gathering information from different sources and using different
methods which are available data was collected through direct observation, experience of the
artificial intelligence system in practice and questionnaires. Data was collected largely on the
implementation of artificial intelligence in the health scenario. Very few health care centers
had implemented and applied artificial intelligence in the practice of medical care even
though the numerous ways of implementation that had been brought forward and developed.
The process of data collection was focused on the scenarios which included implementation
of artificial intelligence. This data collection was instinctively based on the qualitative and
quantitative determining of the insights towards artificial intelligence implementation,

adoption for use into practice and the challenges the organizations faced before, during and
after implementation.
V. Ethical considerations.
Hospital data and patients’ data is private and confidential. Several key ethical issues
were considered in the study that included:
Respect for persons, autonomy, dignity and personal opinions.
Participants were only selected from groups of health facilities that would benefit from the
research in implementation of artificial intelligence to ensure a smooth transition.
Ethical considerations were strictly observed during the research period and data that is
open for public access and non-personal information will be used to analyze the data.
VI. Data analysis.
Since this process involves taking the assortment of the data which is collected and
recorded then now analyzing the data; Data analysis was focused in two phases. In phase one,
the implementation of artificial intelligence in the organizations that had set up a robust
mechanism for smooth transition was reviewed. In phase two, current challenges that were
encountered by the facilities that had not implemented the artificial intelligence solution were
analyzed. In accordance with the main objectives of the study, importance is based upon
finding, understanding and developing solutions to the major challenges that impact
implementation of artificial intelligence.
4.0. LITERATURE REVIEW.
In the paper “Adapting to Artificial Intelligence Radiologists and Pathologists as
Information Specialists” by Saurabh Jha, the author believes that artificial intelligence is
becoming a reality in medicine. The mimicking of human brains by computer robotics would
change the spectrum of practice of medicine. The author outlines how the fifth revolution of
technology would help improve the practice of medicine in radiology and pathology. The

writer uses the example of deep learning. He compares it to the registrar in medical
radiology. Just like how the human brain learns by interpreting more images, so does the
deep learning artificial intelligence computer. However, the author describes a challenge to
the adoption of artificial intelligence in radiology and pathology to the fears attributed to
human radiologists and pathologists being replaced by computers. Although he points out the
fear as being exaggerated, it is a challenge worth to note against implementation of artificial
intelligence.
Artificial intelligence has crept into entirely every aspect of human life. From the text
by Jeffrey Allan “Deep Learning Algorithms for Detection of Lymph Node Metastases from
Breast Cancer” in helping artificial intelligence be seen, the author notes that radiology and
pathology having to use images for diagnostics, can adopt and implement artificial
intelligence. The author describes the advantages of deploying artificial intelligence in the
practice of medicine in radiology and pathology departments by using the digital images
developed over time to develop a computerized robot loaded with the images that would be
able to detect pathologies by comparing new images to its stored images. He however notes
that apart from the advantages, limitations to implementation of the artificial intelligence
program are restrained by practical and financial obstacles. Digitalization of pathological and
radiological images would not reduce the anatomic workflow nor remove the need for glass
slides in pathology, it would therefore require additional personnel, equipment and storage
services that would increase financial requirements on the already stressed health care
system.
In the use of technology to improve the quality of health care delivery, artificial
intelligence has been deployed in the form of a robot that demonstrates ability to converse to
patients about their problems. The technology, named Gabby, is a racially ambiguous female
in her mid-forties. ( Adam Miner, Arnold Milstein and Jeffrey Hancock,2017) The software
program is designed to converse with patients with chronic pain and depression like a human
being would do. The author illustrates the feasibility of its implementation, citing advantages
in the field of psychiatry to do clinical assessments such as mental state exam and treatment.
The paper however, determined that implementation of the program Gabby faced challenges
in the presence of a conversation gap between the machine and human intelligence that would
pose a safety threat to patients and may even cause more harm.

