Decision Making In Healthcare Systems 1st

Edition Tofigh Allahviranloo download

https://ebookbell.com/product/decision-making-in-healthcare-
systems-1st-edition-tofigh-allahviranloo-54730844

Explore and download more ebooks at ebookbell.com

Here are some recommended products that we believe you will be
interested in. You can click the link to download.

Evidence Synthesis For Decision Making In Healthcare Nicky J Welton

https://ebookbell.com/product/evidence-synthesis-for-decision-making-
in-healthcare-nicky-j-welton-4304106

Healthcare Management Engineering What Does This Fancy Term Really

Mean The Use Of Operations Management Methodology For Quantitative
Decisionmaking In Healthcare Settings 1st Edition Alexander Kolker
Auth
https://ebookbell.com/product/healthcare-management-engineering-what-
does-this-fancy-term-really-mean-the-use-of-operations-management-
methodology-for-quantitative-decisionmaking-in-healthcare-
settings-1st-edition-alexander-kolker-auth-4230086

Applications Of Multicriteria Decisionmaking Theories In Healthcare

And Biomedical Engineering 1st Edition Ilker Ozsahin

https://ebookbell.com/product/applications-of-multicriteria-
decisionmaking-theories-in-healthcare-and-biomedical-engineering-1st-
edition-ilker-ozsahin-23888600

Clinical Judgement And Decisionmaking In Nursing And Interprofessional

Healthcare Standing

https://ebookbell.com/product/clinical-judgement-and-decisionmaking-
in-nursing-and-interprofessional-healthcare-standing-5673602
The Patient Multiple An Ethnography Of Healthcare And Decisionmaking
In Bhutan Jonathan Taee

https://ebookbell.com/product/the-patient-multiple-an-ethnography-of-
healthcare-and-decisionmaking-in-bhutan-jonathan-taee-51750862

Multicriteria Decision Making Theory And Applications In Sustainable

Healthcare Mohamed Abdelbasset

https://ebookbell.com/product/multicriteria-decision-making-theory-
and-applications-in-sustainable-healthcare-mohamed-
abdelbasset-48526758

Uncertainty Decisionmaking And Team Work In Hightech Healthcare

Personsoulmindbodyhood Relational Medicine Federica Raia

https://ebookbell.com/product/uncertainty-decisionmaking-and-team-
work-in-hightech-healthcare-personsoulmindbodyhood-relational-
medicine-federica-raia-47686092

Management Options In Breast Cancer Case Histories Best Practice And

Clinical Decisionmaking 1st Edition Benson John

https://ebookbell.com/product/management-options-in-breast-cancer-
case-histories-best-practice-and-clinical-decisionmaking-1st-edition-
benson-john-1813024

Practical Decision Making In Health Care Ethics Cases And Concepts

Third Revised Raymond J Devettere

https://ebookbell.com/product/practical-decision-making-in-health-
care-ethics-cases-and-concepts-third-revised-raymond-j-
devettere-2386504
Studies in Systems, Decision and Control 513

Tofigh Allahviranloo
Farhad Hosseinzadeh Lotfi
Zohreh Moghaddas
Mohsen Vaez-Ghasemi Editors

Decision
Making in
Healthcare
Systems
Studies in Systems, Decision and Control

Volume 513

Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
The series “Studies in Systems, Decision and Control” (SSDC) covers both new
developments and advances, as well as the state of the art, in the various areas of
broadly perceived systems, decision making and control–quickly, up to date and
with a high quality. The intent is to cover the theory, applications, and perspectives
on the state of the art and future developments relevant to systems, decision making,
control, complex processes and related areas, as embedded in the fields of engi-
neering, computer science, physics, economics, social and life sciences, as well
as the paradigms and methodologies behind them. The series contains mono-
graphs, textbooks, lecture notes and edited volumes in systems, decision making
and control spanning the areas of Cyber-Physical Systems, Autonomous Systems,
Sensor Networks, Control Systems, Energy Systems, Automotive Systems, Biolog-
ical Systems, Vehicular Networking and Connected Vehicles, Aerospace Systems,
Automation, Manufacturing, Smart Grids, Nonlinear Systems, Power Systems,
Robotics, Social Systems, Economic Systems and other. Of particular value to both
the contributors and the readership are the short publication timeframe and the world-
wide distribution and exposure which enable both a wide and rapid dissemination of
research output.
Indexed by SCOPUS, DBLP, WTI Frankfurt eG, zbMATH, SCImago.
All books published in the series are submitted for consideration in Web of Science.
Tofigh Allahviranloo · Farhad Hosseinzadeh Lotfi ·
Zohreh Moghaddas · Mohsen Vaez-Ghasemi
Editors

