Available online

Available online at
at www.sciencedirect.com
www.sciencedirect.com
ScienceDirect
ScienceDirect
ScienceDirect
Procedia Computer Science00 (2019) 000–000
Procediaonline
Available
Procedia Computer Science00
at Science00
Computer (2019) 000–000
000–000
www.sciencedirect.com
(2018) www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia
www.elsevier.com/locate/procedia
ScienceDirect
Procedia Computer Science 154 (2019) 361–368

8th International Congress of Information and Communication Technology (ICICT-2019)

8th
8th International
International Congress
Congress of
of Information
Information and
and Communication
Communication Technology
Technology,(ICICT-2019)
ICICT 2019
Machine Learning Approaches in Smart Health
Machine
Research on theLearning
Innovation Approaches
of Protecting in Smart Health
Intangible Cultural
a
Zeina Rayan a *, Marco Alfonse a, Abdel-Badeeh M. Salem aa
a

a
Heritage
Zeina Rayan in the
*, Marco "Internet
Alfonse Plus" Era
, Abdel-Badeeh M. Salem
Department of Computer Science, Faculty of computer and information science, University of Ain Shams, Cairo, Egypt
a a*and information science,
b University of Ain Shams, Cairo, Egypt
Ying Li , Peng Duan
Department of Computer Science, Faculty of computer