Several clinical trials have been attempted to illustrate the need for adoption of
artificial intelligence in health care. The study by David Crawford, Joseph Batuello and Peter
Snow, November 2000 assessed the artificial intelligence methods to identify patients with
prostate carcinoma risks for lymph node metastasis. An artificial intelligence tool was created
using patient data from a database of 4133 patients with prostrate carcinoma to use the
prostate specific antigen values and Gleason sum to identify the risk levels for lymph
metastasis. The simple algorithm showed tremendous success in identifying patients at low
risk of lymph prostate metastasis. However, the tool was unable to determine the relationship
between TNM score and lymph metastasis. This was attributed to either a bug in the
intelligence program or the fluctuating values of the TNM data.
There are several diverse artificial intelligence techniques available capable of solving
many clinical problems from imaging, pathology to clinical assessments. However, despite
the optimism for developing and adoption of artificial intelligence in the practice of medicine,
this has not been met with enthusiasm (AN Ramesh, C Kambhampati, JRT Monson and PJ
Drew, Artificial intelligence in medicine,2004.) this is so because a number of previous
technologies adopted and implemented in the field of medicine such as electronic medical
records did not meet their projected advantages.
Proposed solutions to the issues by authors.
1. Replacement of medical professionals by computers.
Medical data can be categorized into data that can be extracted by the eyes and that
can be extracted by computer software. Use of computer software in extraction of data is the
basis for development of artificial intelligence programs that can use the extracted data to
form decisions concerning diagnosis and management. This description moves the art and
science of medicine from a subjective display of skills to an objective science. The
development and invention of more and more information has equipped medical
professionals with more knowledge. However, this addition of data has challenged the
computational abilities of the physicians, radiologists and pathologists due to the great
quantity and density of data. Thus development creates a path for necessitation of computers
to supplement the medical professionals. The computers would enable of fast and accurate
acquisition of data from medical situations that would otherwise mot be obvious to the human
eye. In order to adapt to artificial intelligence and avoid replacement, the author Saurabh Jha,
2016 recommends that the medical professionals allow to be displaced by the computers.

This new situation would allow categorization of medical into automated computer extracted
data and human analysed data. In this situation, thus, computers would only compliment
medical professionals in the delivery of quality health care. The reservations revolving
around the development and implementation of the artificial intelligence technology should
not be waivered off. These challenges have formed from the basis that most of the medical
professionals do not understand the technology well. As a human instinct, we do not accept
what we do not understand. For a smooth adoption and implementation of the artificial
intelligence technology in medicine, all the health care professionals should be included in
the design framework. This inclusion would include training of the professionals and
equipping them with enough knowledge to understand the complimenting functionalities of
the artificial intelligence. By so doing the most part that will be remaining will be the control
of the computers and also decision making and implementation of the decision of choice that
the computer based system has given out.
2. Increased financial requirements of the implementation.
The current health care system is constrained by financial stresses that prevent
maximization and offering of top quality service of care. Artificial intelligence is a new
technology that comes with its own financial implications and requirements. The
requirements of digitizing many aspects of the medical field in order to accommodate
artificial intelligence and utilize the program maximally would require a huge financial input
into the health care system. For example, from the text by Jeffrey Allan, 2017, digitization of
the radiological images would increase the success of the artificial intelligence. However, in
the same text, the digitization would not work in the field of pathology because there is no
substitute for glass slides in pathology lab diagnoses. The acquisition of more personnel,
computers and machines and storage facilities would be a huge financial constraint. To
maneuver around this task-hill, the author recommends a slow process of implementation of
artificial intelligence that Is spread over a period of time. A number of artificial intelligence
tools acquire donor funding and government support initiatives after showing great steps
towards improving the quality of health care delivery. The quality of service at any health
care is determined by how much have we spend in keeping and loading the facility every
before. This is so important as the facilities of the health care delivery system is very
expensive and therefore needs sacrifices I order to equip the system for smooth handling of
the problem which comes across.

3. Functional gap between machine and human intelligence.
The difference between human responses and machine responses has been greatly
reduced. The way that this has been achieved has never been made clear to the consumers of
the products. The challenge facing implementation of such programs if artificial intelligence
lies within the robots’ inability to respond to the needs both conversational and emotional of
the patients. This shortcoming is projected to cause ineffective care and harm to patients and
therefore affects the streamline implementation of the technology for example The Gabby.
The machines’ inability to pass the Turing test becomes a huge set back.
In the adoption of artificial intelligence, the author suggests randomized clinical trials to
compare the effectiveness of the artificial intelligence technology with its implications on the
cost and safety of the patients. The process of developing such systems is not an easy and
smooth task. Therefore, a lot time and human effort is recommended to write the codes for
such algorithms and data should also be extensively and intensively collected about the
different disease processes to ensure a close to human intelligence performance is obtained.
4. Technological precincts.
Artificial intelligence programs use algorithms for form a set-tree of decision making and
formulations. The algorithms analytical tools that act on data by performing many operations
such as comparisons. However, the data stored in always past data from the patients. The
algorithms perform analytical operations. Since the artificial intelligence are a set of
instructions, are written by human to try and merge the informal language of human to the
formal language of computers to form an organized system. These written algorithms are
prone to system failures and bugs. The delivery of health care and the art and science of
medicine backers for the do no harm, principle. Since the artificial intelligence programs are
manmade computer programs are bound to periodic systemic failures and thus one cannot
predict the extent of damage that might result from such technical errors and glitches. The
medical field is a very conservative field with the primary objective of maintaining patients’
safety. The set of procedures practiced from the old times and are very efficient are
documented. Therefore, the prospect of using machines to carry out the same is a subject of