Decision Making
in Healthcare Systems
Editors
Tofigh Allahviranloo Farhad Hosseinzadeh Lotfi
Faculty of Engineering and Natural Department of Mathematics
Sciences Science and Research Branch
Istinye University Islamic Azad University
İstanbul, Türkiye Tehran, Iran

Zohreh Moghaddas Mohsen Vaez-Ghasemi

School of Information Technology Department of Mathematics
Deakin University, Waurn Ponds Campus Islamic Azad University
Geelong, VIC, Australia Rasht, Iran

ISSN 2198-4182 ISSN 2198-4190 (electronic)

Studies in Systems, Decision and Control
ISBN 978-3-031-46734-9 ISBN 978-3-031-46735-6 (eBook)
https://doi.org/10.1007/978-3-031-46735-6

© The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature
Switzerland AG 2024

This work is subject to copyright. All rights are solely and exclusively licensed by the Publisher, whether
the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse
of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and
transmission or information storage and retrieval, electronic adaptation, computer software, or by similar
or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Paper in this product is recyclable.

Contents

Methodologies for Decision-Making in the Health and Medicine

Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Kemal Gökhan Nalbant and Sevgi Aydin
The Application of System Simulation in the Health Sector:
A Rapid Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Mohammadreza Mobinizadeh, Marita Mohammadshahi,
Parisa Aboee, Zeinab Fakoorfard, Alireza Olyaeemanesh,
and Efat Mohamadi
Data Science in the Field of Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Handan Kulan and Ezgi Özer
Evaluation of Hospitals and Health Care Centers with Ratio Data . . . . . . 29
Mehdi Soltanifar
Multiple Attribute Decision Making in Ranking the Criteria
in Health (with Certain and Uncertain Data) . . . . . . . . . . . . . . . . . . . . . . . . . 49
Mansour Soufi
Healthcare Facility Location . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Hamed Zhiani Rezai and Alireza Davoodi
Fuzzy Transportation Model for Resource Allocation in a Dental
Hospital . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Alize Yaprak Gul and Saliha Karadayi-Usta
Locating Problems for Medical Centers and Emergency Services . . . . . . 173
Mansour Soufi
Budgeting in Healthcare . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
S. Khajavi, M. Etemedy Jooriaby, and E. Kermani

v
vi Contents

Sleep Disorders Detection and Classification Using Random

Forests Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Wadhah Zeyad Tareq Tareq
Green Supply Chain in Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Mehdi Fadaei Eshkiki and Mahdi Homayounfar
Statistical Analysis and Structural Equations on Influential
Parameters in Health . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
Mahdi Homayounfar, Mehdi Fadaei Eshkiki, and Sara Namdar
Boosting Facial Action Unit Detection with CGAN-Based Data
Augmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
Duygu Cakir and Nafiz Arica
Resiliency in Green Supply Chains of Pharmaceuticals . . . . . . . . . . . . . . . . 337
Saliha Karadayi-Usta
Exploring Congestion in Fuzzy DEA by Solving One Model; Case
Study: Hospitals in Tehran . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 355
Saber Saati, Maryam Shadab, and Sajedeh Mohamadniaahmadi
Performance and Managerial Ability Analysis in Health Sector:
A Data Envelopment Analysis Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 373
Alireza Amirteimoori, Sharmineh Safarpour, Sohrab Kordrostami,
and Leila Khoshandam
Mental Health on Twitter in Turkey: Sentiment Analysis
with Transformers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391
Qamar Alshammari and Süreyya Akyüz
Roe v Wade in Twitter: Sentiment Analysis with Machine Learning . . . . 403
Hiba Ayad Allami and Süreyya Akyüz
Time Scheduling for Staff in Hospitals and Health Care Centres . . . . . . . 417
Nursaç Kurt, Ramazan Bakır, and Amir Seyyedabbasi
Transportation Models in Health Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
Nursaç Kurt, Ramazan Bakır, and Amir Seyyedabbasi
Methodologies for Decision-Making
in the Health and Medicine Sector

Kemal Gökhan Nalbant and Sevgi Aydin

Abstract The fields of medicine and healthcare have benefited significantly from
technological advances, investments, and software in our ever-evolving and trans-
forming world, as they have from these same developments in every other industry.
Because of the tremendous strides that have been made in the realm of specialist
software in the health industry, the health system has become significantly more tech-
nologically advanced and functional. As a consequence of advances in technology,
diagnosis and treatment are now more easily available, and favorable outcomes are
being achieved with greater frequency. The growth of Artificial Intelligence (AI)
technology has resulted in the establishment of ultra-modern medical facilities, such
as hospitals and other health institutions. In these hospitals that are getting more and
more advanced, robotic surgery techniques are much better than traditional ones.
This means that patients can recover faster and have a lower chance of complica-
tions. In this chapter, we take a look at the many approaches to decision-making that
are utilized in the field of health and medicine. Machine learning, artificial intelli-
gence, the internet of things, deep learning, and natural language processing are just
a few of the major methods that are utilized in this discipline. This chapter also delves
into the topics of organizational decision-making in healthcare as well as medical
decision-making. In addition, the marketing of healthcare services and some of the
benefits of using marketing tactics for health services are both covered in this chapter.