Abstract
Suzhou University Shandong Business Institute
Abstract
The increase of age average led to an increase in the demand of providing and improving the service of healthcare. The
advancing
The increase
Abstract of the information
of age average and
led communication
to an increase in technology
the demand (ICT)of led to the development
providing and improving of smart cities which
the service have a lotThe
of healthcare. of
components.
advancing of One of those components
the information is Smart Health
and communication (s-Health),
technology (ICT) which is used
led to the in improving
development healthcare
of smart by providing
cities which have a many
lot of
services
components.
This such
article as patient
One
combines themonitoring,
of those most fierceearly
components isdiagnosis
concept Smart of diseases
Health
"Internet inand
(s-Health),
Plus" sowhich
modern on.
eraNowadays
,isFrom
usedtheinthere are many
improving
perspective ofmachine
healthcare learning
"Internetby ittechniques
providing
Plus", many
discusses
that can
services facilitates
such as s-Health
patient services.
monitoring, This
early paper
diagnosis reviews
of recent
diseases published
and so on. papers
Nowadays in the
there
the protection mode, tries to explore the key points for the new model to construct “Internet + intangible cultural area
are of
many smart health
machine starting
learning from the
techniques
heritage
years 2011 to 2017, and a structured analysis for different machine learning (ML) approaches
protection”, provides reasonable practical guidance, and finally creates innovative ideas and methods for the protection the
that can facilitates s-Health services. This paper reviews recent published papers in the area that
of are
smart applied
health in s-Health.
starting from The
of
results2011
years
intangibleshow tothat
2017,
cultural theandMLa approach
heritage.structured isanalysis
Simultaneouslyused initfor
many s-Health
different
makes applications
machine
academic learning such
contributions (ML)toasapproaches
theGlaucoma diagnosis,
that
innovation Alzheimer’s
are applied
and disease,
in s-Health.
inheritance of ChineseThe
bacterial
results show
intangible sepsisthatdiagnoses,
cultural the the Intensive
ML approach
heritage. is usedCareinUnit
many(ICU) readmissions,
s-Health andsuch
applications cataract detection.diagnosis,
as Glaucoma The Artificial Neural Network
Alzheimer’s disease,
(ANN), Support
bacterial Vector Machine
sepsis diagnoses, (SVM)Care
the Intensive algorithm and deep
Unit (ICU) learning models
readmissions, especially
and cataract the Convolutional
detection. The Artificial Neural Network
(CNN)
© 2019 are
(ANN), the
Support
The most commonly
Vector
Authors. Machine
Published used machine
(SVM)
by Elsevier learningand
algorithm
B.V. approaches where they
deep learning modelsproved to get high
especially evaluation performance
the Convolutional in most
Neural Network
cases.is an
(CNN)
This areopen
the most
access commonly usedthe
article under machine learning approaches
CC BY-NC-ND where they proved to get high evaluation performance in most
license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
cases.
Peer-review under responsibility of organizing committee of the 8th International Congress of Information and Communication
© 2019 The (ICICT-2019).
Technology Authors. Published by Elsevier B.V.
©© 2019
2019
This is anThe
The Authors.
open Published
accessPublished
Authors. by Elsevier
article under
by Elsevier Ltd.
the CC BY-NC-ND
B.V. license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
This is an
Peer-review open
underaccess article under
responsibility the
of the CC
organizingBY-NC-ND
committee license
of the(https://creativecommons.org/licenses/by-nc-nd/4.0/)
8th International Congress of Information and Communication
This is an "Internet+",
Keywords: open access article under
intangible cultural CC BY-NC-ND
heritage, innovative license
model; (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection and peer-review under responsibility of the 8th International Congress of Information and Communication Technology,
Technology
Peer-review (ICICT-2019).
under responsibility of organizing committee of the 8th International Congress of Information and Communication
ICICT 2019.
Technology (ICICT-2019).
Keywords: Smart Health; Electronic Health; Medical Informatics; Machine Learning
Keywords: Smart Health; Electronic Health; Medical Informatics; Machine Learning
* a
About the author: Li Ying (1985), female, Han nationality, Yantai, Shandong, Ph.D. student of Suzhou University, lecturer of information art
1.
department Introduction
of Shandong Business Institute, research direction: design science;
1.bDuan Peng Introduction
(1983-), male, Han nationality, Yantai, Shandong, a lecturer at the Innovation and Entrepreneurship Center of Shandong Business
Institute,
Smart research
Health direction: computer
is a field whichapplication
growntechnology;
as a subset of two fields, smart cities and Electronic Health (e-Health) ‎1.
Fund Project: Shandong University Humanities and Social Sciences Research Project “Investigation on the Visualization of Qilu Classic Folk
Smart
Smart Health
cities can beis adefined
field which grown
as “cities as a subset
strongly founded of on
twoinformation
fields, smartandcities and Electronic
communication Health (e-Health)
technologies ‎1.
that invest
Art” (J16WH12); Shandong Provincial Social Science Planning Research Project “Qilu Folk Art Narrative Research Based on Information
Smart
in humancities can
and be
socialdefined as
capital “cities
to strongly
improve the founded
quality on
of information
life of and
their communication
citizens by technologies
fostering economic
Interaction Design” (18DWYJ01);Jiangsu Province Academic Degree College Graduate innovation projects "Design Art Research of Su Zuo that invest
growth,
in human
Latticed windowsand" (KYZZ16_0073)
social capital to improve the quality of life of their citizens by fostering economic growth,
Detailed address: Shandong Business Institute, Jinhai Road, High-tech Zone, Yantai, Shandong, China,zip code: 264670, contact number:
* Corresponding
15954549212, author
E-mail: : Zeina Rayan
[email protected]
*E-mail address:[email protected]
Corresponding author : Zeina Rayan
E-mail address:[email protected]
2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
https://creativecommons.org/licenses/by-nc-nd/4.0/)
2019 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection and peer-review under responsibility of the scientific committee of the 8th International Congress of Information and Communication
https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection and peer-review under responsibility of the scientific committee of the 8th International Congress
Technology
Selection and peer-review under responsibility of the scientific committee of the 8th International Congress of Information and Communication
of Information and Communication Technology
Technology

1877-0509 © 2019 The Authors. Published by Elsevier Ltd.