serious debate due to the argument that technology fails. This time of belief protects the
patients but reduces the chances of innovation as the medical field is not ready to risk new
technology.
5. Data availability limitations.
Artificial intelligence systems are built on algorithms that depend on hospital and patient
data to form bases for analysis and decision are made from the analysis. There data
challenges are primarily with the highest impact in the development of artificial intelligence.
Although steps have been made using the available data, the systems manufactured and
implemented in the health facilities are not on 100% data and information concerning the
diseases. Data has been collected about diseases and forms the biggest part of the resource for
the decision making process of the artificial intelligence systems. However, the data is not
sufficient enough to design a system of artificial intelligence that would otherwise replace or
displace the medical professionals. Artificial intelligence systems therefore require more data
that is efficiently organized, that is currently not available.
While the process of acquiring of data in the health care sector is pretty difficult with a lot
legal and ethical consideration, a lot has been achieved in the development of artificial
intelligence. However, such systems are not set at optimal working capabilities hence failures
in the Turing test or in the clinical trials renders them disapproved by the FDA.
6. Misapprehensions and exaggerated capabilities of technology in medical practice
reduces the enthusiasm towards the technology.
The term artificial intelligence is misleading in the field of medicine. At the moment, the
system is not that highly developed as the term itself hypes. Artificial intelligence as at the
moment is as system that is capable of detecting changes in patterns. These capabilities are
used to create reports of medical assessments such deviations from the normal. Artificial
intelligence is hyped with the belief that the system can extract concepts and knowledge and
make informed decisions about life and medical wisdom. With the failures of the previous
technologies that were hyped in advertisements and the media such as electronic health
records, that destroyed the healthy and public image about artificial intelligence. The solution
to this is to provide understanding of the capabilities of the artificial intelligence technology.

Consumers need to understand that the system is dependable on human intelligence to
achieve ,maximum performance.
4.0 EXPERIMENTAL ANALYSIS
4.1 Results.
The projected took a duration of 31days to completion. A total of 112 respondents,
managing directors and chief executive officers of the major health centers in the country
were interviewed on the factors they would consider necessary before adopting the artificial
intelligence system. The responses were analyzed and categorically placed according to rank.
The respondents who had adopted and implemented artificial intelligence in their practice of
medicine were asked to narrate their experience in the use of the system focusing on the
challenges they faced during the process. In centers without the artificial intelligence
programs in place, respondents described the factors that inhibited the adoption process or the
challenges they faced during the adoption process that resulted in ditching the whole system
all together.
4.2 Factors.
The following factors were identified as requirements for the smooth implementation
of the artificial intelligence.
Trained personnel and technical expertise: 81% of the respondents ranked technical
expertise as one of the primary factors that enabled the adoption and implementation of
artificial intelligence. The chief executive officer and medical supervisor of the brain and
spine hospital complained of limited human resources to the adoption of the artificial
intelligent system.
Adequate infrastructure and computing power: artificial intelligence require specific
machines and an ambient environmental set up that would enable reliability, efficiency and
speed of the system. Since the technology also involves massive computations and data
analysis that should be accurate and fast, 795 of the respondents ranked the need for super

computers with high power computational abilities as a factor necessary for adoption and
implementation of artificial intelligence.
Financial muscle: financial resources in form of startup costs, installation costs,
maintenance costs and repair costs were ranked top of the factors affecting adoption and
smooth implementation of the artificial intelligence.
Enthusiasm, desire and willingness to risk new technology: The medical profession
has remained strictly conservative over the years. Strict observation of medical procedures to
protect the patients are practiced. Therefore, the health care professionals and health facility
management are very reluctant to risk new technology on the patients and thus this practice
kills the innovative drive in the medical sector.
4.3 Challenges of implementation.
I. Job losses and relocations.
Cited as the major complaint against implementation of medical artificial intelligence
system, the human fear of losing jobs in large scale due to their jobs being substituted by the
computers. Another concern raised that impacted negatively on the implementation of the
artificial intelligence system was the possibility of job relocation to other positions of work
that would require more creative skills than the knowledge acquired through the education
system. Despite the fact that there would be a need for employment of technical expertise or
training on skills related to artificial intelligence, the idea of implementing the system that
would see “computers take away” peoples jobs, was a matter of fierce debate. The
implementation and introduction of computer based systems in most of the fields which are
related to medical has been receiving set backs since human beings could replaced with the
machine hence creating unemployment to the country. This is the major reason as to why the
implementation is taking too long to be completed
II. Medico-legal and ethical concerns of responsibility.
Medical practice is a very conservative field. The principle of “do no harm” to the patient
are strongly held and passed from one generation of medical professionals to the next. With
this principle, issues of professional negligence, errors and/ natural disasters were easier to
determine and find the individual for legal or ethical punishment in case of intentional human