Keywords Decision-making · Artificial intelligence · Healthcare marketing ·

Machine learning · Health sector

K. G. Nalbant (B)
Istanbul Beykent University, Software Engineering, Istanbul, Turkey
e-mail: [email protected]
S. Aydin
Istanbul Beykent University, Business, Istanbul, Turkey

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 1

T. Allahviranloo et al. (eds.), Decision Making in Healthcare Systems,
Studies in Systems, Decision and Control 513,
https://doi.org/10.1007/978-3-031-46735-6_1
2 K. G. Nalbant and S. Aydin

1 Introduction and Motivation

Decision-making is required in virtually every aspect of an individual’s life. There-

fore, developing a theory about decision-making is essentially the same as devel-
oping a theory about human activities. However, the choice theory is not exactly as
all-encompassing as that statement suggests. It concentrates only on certain elements
of human behavior. Particular attention is paid to the ways in which we make use of
the freedom we have. In the scenarios that are examined by decision theorists, there
are several courses of action from which to select, and the selections that we make
are not made at random. The actions we take in response to these circumstances are
ones that are goal-directed. So, decision theory looks at how people act toward their
goals even when they have choices [19].
“The act of deciding between two or more possible courses of action is known as
decision-making.” However, one thing that can never be forgotten is that there is not
necessarily a “right” choice to be made out of the options that are presented to one.
There may have been a better option that had not been examined, or the appropriate
knowledge may not have been accessible at the time. Both of these things might have
contributed to the situation. Problems that require the examination of multiple criteria
always include a set number of potential solutions, all of which are specified before
the issue-solving process begins. In multiple-criteria design challenges, also known
as mathematical programming problems with numerous objectives, the choices are
not always made clear. Through the process of solving a mathematical model, one
might discover an alternative solution. The number of possibilities is either infinite
or cannot be counted (where certain variables are continuous), or it is often very high
if it can be counted (when all variables are discrete). But each kind of problem can
be put into a subclass of problems that have to do with making decisions based on
more than one factor [27].
Making intelligent choices is something that everyone strives for. To be more
specific, those in decision-making roles have an incentive to make choices that will
result in favorable outcomes. What exactly does it mean to be “good”? In most cases,
it is a subjective measurement that represents the beliefs or inner reflections of the
one who is making the choice. However, the notion that the quality of a choice is
entirely dependent on the wishes of the person making the decision is not scientific.
This is because the only person who can objectively assess the desires of the person
making the decision is the person making the decision themselves. Even though
the person making the choice is the only one who can choose the “best” alternative
and has the final say, science may evaluate the method by which the decision is
reached. This is the scientific perspective, according to which a sound decision is
the result of sound procedures for making decisions. Analysts and academics should
be interested in ensuring that there is a sound process of decision-making, even
if decision-makers are often more concerned with the outcomes of their decisions
than the processes by which they get to those outcomes [21]. Methods referred
to as “multi-criteria decision making” (MCDM) deal with the process of making
judgments when there are a number of different factors to consider. The individuals
Methodologies for Decision-Making in the Health and Medicine Sector 3

in charge of making decisions have to pick one of three options: measurable, non-
quantifiable, or many criteria. As a rule, the aims are in direct opposition to one
another, hence, the answer must be some kind of accommodation and is heavily
reliant on the personal preferences of the one making the choice [20, 30]. One of
the techniques for decision making is Data Envelopment Analysis (DEA), and many
researchers have conducted research on this topic. Several recent papers have been
cited to mention their research area, [6–8], [4]. Some authors have explored the topic
in a different way: safety analysis and reliability, [2, 3, 5].

2 Literature Review

Drake et al. [15] proposed Multiple-Criteria Decision Analysis (MCDA) as a

decision-making tool applicable to the healthcare industry because of its complete,
consistent, adaptable, and transparent methodology, which encourages collaboration
among all healthcare stakeholders. Baek et al. [9] made use of the hybrid clustering-
based food recommendation method, which is characterized by the utilization of diet
and nutrition ontology, a diet and nutrition knowledge base, and chronic disease-
based clustering. Glaize et al. [16] examined the applications of MCDA methods in
order to provide structure and practical insights regarding the use of these methods
in a variety of healthcare settings. Domínguez and Carnero [14] decided to use the
Fuzzy Analytic Hierarchy Process (FAHP) as the basis for the design of a model that
would support their choice to upgrade the technology of medical equipment found
in hospitals. Loftus et al. [25] outlined the shortcomings of conventional clinical
decision-support systems and made the case for incorporating artificial intelligence.
Chowdhury et al. [12] presented an ensemble-based multi-criteria decision-making
(MCDM) method as a way to pick the top-performing machine learning technique(s)
for COVID-19 cough classification.