This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
Selection and peer-review under responsibility of the 8th International Congress of Information and Communication Technology,
ICICT 2019.
10.1016/j.procs.2019.06.052
362 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000

participatory governance, wise management of resources, sustainability, and efficient mobility, whilst they
guarantee the privacy and security of the citizens.”‎2. e-Health can be defined as “an emerging field in the
intersection of medical informatics, public health and business, referring to health services and information
delivered or enhanced through the Internet and related technologies. In a broader sense, the term characterizes not
only a technical development, but also a state-of-mind, a way of thinking, an attitude, and a commitment for
networked, global thinking, to improve health care locally, regionally, and worldwide by using information and
communication technology.”‎3. There is an intersection between s-Health and Mobile Health (m-Health); m-Health
can be defined as “emerging mobile communications and network technologies for healthcare systems”‎4.
Machine Learning (ML) is a field that grew out of Artificial Intelligence (AI). It is concerned with designing and
developing algorithms that enable the computers to evolve their behaviors according to empirical data. The ML
approach is evolving rapidly as a result of the improvement of the ML algorithms, enhanced methods of capturing
data, improved computer networks, new sensors/IO units, and the interest in self-customization to users’ behavior ‎5.
ML plays an important role in s-Health, it can improve the quality of healthcare services by providing accurate
medical diagnosis, predicting diseases in early stages and disease analyses.
This paper is structured as follows: section 2 presents the components of a smart health system, and the pipeline
of developing it, section 3 contains a comparative study between different machine learning approaches that are
applied in s-Health, section 4 presents the conclusions and future work.

2. Smart Health

s-Health is a new form of healthcare which is a subfield of e-Health using Electronic Health Records (EHR) and
other variables coming from the smart city’s infrastructure; in order to improve the healthcare. The idea of a smart
health system is that it uses all data coming from sensors on the patient body, smart homes, smart city infrastructure,
and robots to help make better decisions and improve healthcare by providing emergency response and paging
doctors, nurses, and technicians (see Fig. 1). It can also support self-diagnosis, monitoring, early detection and
treatments.

Fig. 1 The components of a smart health system

The smart health system can be described as a pipeline that includes data acquisition, networking and computing
technologies, data security and privacy, data processing, and data dissemination. Fig. 2 shows the pipeline for
developing a smart heath system.
 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368 363
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000 3

Fig. 2 Smart health pipeline

The data acquisition means collecting data from different sources such as sensor network ‎6,7,8, Mobile Ad hoc
Networks ‎9,10 , Vehicular Ad hoc Networks ‎11,12, Social networks ‎13, Internet of Things (IoT) ‎14, 5G
devices ‎15,16, Device-to-Device (D2D) ‎17, Unmanned Aerial Vehicles (UAVs) ‎18, or may be a combination of a
set of them. The networking and computing technologies are applied to the data gathered in the data acquisition
phase, since this data is a bit complex, we need a server or a computing technology where the data can be processed
such as cloud computing ‎19, or fog computing ‎20. The data security and privacy plays a big role in both s-Health
and smart cities where gathering so many information about citizens could possibly violate the citizen privacy ‎21.
The data processing involves preprocessing to remove the noise and clean the data, feature extraction to find the
relevant features, and ML technique to process the data and perform the required task such as
classification/clustering …etc. The data dissemination is responsible for providing the output of the data processing
phase to the target parties by means of direct access, push notifications, pub/sub, or opportunistic routing ‎22.