error. However, the implementation of artificial intelligence raises the question of “who
should be implicated when technology fails and does harm to the patients?”.
III. Issues of technology and robotics.
Artificial intelligence is unable to display other aspects of human intelligence apart from
knowledge, such as emotions. Such disabilities question the ability of the artificial
intelligence to make decisions that affect patients that require the basis of emotions such as
love, sympathy and empathy. Another serious issue of contention is when the artificial
intelligence system or bots become more intelligent than humans and become dangerous or
some malicious individual takes control to cause destruction. Most of the robotics are made
by humans and can be malfunctioned or confused, this machines are not perfect and cannot
employ full reasoning in decision making. For complex diagnosis the artificial intelligence
system might give out wrong findings in relation to the specific fact impacted.
IV. Patients’ data privacy and confidentiality.
Serious uncertainties are placed on the implementation of artificial intelligence and its
effects on the confidentiality and privacy of patients’ data and medical records. The use of
technology, as determined by the failures of previous technologies such electronic medical
records, on the protection of individual privacy evokes extreme reservations for the
implementation of artificial intelligence.
5.0 CONCLUSIONS
Artificial intelligence is the future of medical practice. Features of human intelligent
can be represented by the formal computer language and form algorithms that would be used
to analyze medical aspects and form decisions like the expert medical professionals do. From
the clinical trials, artificial intelligence can outperform medical professionals and thus
becomes a challenge against itself implementation as a new form of technology. The factors
limiting the smooth adoption and implementation of artificial intelligence in healthcare are
budgetary requirements, infrastructure and computing power, technical expertise and the will
to try new technology. The top ranked challenges impacting the implementation of the
artificial intelligence is the fear for job losses due to substitution by computers, technological
limitations such as robots being excessively intelligence than humans and thus becoming
autonomous weapons of havoc and mass destruction and the privacy reservations once the

artificial systems have been implemented. Artificial intelligence spells quality, fast and
affordable future health care. However, the process requires a lot of input in terms of data
about diseases, patients’ data including risk factors and prevalence and the computed
algorithms to enable analytic calculations of determinants of medical judgement and
decisions to compliment the medical professionals in delivery quality health care.
From this , its is therefore recommended that the implementation of artificial
intelligence should be employed with no intentions of replacing human effort and assistance
in the health sector as this can make matters more worse than the way it is now.
The employment of the robots to take part most of the minor processes will fasten and lighten
the whole processes that are taking place in the health care delivery system.

References.
Bennett, C. C., & Hauser, K. (2013). Artificial intelligence framework for simulating clinical decision-
making: A Markov decision process approach. Artificial intelligence in medicine, 57(1), 9-19.
Greenberg, D. A. (2017). Artificial Intelligence in Health Care: Are the Legal Algorithms
Ready for the Future? McGill Journal of Law and Health.
Hamid, D. S. (2016). The Opportunities and Risks of Artificial Intelligence in Medicine and
Healthcare. Communications Summer, 1-4.
Patel, V. L., Shortliffe, E. H., Stefanelli, M., Szolovits, P., Berthold, M. R., Bellazzi, R., & Abu-Hanna,
A. (2009). The coming of age of artificial intelligence in medicine. Artificial intelligence in
medicine, 46(1), 5-17.
Hider, P., & Pymm, B. (2008). Empirical research methods reported in high-profile LIS
journal literature. Library & Information Science Research 30, 1-7.
Jeremy Corbett. (2017). Is AI for Healthcare Going Too Far? HealthTech.
Jiang, F., Jiang, Y., & et.al. (2017). Artificial intelligence in healthcare: past, present and
future. BMJ Journals.
Johnson, S. R. (2018). Realizing AI: AI's role in healthcare starts small, gets bigger. Modern
Healthcare.
Krisberg, K. (2017). Artificial Intelligence Transforms the Future of Medicine. AAMC.
Lee, B., & Cassell, C. (2013 ). Research Methods and Research Practice: History, Themes
and Topics. International Journal of Management Reviews; Volume15, Issue2.
Mesko, B. (2017). The role of artificial intelligence in precision medicine. Expert Review of
Precision Medicine and Drug Development;Volume 2, Issue 5, 239-241.
Peytchev, A., & Peytcheva, E. (2017). SURVEY RESEARCH METHODS. journal of the
European Survey Research Association; Vol 11, No 4.
Zang, Y., Zhang, F., Di, C. A., & Zhu, D. (2015). Advances of flexible pressure sensors toward
artificial intelligence and health care applications. Materials Horizons, 2(2), 140-156.