AI in health care
AI in health care
AI in health care
A review of literature
Viet Le
Abstract
2
This literature review compares the potential of AI in healthcare support with the obstacles they
bring about whether it is worth integrating them or not. Although the potential of AI in healthcare
services is undeniable, there are still many problems discovered during the application process in
services that have not been completely resolved. By reviewing a number of studies to find the
personalize treatment, assisting to improve users’ experiences and lifestyle. Along with that,
issues were pointed out such as security and privacy issues, transparency of AI, quality of
information provided by AI and lack of experts in verifying instructions given by AI. This review
shows that AI’s ability to analyze large amounts of data and identify certain patterns can help
diagnose diseases faster and more accurately. With patient information and data, AI can help
provide better personalized treatments (suitable for each patient’s condition). The availability of
addition, this research clearly points out the drawbacks along with several solutions to improve
the integration of AI into healthcare support. Therefore, although there are still problems to be
solved, the potential of AI is still the most important because the benefits they bring are also
significant.
A review of literature
Artificial intelligence (AI) is a field of computer science and is made up of various pre-
written algorithmic techniques that enable them to function like human intelligence in certain
specific processes. AI was first developed in 1950 by Christopher Strachey (Alowais A., 2023).
Because it is a newly developed program that is still in its early stages, AI is just a topic for
researchers. Until now, AI has been attracting attention from many people in different
The techniques that make up AI such as machine learning (ML), deep learning (DL), and
large language models (LLMs) provide many features that can be applied to many different
situations. For example, analyzing large data sets, analyzing image recognition, or analyzing
images based on thermal sensors, communicating in human language, etc. are all capabilities that
the above techniques bring to AI. Not only that, AI is also an attractive research topic in science
and technology to support and perform various human tasks with the support of machines. They
can also perform many different tasks at the same time without making any mistakes. With such
capabilities, AI can help people cut down on boring and labor-intensive tasks, as well as save a
Because of these features, the potential of AI is enormous, not only for computer science
but also for all other professions, specifically healthcare. This research paper will focus mainly
on the potential of AI in healthcare and compare it with the limitations and challenges in
integrating them. By doing so, this review aims to contribute a better understanding of the role
and importance of AI and provides evidence regarding the benefits of integrating AI into
healthcare practice.
Disease diagnosis
Healthcare is one of the most indispensable services in everyone’s life, especially in this
era when there are more and more diseases as well as living environments that have negative
effects on human health. Therefore, the increasing demand for health care services of every
person in the world leads to many different problems that doctors and health experts have to face,
such as having too many patients may lead to diagnostic errors or delays due to overload. By
using AI, health care practitioners will be empowered to deal with vast amounts of data from
patients, thereby being able to give earlier and more accurate diagnostics for many kinds of
diseases.
To demonstrate that AI analysis is accurate and fast, a study was conducted on the
diagnosis and treatment of breast cancer. Research published in the UK shows that, by using AI
to analyze identification based on patients’ visual input, the rate of correct diagnosis was
significantly increased, along with that minimizing misdiagnosis rates. (Alowais A., 2023).
Another method is also mentioned by Kitsios F., Kamariotou M. et al. (2023) in their research
paper, a model using linear discriminant analysis (LDA) as a feature reduction method is
proposed by Gayathri to determine a woman’s odds of developing breast cancer. The result
shows that this model is successful in its efforts to shorten the amount of time needed for
(2022) stated that doctors can be widely assisted by using machine learning techniques for
diagnosis, classifying, and predicting tumors during early detection and decision making. All the
above studies tested the image analysis and recognition features of AI, more specifically machine
Personalized treatment
5
Personalized treatment is one of the services where AI is expected to have the most
significant changes and developments. This process requires close monitoring by medical staff
and a large amount of information about each patient so that staff can take care of each patient
based on their condition, such as according to genes, environment, or their lifestyle, etc.
Therefore, using AI will save human resources, a huge amount of costs and time for healthcare
services. Experts can use AI or integrate it into technology and machines to monitor and then
analyze patient behavior and data, so that decisions are then made regarding what action is
necessary or can be used to alarm warnings and preventative measures (Rong G., A. Mendez, E.
Bou Assi et al., 2019). In in research article by Alowais A. et al. (2023), the authors discussed
treatments based on their genes, medical history, and other factors. Two treatment approaches,
using chemotherapy for cancer patients and antidepressants for depressed patients, were carried
out by Huang et al and Sheu et al., respectively, all affirm the important role of AI when their
In addition, AI can also be combined with other technological machines to better support
human life. An experiment was conducted with smart robot systems with a proposal to create
smart home functions and tools specifically for the elderly, disabled people, or pregnant women
(Rong G. et al., 2019). The authors argue that by monitoring and analyzing the user’s actions and
gestures with various intelligent systems of the developed “ambient assisted living” (AAL)
model that enables AI to process information collected from various sources, machine learning
techniques can provide warnings about dangerous situations and make clear preventive
measures.
References
Artificial intelligence (AI) in healthcare and research (2018, May). Nuffield Council on
Bioethics.
7
Davenport T. (2019 June). The potential for artificial intelligence in healthcare. Future Healthc J.
https://doi.org/10.1080/20479700.2021.1886478
Kitsios F. (25 June 2023). Recent Advances of Artificial Intelligence in Healthcare: A Systematic
Rong G. (2019). Artificial Intelligence in Healthcare: Review and Prediction Case Studies.
Elsevier LTD on behalf of Chinese Academy of Engineering and Higher Education Press
Secinaro S. (2021). The role of artificial intelligence in healthcare: a structured literature review.
01488-9