3. ML Approaches in Smart Health

This section presents a comparative study between different machine learning approaches that are applied in s-
Health applications and systems.
Chai, et. al 23 proposed a deep-learning model using CNN for Glaucoma diagnosis, using data which was
collected from Beijing Tongren Hospital in China. The dataset includes 3554 images of retinal fundus collected
from 2000 patients who suffered from numerus types of eye diseases. This data set contains 1391 images that are
diagnosed with glaucoma, and the remaining images (2163) are not. The proposed model gives accuracy 81.69%.
Zhang, et. al 24 proposed a model to predict the Mini-Mental State Examination (MMSE) scores in Alzheimer’s
disease (AD) based on multi-granularity whole-brain segmentation volumes through images. They used the
Alzheimer's disease Neuroimaging Initiative (ADNI) database, which includes healthy control cases (190), mild
cognitive impairment cases (331) and AD cases (157). The proposed model gives Area Under Curve (AUC) 0.554.
Liu, Y. and Choi, K.S. 25 proposed a model to diagnose the bacterial sepsis in critically ill patients. The data was
collected from Medical ICU (MICU) and ICU departments of the General Hospital of Guangzhou, China (185
inpatients). The proposed model gives accuracy 90.8%.
Viegas, R., et. al 26
‎ proposed an ensemble model to predict readmissions of ICU utilizing feature selection and
fuzzy modeling approaches. They used the Multi–Parameter Intelligent Monitoring for Intensive Care (MIMIC II)
database, which is an archive of data, publically available, collected from more than 32,000 ICU patients within
2001 and 2008 27‎ . The proposed model gives AUC 0.79.
Dong, Y., et. al ‎28 proposed a model to use the information of retinal vascular for the detection of cataract, which
is based on the retinal images classification. The dataset used contains 445 fundus images (199 normal images, 148
slight images, 71 images are medium and 27 images are severe). The classification of these fundus images are
364 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000

performed by the professional ophthalmologist, the data is divided into 2 sets (a set contains 70% of the data that are
randomly selected are used for training the classification algorithm, and the remaining 30% of the data is used to test
the model performance). The proposed model gives accuracy 94.91%.
Zheng, B., et. al 29
‎ proposed a model to study the hospital readmissions risk prediction. The dataset was collected
from medical records of different hospitals, which contains 1641 records (316 records of them are readmitted to
hospitals during 30 days of discharge). The proposed model gives accuracy 83.8%.
Fialho, A.S., et. al ‎30 proposed a model to predict the readmission of ICU between 24 and 72 hours after the
discharge from ICU, they used the MIMIC II database. The proposed model gives accuracy 74%.
Table 1 shows a comparison between different machine learning approaches used in smart health.

Table 1 ML approaches in s-Health

Authors Objective Data Set Preprocessing Feature Machine Evaluation

Extraction Learning
Technique
Chai, et. Glaucoma Data Equalize the size Scale-Invariant Logistic Accuracy =
al, 2017 ‎23 Diagnosis collected of images Feature regression 60.84%
from (Normalization) Transform
Beijing (SIFT) features
Tongren Local Binary Logistic Accuracy =
Hospital. Patterns (LBP) regression 62.67%
features
Normalization whole image One-branch Accuracy =
and CNN (5-conv 71.36%
Segmentation layers)
extracted image One-branch Accuracy =
CNN (5-conv 74.95%
layers)
Two-branch Accuracy =
CNN (six- 74.89%
conv layers)
Two-branch Accuracy =
CNN (5-conv 81.69%
layers)
Zhang, et. Prediction of ADNI segmented each Level 1 Linear AUC =
al, 2017 ‎24 the MMSE database. image into 5- Regression 0.318
scores in level whole- Ridge AUC =
Alzheimer’s brain Regression 0.530
disease. segmentations Lasso AUC =
utilizing the Regression 0.520
Multi-Atlas Elastic-net AUC =
Likelihood 0.527
Fusion algorithm Level 2 Linear AUC =
Regression 0.453
Ridge AUC =
Regression 0.542
Lasso AUC =
Regression 0.550
Elastic-net AUC =
0.554
 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368 365
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000 5

