MACHINE LEARNING FOR SMART HEALTHCARE
ANALYSIS: A MULTIMODAL APPROACH.
Abstract:
Contents
INTRODUCTION ....................................................................................................................................... 4
Background to the Study ..................................................................................................................... 4
Significance of the Study ..................................................................................................................... 5
Aims and Objecves ............................................................................................................................ 5
Research Quesons ............................................................................................................................. 6
LITERATURE REVIEW ............................................................................................................................... 6
Healthcare Current Status and Challenges ......................................................................................... 6
Smart Healthcare System .................................................................................................................... 7
Definion and Types ....................................................................................................................... 7
History ............................................................................................................................................. 8
Framework and Advancement in Smart Healthcare System. ............................................................. 8
Smart City for Smart Healthcare ..................................................................................................... 8
Internet of Things for Smart Healthcare. ............................................................................................ 9
Arficial Intelligence for Smart Healthcare. .................................................................................. 11
Machine Learning for Smart Healthcare ....................................................................................... 12
Deep Learning and Machine Learning Algorithms are used in Healthcare Systems. ........................ 14
Methodology Principles for Smart Healthcare .................................................................................. 16
User Centred Principle .................................................................................................................. 16
Mul-Modal Principles .................................................................................................................. 16
RESEARCH METHODOLOGY ................................................................................................................... 17
Research Philosophy ......................................................................................................................... 18
1. Positivism .............................................................................................................................. 18
2. Pragmatism ........................................................................................................................... 18
Research Approach: I shall adopt the deducve approach. ............................................................. 19
Deducve ...................................................................................................................................... 19
Research Strategy ............................................................................................................................. 19
Case Study ..................................................................................................................................... 19
Research Method .............................................................................................................................. 19
Mixed method ............................................................................................................................... 19
Research Time Horizon ..................................................................................................................... 19
Research Techniques and Procedures: .............................................................................................. 20
Data Collecon: ............................................................................................................................. 20
Data Analysis: ................................................................................................................................ 21
CONCLUSION ......................................................................................................................................... 22
REFERENCES .......................................................................................................................................... 23
INTRODUCTION
Background to the Study
Good health constitutes an essential aspect of socio-economic development and a
prerequisite for high levels of productivity. The basic indicator of the quality of life, and
economic and social well-being in a society is the quality of its population’s health, that
is, the extent to which the different strata of its population cohorts enjoy good health
and have access to good health service infrastructure (Nian et al., 2015; Pasluosta et
al, 2015).
The desire of nations to give their citizens good healthcare service has been
challenged by several factors occasioned by a rapidly increasing and aging population,
a rise in chronic diseases, and a shortage of manpower. The strain on healthcare has
brought into focus the potential benefits that can be gained from the application of
digital technology, Machine Learning, Artificial Intelligence (Australian Institute of
Health and Welfare, 2014; Sicari et al., 2017; Gope and Hwang, 2016)
The Internet of Things (IoT) is a system made up of actual objects in the real world
that use networking technologies to interact and communicate with one another. IoT
has fully involved in the healthcare domain because it is suitable for remote health
monitoring and cloud services. Artificial intelligence approaches have also been
integrated into medical operations, supporting medical staff in image processing,
diagnostics, and prophylaxis. Machine learning is a subset of artificial intelligence (AI)
that focuses on the development of algorithms and statistical models that enable
computer systems to learn and improve their performance on a specific task (Shinde,
A., et al., 2022).
Smart healthcare is a term that has been coined to explain the application of
technology in healthcare service, research, development, and management. Tian et
al (2019) define smart healthcare as the use of front-line technology, like the Internet
of Things, artificial intelligence, cloud computing, and a set of digital tools to better the
ways we maintain our health and manage our healthcare systems. The group of
people saddled with the responsibility of implementing the smart health make up the
smart healthcare system. A healthcare system, therefore, consists of certain groups
(such as patients, primary care physicians, pharmacists, and other specialists) and a
variety of stages (such as medical issue screening, sickness determination, clinical
therapy, and recovery). Raju and Kumar (2022). Together, all these parties contribute
to the implementation of the smart healthcare ecosystem.