Level 3 Linear AUC =

Regression 0.388
Ridge AUC =
Regression 0.427
Lasso AUC =
Regression 0.419
Elastic-net AUC =
0.424
Level 4 Linear AUC =
Regression 0.309
Ridge AUC =
Regression 0.318
Lasso AUC =
Regression 0.324
Elastic-net AUC =
0.322
Level 5 Linear AUC =
Regression 0.266
Ridge AUC =
Regression 0.273
Lasso AUC =
Regression 0.262
Elastic-net AUC =
0.268
Liu, Y. Diagnose Data Normalization Procalcitonin Accurac
and Choi, bacterial collected for the age (PCT) (Cutoff y = 78.4%
K.S., 2017 sepsis. from the attribute = 3.40)
25
General
Hospital of PCT (Cutoff Accuracy =
Guangzhou, = 8.56) 82.2%
China. SVM Accuracy =
88.6%
Random Accuracy =
Forest (RF) 89.2%
ANN Accuracy =
90.8%
Viegas, R., Prediction of MIMIC II exactly define Area under the Average AUC =
et. al, ICU the scope of sensitivity and ensemble 0.77±0.02
2017 ‎26 readmissions. allowable specificity decision
qualities as curves criteria
indicated by the Maximum AUC =
expert distance 0.77±0.02
ensemble
decision
criteria
Weighted AUC =
distance 0.77±0.02
ensemble
decision
criteria
366 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000

Sensitivity and Average AUC =

specificity at ensemble 0.76±0.02
the intersection decision
threshold criteria
Maximum AUC =
distance 0.75±0.02
ensemble
decision
criteria
Weighted AUC =
distance 0.76±0.02
ensemble
decision
criteria
Sensitivity and Average AUC =
specificity close ensemble 0.75±0.02
to the decision
intersection criteria
threshold Maximum AUC =
distance 0.74±0.02
ensemble
decision
criteria
Weighted AUC =
distance 0.75±0.02
ensemble
decision
criteria
Dong, Y., cataract Data Wavelet Accuracy =
et. al, detection collected features SVM 81.81%
2016 ‎28 from Texture features Accuracy =
different 80.33%
sources staking Accuracy =
algorithm 84%
Zheng, B., The risk Data was Using a range Radial Basis Accuracy =
et. al, prediction of collected defined for Function 56.1%
2015 ‎29 hospital from admissible Neural
readmissions. different values. Network
hospitals. Particle Accuracy =
Swarm 83.8%
Optimization–
SVM with
Radial Basis
Function
Linear SVM Accuracy =
50.6%
Polynomial Accuracy =
SVM 52.7%
RBF-based Accuracy =
SVM 69.5%
 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368 367
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000 7

Fialho, Predict ICU MIMIC II • Data sequential Accuracy =

A.S., et. al, readmission. missing for forward 71 ± 3%
2012 ‎30 an selection
intentional
reason was Intellegent tool Accuracy =
deleted. titeled: Acute Data mining 61 ± 2%
• Data Physiology And approach
missing for Chronic Health
unintentiona Evaluation
l reason was (APACHE)
given the version II
last
APACHE Accuracy =
available
version III 65 ± 2%
value was
Modified tool
used.

From the above table, we can find that the ML approach is used in many s-Health applications such as Glaucoma
diagnosis, Alzheimer’s disease, bacterial sepsis diagnoses, ICU readmissions, and cataract detection. The ANN,
SVM algorithm and deep learning models especially CNN are the most commonly used machine learning
approaches where they proved to get high evaluation performance in most cases.

4. Conclusions and Future Work

Smart health is a developing and exceedingly critical research field with a possibly noteworthy effect on the
conventional healthcare industry. This work presents an overview of the challenges, pipeline, and techniques of
smart health. A systematic pipeline of data processing is accommodated for conventional smart health, covering data
acquisition, data processing, data dissemination, data security and privacy, and networking and computing
technologies. In spite of numerous chances and methodologies for data analytics in healthcare presented in this work,
there are numerous different bearings to be investigated concerning different aspects of healthcare data such as
quality, privacy and so on. In future work, we are going to apply Machine Learning algorithms to detect anomaly
over intensive care patient data.