Significance of the Study
Healthcare is an important aspect of smart city development, and the incorporation of
AI technologies has the potential to revolutionize healthcare service delivery in these
areas. This study presents a multi-modal approach to analysing the various machine
learning algorithms for the development of a smart healthcare system extending its
result to Cardiff, a city that has shown concerted efforts to improve the quality of life
and well-being of its people but lacking in specific scientific test models. It is believed
that the ideas that will be presented in this study will aid further research. It will form
the blueprint for the development of a medical app and guide the healthcare system,
Cardiff city planners, policymakers, and urban and health researchers on the cutting-
edge machine learning models in smart city and smart healthcare
Aims and Objectives
This research aims to examine various machine learning approaches for smart
healthcare systems and draw conclusions about Cardiff. This will be achieved with the
following objectives:
1. To identify suitable machine learning algorithms for smart healthcare analysis
2. To predict risk to stroke based on multi-modal patient data.
3. To predict risk to heart disease based on multi-modal patient data.
4. To examine the degree of accuracy of predicting prostate cancer with deep
learning models.
5. To determine the most suitable machine learning algorithms for different types
of health output data type.
6. To advise Cardiff city planners and policymakers on proven machine learning
algorithms for smart healthcare and smart cities.
Research Questions
1. What are the suitable machine learning algorithms for smart healthcare
studies?
2. To what degree can biomedical factors predict the risk of stroke?
3. To what degree can biomedical factors predict the risk of heart disease?
4. How accurately can the deep learning model predict the risk of prostate cancer?
5. Which are the most suitable machine learning algorithms for different types of
output data type?
6. What outcomes of the study can be best applied to smart care development in
Cardiff
LITERATURE REVIEW
Healthcare Current Status and Challenges
1. Low Quality
The lack of access to good-quality healthcare results in the deaths of over 30,000
people annually (Demirkan, 2013). According to the Institute of Medicine (2001), six
factors make up the quality of healthcare: safety, efficacy, efficiency, personalization,
timeliness, and equity. The primary reason for the low quality is the dearth of reputable
medical facilities and qualified doctors. Professional doctors are willing to diagnose
and treat patients more extensively, but their time and energy are limited. As a result,
only a certain percentage of patients in most of the world can receive the anticipated
diagnosis and treatment (Huan, et al., 2018)
2. Ineffective Method of Diagnosis:
Information retrieval from patients is one of the factors contributing to ineffective
diagnosis and treatment processes. Doctors always begin the diagnosis process by
obtaining the necessary information from their patients, particularly when the patient
is seeing a new doctor or hospital. Both information related to treatment and
information unrelated to treatment are required. If there is a secure information-sharing
platform, the time spent repeatedly obtaining and recording patient information could
be saved because the information typically remains the same within a given amount
of time, (Huan, et al., 2018).
Not using prior diagnosis data is another factor contributing to ineffective diagnosis
processes. Patients frequently have to undergo additional testing as a result of lost
data. The deficiency may be the cause of the loss. (Huan, et al., 2018).
3. Aging Population
The aging population is contributing to an increase in certain disease types, which
means that there will be a shortage of physicians in general and specialist physicians
in particular for diseases like diabetes, osteoporosis, stroke, cancer, heart disease,
arteriosclerosis, and chronic renal failure. The conventional healthcare system gives
little thought to the numerous remote health condition monitoring, disease condition
monitoring, and psychological support (Huan, et al., 2018).
Smart Healthcare System
Definition and Types
Huan, et al., (2018) define smart healthcare as an integrated system in which medical
professionals, patients, monitors, AI, and other components are all linked and
communicate with one another to improve healthcare quality, lower costs, and
personalize treatment.
Asim
(2022) defines smart healthcare as the seamless integration of intelligent
technology with healthcare systems to facilitate the retrieval of healthcare data and
establish connections between resources to effectively manage fluctuations in
healthcare demand.
Huan, et al., (2018) believe that there are two different kinds of smart healthcare
systems based on the functions they perform: patient-assisting smart healthcare
systems and physician-assisting smart healthcare systems.