References

1. Solanas, A., Patsakis, C., Conti, M., Vlachos, I.S., Ramos, V., Falcone, F., Postolache, O., Pérez-Martínez, P.A., Di Pietro, R., Perrea, D.N.
and Martinez-Balleste, A., Smart health: a context-aware health paradigm within smart cities. IEEE Communications Magazine, 52(8),
pp.74-81, 2014.
2. P. A. Pérez-Martínez, A. Martínez-Ballesté, and A. Solanas, “Privacy in Smart Cities — A Case Study of Smart Public Parking,” Proc. 3rd
Int’l Conf. Pervasive Embedded Computing and Commun. Sys., pp. 55-59, 2013.
3. G. Eysenbach, “What Is e-Health?” J. Medical Internet Research, vol. 3, no. 2, pp. 20-21, Apr-June 2001.
4. Istepanian, Robert, Swamy Laxminarayan, and Constantinos S. Pattichis. M-health. New York, NY: Springer Science+ Business Media,
Incorporated, 2006.
5. John D. Kelleher, Brian Mac Namee, Aoife D'Arcy, Fundamentals of Machine Learning for Predictive Data Analytics: Algorithms,
Worked Examples, and Case Studies, 1st Edition, The MIT Press, 2015.
6. Y. Fujii, N. Yoshiura, A. Takita, and N. Ohta, “Smart street light system with energy saving function based on the sensor network,” in
Proceedings of the Fourth International Conference on Future Energy Systems, ser. e-Energy ’13, New York, NY, USA, pp. 271–272,
2013. [Online]. Available: http://doi.acm.org/10.1145/2487166.2487202
7. N. Yoshiura, Y. Fujii, and N. Ohta, “Smart street light system looking like usual street lights based on sensor networks,” in 13th IEEE
International Symposium on Communications and Information Technologies, pp. 633–637, 2013.
8. Abinaya, R., V. Varsha, and Kaluvan Hariharan. "An intelligent street light system based on piezoelectric sensor networks." In 4th
International Conference on Electronics and Communication Systems (ICECS), pp. 138-142. IEEE, 2017.
A. Didolkar and F. Zama, “SMARTIE: A security solution for a smart city using internet of things - architecture reference model,”
International Journal of Advance Research and Innovative Ideas in Education, vol. 2, no. 4, pp. 231-236, 2016.
368 Zeina Rayan et al. / Procedia Computer Science 154 (2019) 361–368
Zeina Rayan/ Procedia Computer Science00 (2019) 000–000