History
Huang, et al., (2018) traced the history of smart healthcare to Snell's (1996) censor-
based monitoring method and the device-based patient management platform of
Costanzo and S. Pamboukian (2009). Big data analysis, machine learning, cloud
computing technology, and the Internet of Things (IoT) are all developing quickly in
the twenty-first century as a result of the exponential growth in device processing
power, and the impact on smart healthcare systems has been enormous.
Framework and Advancement in Smart Healthcare System.
Smart City for Smart Healthcare
Smart healthcare is a major component of the parent concept of a smart city.
Figure 1: Components of a smart city.
Source: Hassan et al. (2023)
According to Herath and Mittal (2022), a smart city is a framework that makes use of
data and technology to address issues like waste management, traffic congestion, air
pollution, health issues, energy use, etc to improve the quality of life of a people.
The smart city concept argues that a city consists of interacting and interconnected
entities and that the best way to manage these entities and improve the quality of life
of city dwellers with the help of artificial intelligence and the IoT enabled devices,
(Hassan et al, 2023; Al-Azzam and Alazzam, 2019).
There has been some research done on the smart city concept:
1. Huang et al., (2018) attempted a holistic study of smart city systems, the study
outlined how smart healthcare systems interact and cooperate with smart city
infrastructure, as well as how it benefits the healthcare industry. Huang et al.,
(2018) also reviewed several case studies, which offer insights into how smart
healthcare systems can be combined to create a more potent, cohesive, and
efficient system.
2. Grunhut et al., (2021) propose a smart home healthcare monitoring system that
can be installed with sensors and devices to meet the needs of senior citizens
who require ongoing at-home care.
3. Cugurullo (2020) conducted research into the theory and application of artificial
intelligence to smart city development. The paper offered a theoretical
framework for the understanding of AI in the context of urbanism. Cugurullo
(2020) suggested a research agenda for the study of an autonomous city after
looking at the case study of Masdar City, an Emirati urban experiment.
Internet of Things for Smart Healthcare.
Patel and Patel (2016) believe that IOT is a network of physical objects. In addition to
being a network of computers, the internet has grown to be a network of all kinds of
devices, including smartphones, cars, toys, medical equipment, cameras, industrial
systems, and buildings. These devices communicate and share information according
to predetermined protocols, enabling real-time online health monitoring, process
control, and administration
Figure 2: The Internet of Things
Source: Patel and Patel (2016)
Figure 3: The Value Potential of the Internet of Things
Source: International Society of Automation (2015)
The importance of the IoT mechanism is felt in every our present day society. IoT
solutions have made significant entry into the healthcare sector, where they are now
incredibly essential. Some research in IoT includes:
1. In November 2017, Abilify MyCite launched the first FDA-cleared smart pill that
timestamped only when the drug was taken. As soon as the pill comes into
contact with the patient’s gastric acid, it triggers a sensor that marks the time of
contact and first forwards this information to the wearable sensor and finally to
the mobile phone app. (Shah and Chircu, 2018; Ghazal, T.M. et al., 2021).
2. Bhandari et al (2020) examined the use of robotic surgery and mixed reality
technology to facilitate more accurate operations.
3. Using a voice signal as input, Alhussein (2017) created a remote monitoring
system for patients with Parkinson's disease.
Artificial Intelligence for Smart Healthcare.
Artificial intelligence (AI) describes systems that exhibit intelligent behaviour by
analyzing their surroundings and acting, sometimes autonomously, to accomplish
predetermined objectives, COM, 2018).
Artificial intelligence (AI) can be found in both software and hardware systems.
Software-based systems can function in the virtual world and include voice assistants,
image analysis software, search engines, speech and face recognition systems,
advanced robots, autonomous cars, drones, and Internet of Things applications.
(COM, 2018).
Figure 4: Components of Artificial Intelligence
Source: Kumar V and Sheshadri K., (2019)
Artificial Intelligence has several components, this component has the potential to
revolutionize our healthcare systems. Research in artificial intelligence for smart
healthcare system include:
1. Kharel et al. (2017) proposed a model for a smart health monitoring system
based on fog computing. The proposed architecture acknowledges the
underlying problems of a clinic-centric healthcare system and advocates a
change to a smart patient-centric healthcare system.
2. Bhandari et al. (2020) studied how Al-based clinical decision support systems
are continuously used to aid and improve diagnosis. The paper also examined
the use of robotic surgery and mixed reality technology to facilitate more
accurate operations
3. Ashfaq et al. (2022) employed artificial intelligence (AI) and the Internet of
Things (IoT) to remotely monitor cardiovascular patients using a gadget that
enables intelligent monitoring of human vitals like body temperature, heart rate,
and oxygen levels.
Machine Learning for Smart Healthcare
Machine learning was coined in 1959 by Arthur Samuel, an American pioneer in
artificial intelligence and computer gaming. Samuel defined machine learning as "the
field of study that gives computers the ability to learn without being explicitly
programmed." (ElNaqa & Murphy, 2015).
Machine Learning (ML) is a very important component of artificial intelligence and the
algorithms for Machine Learning are into three categories:
(A) Supervised learning: In supervised machine learning, programmers give the
algorithms a ready-made dataset to train from, along with an output that is
known to be correct. Regression and classification models are the two types of
supervised learning categories.
A predetermined data set with known outputs and inputs, that is where the
input and the output are known, is necessary for supervised learning to take
place. Supervised learning models are grouped into regression and
classification models.
(b) Unsupervised learning: Unsupervised machine learning uses provided unlabeled
data as its basis. Thus, the learning algorithm looks for structure in the provided data.
The clustering algorithm is the primary unsupervised machine learning algorithm.
(c) Reinforcement learning: No prior data is needed for reinforcement learning.
Rather, during training, they are created and labeled through multiple runs in a
simulation environment using a trial-and-error method, (Ghazal, et al, 2021)
Figure 5: Classification of Machine Learning Algorithms
Source: Ghazal, et al, (2021)
Deep Learning and Machine Learning Algorithms are used in Healthcare
Systems.
According to Molly (2021) deep learning is a subset of Machine learning using large
artificial neural networks. These neural networks are made up of multiple layers,
usually comprising input and output layers with a few "hidden" layers inserted in
between. Each of the many artificial neurons that make up each layer is numerically
weighted, connected to other neurons, and capable of responding to input from
neurons in the layer above, (Ghazal, et al, (2021)
Figure 6: Relationship between Artificial Intelligence, Machine Learning, and Deep
Learning
Source: Molly (2021)
Today machine learning and its subset of deep learning have gained significant
importance in health monitoring and prognosis, treatment of the acutely ill, decision
support systems, treatment of chronic illnesses, and in Respite Care. Some of the
machine learning algorithms used in health care are shown below:
Figure 7: Some machine learning algorithms in the healthcare industry
Source: Jiang, et al, (2017).
Some research in machine learning for healthcare is presented below:
1. Li, et al. (2021) did a thorough analysis of the uses of machine learning methods
for big data analysis in intelligent healthcare systems, the paper outlined a
number of the advantages and disadvantages of the current strategies, with an
emphasis on the difficulties facing this area of research.
2. A thorough analysis of published surveys utilizing deep learning-based
techniques for classifying brain tumors was provided by Muhammad et al.
(2021). The paper covered all of the key stages, such as preprocessing, feature
extraction, classifications, accomplishments, and limitations. The paper also
conducted in-depth experiments using transfer learning with and without data
augmentation to look into cutting-edge convolutional neural network models for
BTC.
3. Hassan et al (2019) believe that by using artificial intelligence technology,
health- related data such as vital signs, physical activity levels, and sleep
patterns can be automatically collected, analyzed, and used to detect health-
related issues.
Methodology Principles for Smart Healthcare
User Centred Principle
Health monitoring systems that are created with the end user in mind are known as
user-centered Al monitoring systems. To enhance patients' general health and
wellbeing, these systems use artificial intelligence (AI) technology to gather, process,
and analyse health-related data, (Hassan et al, 2023).