9. W. A. Jabbar, M. Ismail, and R. Nordin, “On the performance of the current MANET routing protocols for VoIP, HTTP, and FTP
applications,” Journal of Computer Networks and Communications, Volume 2014, Article ID 154983, 2014.
10. S. Al-Sultan, M. M. Al-Doori, A. H. Al-Bayatti, and H. Zedan, “A comprehensive survey on vehicular ad hoc network,” Journal of
Network and Computer Applications, vol. 37, pp. 380-392, 2014.
11. K. Mershad and H. Artail, “A framework for secure and efficient data acquisition in vehicular ad hoc networks,” IEEE Transactions on
Vehicular Technology, vol. 62, no. 2, pp. 536-551, Feb 2013.
12. E. Psomakelis, F. Aisopos, A. Litke, K. Tserpes, M. Kardara, and P. M. Campo, “Big IoT and social networking data for smart cities:
Algorithmic improvements on big data analysis in the context of radical city applications,” arXiv preprint arXiv: 1607.00509, 2016.
13. Kumar, S.P., Samson, V.R.R., Sai, U.B., Rao, P.M. and Eswar, K.K. Smart health monitoring system of patient through IoT. In
International Conference on IoT in Social, Mobile, Analytics and Cloud (I-SMAC), pp. 551-556, IEEE, 2017.
A. Orsino, G. Araniti, L. Militano, J. Alonso-Zarate, A. Molinaro, and A. Iera, “Energy efficient IoT data collection in smart cities
exploiting D2D communications,” Sensors, vol. 16, no. 6, p. 836-851, 2016.
14. Beard Cory, "High Reliability 4G and 5G Cellular Wireless Services for Smart Cities", IEEE Smart Cities Kansas City Kickoff Workshop,
February 2016.
15. M. N. Tehrani, M. Uysal, and H. Yanikomeroglu, “Device-to-Device communication in 5g cellular networks: challenges, solutions, and
future directions,” IEEE Communications Magazine, vol. 52, no. 5, pp. 86–92, 2014.
16. G. Ermacora, S. Rosa, and A. Toma, “Fly4smartcity: A cloud robotics service for smart city applications,” Journal of Ambient Intelligence
and Smart Environments, vol. 8, no. 3, pp. 347–358, 2016.
17. Mell, Peter, and Tim Grance. "The NIST definition of cloud computing", Computer Security Division, Information Technology Laboratory,
the National Institute of Standards and Technology, Gaithersburg, 2011.
18. Bonomi, F., Milito, R., Zhu, J. and Addepalli, S., August. Fog computing and its role in the internet of things. In Proceedings of the first
edition of the MCC workshop on Mobile cloud computing, ACM, pp. 13-16, 2012.
A. Martínez-Ballesté, P. A. Pérez-Martínez, and A. Solanas, “The Pursuit of Citizens’ Privacy: A Privacy-Aware Smart City is Possible,”
IEEE Communication Magazine, vol. 51, no. 6, pp. 136-141, 2013.
19. Gharaibeh, A., Salahuddin, M.A., Hussini, S.J., Khreishah, A., Khalil, I., Guizani, M. and Al-Fuqaha, A. Smart Cities: A Survey on Data
Management, Security, and Enabling Technologies. IEEE Communications Surveys and Tutorials, 19(4), pp.2456-2501, 2017.
20. Chai, Y., He, L., Mei, Q., Liu, H. and Xu, L., Deep Learning Through Two-Branch Convolutional Neuron Network for Glaucoma
Diagnosis. Proceedings of International Conference on Smart Health, pp. 191-201, Springer, 2017.
21. Zhang, J., Luo, Y., Jiang, Z. and Tang, X., Regression Analysis and Prediction of Mini-Mental State Examination Score in Alzheimer’s
Disease Using Multi-granularity Whole-Brain Segmentations. Proceedings of International Conference on Smart Health, pp. 202-213.
Springer, 2017.
22. Liu, Y. and Choi, K.S., Using Machine Learning to Diagnose Bacterial Sepsis in the Critically Ill Patients. Proceedings of International
Conference on Smart Health, pp. 223-233, Springer, 2017.
23. Viegas, R., Salgado, C.M., Curto, S., Carvalho, J.P., Vieira, S.M. and Finkelstein, S.N., Daily prediction of ICU readmissions using feature
engineering and ensemble fuzzy modeling. Expert Systems with Applications, 79, pp.244-253, 2017.
24. M. Saeed, M. Villarroel, A.T. Reisner, G. Clifford, L. Lehman, G.B. Moody, T. Heldt, T.H. Kyaw, B.E. Moody, R.G. Mark.
Multiparameter Intelligent Monitoring in Intensive Care II (MIMIC-II): A public-access ICU database. Critical Care Medicine 39(5):952-
960; doi: 10.1097/CCM.0b013e31820a92c6, 2011.
25. Dong, Y., Wang, Q., Zhang, Q. and Yang, J., Classification of Cataract Fundus Image Based on Retinal Vascular Information. Proceedings
of International Conference on Smart Health, pp. 166-173, Springer, 2016.
26. Zheng, B., Zhang, J., Yoon, S.W., Lam, S.S., Khasawneh, M. and Poranki, S., Predictive modeling of hospital readmissions using
metaheuristics and data mining. Expert Systems with Applications, 42(20), pp.7110-7120, 2015.
27. Fialho, A.S., Cismondi, F., Vieira, S.M., Reti, S.R., Sousa, J.M. and Finkelstein, S.N., Data mining using clinical physiology at discharge
to predict ICU readmissions. Expert Systems with Applications, 39(18), pp.13158-13165, 2012.