Multi-Modal Principles
Single-modality approaches have been the foundation of healthcare for a long time,
which limits the information available to make medical decisions. But thanks to
technological developments and the availability of a variety of data sources, it is now
possible to integrate different modalities and obtain a more thorough understanding of
patients' conditions. In multi-modality principles, different data types are combined and
analyzed. Multi-modality principles involve the fusing and analysis of several data
types and data sets such that the different data sets and types can interact and inform
each other, and give a better understanding of patients' conditions (Lahat, Adalı and
Jutten, 2015; Salvi et al., 2024)
RESEARCH METHODOLOGY
I have adopted a research methodology suitable for answering my research questions
and achieving my research aims and objectives. I shall discuss my research
methodology taking a cue from the onion framework popularised by
Figure 8: Onion Research Methodology Framework
Source; Saunder et al., (2015)
Research Philosophy: I shall adopt two philosophies
1. Positivism: Rooted in empirical observation and the scientific method,
positivism involves quantifiable data and objective analysis. Here, the focus
would be on measuring and quantifying the impact of machine learning
algorithms on healthcare metrics such as accuracy, speed of diagnosis,
treatment effectiveness, etc. I shall provide answers to my research questions
and attempt to meet my research objectives using quantitative data analytics
techniques, (Saunder, 2015).
2. Pragmatism: This approach has a combination of qualitative and quantitative
approaches. I shall be qualitative to draw empirical findings to Cardiff.
Research Approach: I shall adopt the deductive approach.
Deductive
In a machine learning analysis project, a deductive research approach begins with a
well-defined theory or set of hypotheses, which are then tested and confirmed through
data collection and analysis. This method is distinguished by a top-down procedure in
which you start with a broad assumption and work to derive conclusions from it
(Saunder, 2015).
Research Strategy – I shall adopt the case study strategy.
Case Study
I shall adopt the case study strategy. According to Saunder (2015), case studies
concentrate on a close analysis of a particular person, organization, or phenomenon
in the context of real-world events. This tactic entails in-depth analysis, frequently
drawing from a variety of data sources, including observations, interviews, and
documentation, to provide thorough answers to the research questions
Research Method: I shall adopt the mixed method.
Mixed method
My research methods shall be a combination of both qualitative and qualitative
research methods. To get the desired result I intend to be numerical and non-
numerical, this is because I am applying machine learning to healthcare, and man is
at the centre of my studies. Since man is a complex being, being quantitative alone
will not address the problem statement, (Saunder, 2015).
Research Time Horizon
Cross-sectional: I am adopting the cross-sectional, this is because:
1. This study is carried out over a brief period and I am doing everything at once.
2. To comprehend the relationships between variables and answer my research
questions I shall examine data relating to diverse population cohorts in time.
Research Techniques and Procedures:
This study attempts an analysis of machine approaches to a smart healthcare system.
Its will conclusions will be drawn to aid healthcare stakeholders in developing a
healthcare system that is viable and sustainable for the City of Cardiff. The study will
adopt a multimodal principle to its investigation. Multi-modality principles involve the
fusing and analysis of several data types and data sets such that the different data
sets and types can interact and inform each other, and give a better understanding of
patients' conditions. The study attempts to be holistic as much as possible, attempting
smart healthcare study within a broader concept of a smart city, relying on the
principles of the Internet of Things; that real‐world objects communicate and interact
with themselves via networking technologies. (Lahat, Adalı and Jutten, 2015; Salvi et
al., 2024).
Data Collection:
In its attempts to be a one-stop reference point, the uses three datasets, where
dependent variables are discreet continuous, and image-processed. All three datasets
were obtained from Kaggle. Kaggle is an open data repository that meets the standard
of data quality assurance.
My three datasets are certified by the owners for use in research. For data security
purposes, the customer's information was anonymized.
The first dataset predicts the risk of Atherosclerotic Cardiovascular Disease (ASCVD),
which is known as heart disease. The dependent variable, ASCVD, is continuous. A
continuous variable according to Urdan (2010) is a of numbers with an infinite set of
values. The second data predicts the risk of cerebrovascular accident, CVA, also
known as stroke, the dependent variable on this data set, stroke, is a discrete variable.
The third data relates to medical diagnostics of prostate cancer. The input variables
were obtained using processes known as image processing, and machine vision.
The essence of a three dataset is to succinctly answer the research questions of this
study beyond reasonable doubt and in compliance with the multi-modal principles.
Data Analysis:
1. This study will utilize some contemporary tools, techniques, models, and
development environments such as machine learning algorithms and deep
learning models, python programming language, python libraries, and Jupyter
Notebook to achieve its objectives
2. The project will utilize Tableau and Python libraries such as pandas for data
manipulations and analysis, matplotlib, seaborn, and plot. express for data
visualization.
3. I shall perform data cleaning, feature selection, imbalanced data handling
techniques, and exploratory data analysis on my data sets.
4. I will then fit my data for machine learning algorithms, this will involve label
encoding, variable splitting, training, and testing
5. I shall utilize a wide range of machine learning algorithms such as and not
limited to:
a. Decision Trees
b. Random Forest
c. Logistic Regression
d. Support Vector Machine
e. K-nearest neighbour
f. Naive Bayes
g. Multiple Regression
h. Extreme Gradient Boosting (XGBoost)
i. Convolutional neural networks (CNNs)
CONCLUSION
It is undeniable that each of us needs good health to continue being highly productive
and to continue having communal relationships with our loved ones. The status of our
health care systems, not just in Cardiff, United Kingdom, but globally, is undoubtedly
fraught with enormous challenges. Digital technology—deep learning, machine
learning, cloud computing, artificial intelligence, and the Internet of Things—is the only
way to overcome this obstacle.
Using machine learning techniques, this research aims to preserve physicians, our
society, and the overall healthcare system. It aims to be as comprehensive as feasible.
It analyses various datasets and data types using multimodal principles for
comparison, interpretation, and analysis. It obtains its data from reputable source
Kaggle, cleans it up, and then feeds it to a variety of deep learning models and
machine learning algorithms. It plans to implement its conclusions in Cardiff, a city that
people adore. As a result, offering guidance to healthcare stakeholders on tried-and-
true methods to raise Cardiff's standard of living and healthcare.
RESEARCH GANTT CHART
I shall start my dissertaon on January 8, 2024, and Finish April 02, 2024
REFERENCES
1. Al-Azzam, M. K. A., & Alazzam, M. B. (2019). Smart City and Smart-Health
Framework, Challenges and Opportunities. International Journal of Advanced
Computer Science and Applications, 10(2).
https://doi.org/10.14569/ijacsa.2019.0100223
2. Alhussein, M., (2017). "Monitoring Parkinson's Disease in Smart Cities," IEEE
Access, vol. 5, 2017, doi: 10.1!09/ACCESS.2017.2748561.
3. Ashfaq, Z., Mumtaz, R., Rafay, A., Zaidi, S. S. H., Saleem, H., Mumtaz, S.,
Shahid, A., De Poorter, E., & Moerman, I. (2022). Embedded AI-Based Digi-
Healthcare. Applied Sciences, 12(1), 519.
https://doi.org/10.3390/app12010519
4.
Asim, Z., (2022). "Shaping Healthcare System Under Industry 5.0: Trends
and Barriers," Sudan Journal of Medical Sciences, Sep. 2022, doi:
10.18502/sjms.vl 7i3.12115.
5. Australian Institute of Health and Welfare. (2014). Australia’s Health. [Online].
Available at: http://www.aihw.gov.au/WorkArea/ DownloadAsset.aspx-
?id=60129548150
6. Bhandari, M., Zeffiro, T. and Reddiboina, M. (2020) 'Artificial intelligence and
robotic surgery,' Current Opinion in Urology, 30(1), pp. 48–54.
https://doi.org/10.1097/mou.0000000000000692.
7. Chang, S., Chiang, R., Wu, S., & Chang, W. (2016). A Context-Aware,
interactive M-Health system for diabetics. IT Professional, 18(3), 14–
22. https://doi.org/10.1109/mitp.2016.48
8. COM (2018) - Artificial Intelligence for Europe - EU monitor. 237.
https://www.eumonitor.eu/9353000/1/j9vvik7m1c3gyxp/vknuqttbx4zb.
9. Costanzo, M. R. and S. Pamboukian (2009), “Device-Based Patient
Management,” Cardiac Resynchronization Therapy, pp. 404–418.
10. Cugurullo, F., (2020). Urban artificial intelligence: From automation to
autonomy in the smart city. Front Sustain Cities
11. Demirkan, H, (2023). “A Smart Healthcare Systems Framework,” IT
Professional, vol. 15, no. 5, pp. 38–45.
12. Fan, Y. J., Yin, Y., Da Xu, L., Zeng, Y., & Wu, F. (2014). IoT-Based Smart
Rehabilitation System. IEEE Transactions on Industrial Informatics, 10(2),
1568–1577. https://doi.org/10.1109/tii.2014.2302583
13. Ghazal, T.M.; Hasan, M.K.; Hassan, R.; Islam, S.; Abdullah, S.; Afifi, M.A.;
Kalra, D. (2020). Security vulnerabilities, attacks, threats and the proposed
countermeasures for the Internet of Things applications. Solid State Technol.
63, 2513–2521
14. Ghazal, T. M., Hasan, M. K., Alshurideh, M., Alzoubi, H. M., Ahmad, M.,
Akbar, S. S., Kurdi, B. A., & Akour, I. (2021). IoT for Smart Cities: Machine
Learning Approaches in Smart Healthcare—A Review. Future Internet, 13(8),
218. https://doi.org/10.3390/fi13080218
15. Gope and Hwang T., (2016): ‘‘BSN-care: A secure IoT-based modern
healthcare system using body sensor network,’’ IEEE Sensors J., vol. 16, no.
5, pp. 1368–1376, Mar. 2016.
16. Grunhut, J. A., Wyatt, T, and Marques, 0., (2021, ). "Educating Future
Physicians in Artificial [ntelligence (AI): An [ntegrative Review and Proposed
Changes," J Med Educ Currie Dev, vol. 8, p. 238212052110368, doi:
I0.1177/23821205211036836.
17. Hassan, A., Sulaiman, R. B., Abdulgabber, M. A., & Kahtan, H. (2023).
Balancing Technological Advances with User Needs: User-centered Principles
for AI-Driven Smart City Healthcare Monitoring. International Journal of
Advanced Computer Science and Applications, 14(3).
https://doi.org/10.14569/ijacsa.2023.0140341
18. Hassan, M. K., Desouky, A. I. E., Elghamrawy, S. M., & Sarhan, A. M. (2019).
A Hybrid Real-time remote monitoring framework with NB-WOA algorithm for
patients with chronic diseases. Future Generation Computer Systems, 93,
77–95. https://doi.org/10.1016/j.future.2018.10.021
19. Herath, H.M.R.P. and Mittal, M. (2022) 'Adoption of artificial intelligence in smart
cities: A comprehensive review,' International Journal of Information
Management Data Insights, 2(1), p. 100076.
https://doi.org/10.1016/j.jjimei.2022.100076.
20. Huang, G.; Fang, Y.; Wang, X.; Pei, Y.; Horn, B (2018). A Survey on the status
of smart healthcare from the universal village perspective. In Proceedings of
the 4th IEEE International Conference on Universal Village, Boston, MA, USA,
21. Institute of Medicine, (2001). Crossing the Quality Chasm: A New Health
System for the 21st Century. Washington, DC: The National Academies Press.
https://doi.org/10.17226/10027.
22. International Society of Automation (2015) 'Unlocking the potentials of the
internet of things,' Pinterest.
https://www.pinterest.com/pin/603763893772357033/.
23. Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., Wang, Y., Dong, Q.,
Shen, H., & Wang, Y. (2017). Artificial intelligence in healthcare: past, present
and future. Stroke and Vascular Neurology, 2(4), 230–243.
https://doi.org/10.1136/svn-2017-000101
24. Kharel, J.; Reda, H.T.; Shin, S.Y. An architecture for smart health monitoring
system based on fog computing. J. Commun. 2017, 12, 228–233.
25. Kumar V and Sheshadri K., (2019). Applications of artificial intelligence in
academic libraries. ResearchGate. Available at:
https://www.researchgate.net/publication/332876721_Applications_of_artificial
_intelligence_in_academic_libraries.
26. Lahat, D., Adalı, T. and Jutten, C. (2015) 'Multimodal Data Fusion: An
overview of methods, challenges, and prospects,' Proceedings of the IEEE,
103(9), pp. 1449–1477. https://doi.org/10.1109/jproc.2015.2460697.
27. Li, W.; Chai, Y.; Khan, F.; Jan, S.R.U.; Verma, S.; Menon, V.G.; Kavita; Li, X.
A Comprehensive survey on machine learning-based big data analytics for iot-
enabled smart healthcare system. Mob. Netw. Appl. 2021, 26, 234–252
28. Muhammad, K.; Khan, S.; Del Ser, J.; de Albuquerque, V.H.C., (2021) Deep
learning for multigrade brain tumor classification in smart healthcare systems:
A prospective survey. IEEE Trans. Neural Netw. Learn. Syst. 2021, 32, 507–
522.
29. Pasluosta, C., Gaßner, H., Winkler, J., Klucken, J., & Eskofier, B. M. (2015).
An emerging era in the management of Parkinson’s disease: wearable
technologies and the internet of things. IEEE Journal of Biomedical and
Health Informatics, 19(6), 1873–1881.
https://doi.org/10.1109/jbhi.2015.2461555
30. Patel, K. K. and Patel, S.M. (2016). 'Internet of Things-IOT: definition,
characteristics, architecture, enabling technologies, application &...,'
ResearchGate [Preprint].
https://www.researchgate.net/publication/330425585_Internet_of_Things-
IOT_Definition_Characteristics_Architecture_Enabling_Technologies_Applicat
ion_Future_Challenges.
31. Pitchaimani, J. and Rajan, S.E. (2019) 'Smart-Monitor: patient monitoring
system for IoT-Based healthcare system using deep learning,' IETE Journal of
Research, 68(2), pp. 1435–
1442. https://doi.org/10.1080/03772063.2019.1649215.
32. Raju, S.S. and Kumar, R. C., (2022). "EAI Endorsed Transactions on Smart
Cities Artificial Intelligence in Smart cities and Healthcare," doi: I
0.4108/eetsc.v6i3.2275.
33. Salvi, M. et al. (2024) 'Multi-Modality Approaches for Medical Support Systems:
A Systematic Review of the last decade,' Information Fusion, 103, p. 102134.
https://doi.org/10.1016/j.inffus.2023.102134.
34. Saunders, M.N.K., Lewis, P. and Thornhill, A. (2015). Research Methods for
Business Students. New York: Pearson Education.
35. Sicari, S., Rizzardi, A., Grieco, L., Piro, G., & Coen-Porisini, A. (2017,
September). A policy enforcement framework for Internet of Things applications
in the smart health. Smart Health, 3–4, 39–74.
https://doi.org/10.1016/j.smhl.2017.06.001.
36. Shah, R.; Chircu, A. IoT and AI in healthcare: A systematic literature review.
Issues Inf. Syst. 2018, 19.
37. Snell, J. D., (1996). “Cardiac device with patient-triggered storage of
physiological sensor data,”. https://patents.justia.com/patent/5518001
38. Tian, S., Yang W., Grange J, P. Wang, W. Huang, and Z. Ye, (2019). "Smart
healthcare: making medical care more intelligent," J Glob Health, vol. 3, no.
3, pp. 62-65, 2019, doi: l 0.1016/J.GLOHJ.2019.07.00 I.
39. Nian, Z., Diethe, T., Camplani, M., Tao, L., Burrows, A., Twomey, N., Kaleshi,
D., Mirmehdi, M., Flach, P. A., & Craddock, I. J. (2015). Bridging e-Health and
the internet of things: the SPHERE Project. IEEE Intelligent Systems, 30(4),
39–46. https://doi.org/10.1109/mis.2015.57