3 Decision-Making Techniques in the Medicine and Health

Sector

The use of machine learning (ML) in image retrieval systems for the purpose of
medical decision-making is becoming increasingly common. One use of machine
learning is to collect visually comparable medical pictures from previous patients (for
example, tissue from biopsies), which may then be used as a point of reference when
making a judgment on a new patient. However, it is impossible for any algorithm to
precisely capture an expert’s ideal concept of similarity in every situation: a picture
that is algorithmically found to be similar may not be medically relevant to the
specific diagnostic needs of a clinician [10].
4 K. G. Nalbant and S. Aydin

The Internet of Things (IoT), which has emerged as one of the most innovative
technologies in recent years, has been largely responsible for the paradigm shift that
has been brought about in conventional methods of providing medical treatment. IoT-
based eHealth aims to provide intelligent and individualized medical care services
by utilizing the principle of frictionless data sharing across connected devices, which
is then followed by effective data analytics [11].
The goal of what is known as “artificial intelligence” (AI) is to simulate human
cognitive abilities. It is bringing about a paradigm shift in the healthcare industry,
which is being propelled by the expanding availability of healthcare data and the
quick advancement of analytical tools. There are many different kinds of medical
data that can be processed by AI (structured and unstructured). Popular applications
of artificial intelligence include methods of machine learning for structured data, such
as the traditional support vector machine and neural network, as well as the more
recent deep learning and natural language processing for unstructured data. Both of
these methods can be used with structured data. Cancer, neurology, and cardiology
are examples of important illness areas that make use of AI capabilities [23].
It is becoming increasingly clear that the application of artificial intelligence (AI)
approaches will be critical to the improvement of clinical research and care. Natural
Language Processing (NLP) and Deep Learning (DL) techniques have been utilized
in order to extract information from a significant number of electronic health records
(EHR), the majority of which are locked in clinical narratives. In the context of
robotics-assisted operations, computer vision techniques can be used for medical
imaging, natural language processing techniques can be used for analyzing unstruc-
tured information in electronic health records, and reinforcement learning techniques
can be used as well [32]. While analyzing text and determining the grammatical rela-
tionships between phrases, NLP algorithms can be used to find clinically relevant
phenotypes. This can be done while the algorithms are also analyzing the text. In
clinical records, rule-based natural language processing approaches may be utilized
to achieve high sensitivity (the identification of a significant proportion of actual
cases) as well as high positive predictive value. One of the fields in which computer
science is becoming increasingly helpful in a wide variety of activities is the health-
care industry. Language processing is at the heart of many of the most exciting
new applications for artificial intelligence, and this trend is sweeping the health-
care industry from the most fundamental level practices all the way up to the most
specialized areas. The capabilities of these AI algorithms may enable the detection
of distinguishing clinical traits among patients, which aids in clinical treatment and
reduces methodological heterogeneity in medical research on a wide range of health
conditions [22].
Methodologies for Decision-Making in the Health and Medicine Sector 5

4 Medical Decision Making

The process of making medical decisions encompasses decisions made in medicine,

policy, or daily contexts that have an effect on the health of individuals or the general
population. When compared to other sorts of judgments, the process of making
medical decisions is usually fraught with risk and uncertainty, challenging trade-offs,
information overload, decisions regarding future consequences, and interdependent
behaviors, in addition to a dependence on data. In the field of medical decision-
making, research uses people’s psychological processes, especially the systematic
ways they deviate from rationality, to better understand and improve the health
outcomes of decisions, such as by “nudging” people toward healthy choices [24].
Since over half a century ago, formal research and other relevant applications
have focused on the process of decision-making in the medical field. There is a
professional and academic society devoted to studying and improving decision prac-
tices called the Organization for Medical Decision Making (SMDM). This society
hosts yearly meetings and publishes a magazine called Sage Publications: Medical
Decision Making (Sage Publications). There is also a paradigm for investigating
medical decision-making processes that is founded on the normative comparative
method. This paradigm has been around for a while and is pretty well established.
The archetype of the person who makes medical decisions is that of a stoic, emotion-
less, and utterly logical doctor who methodically considers well-defined possibili-
ties (i.e., therapeutic choices or diagnostic alternatives) on the basis of a meticulous
weighing of the facts. Equally typical is his or her colleague, who is definitely less
skilled: a faulty reasoner who is susceptible to biases and particularly incompetent
in applying probability theory to choice difficulties. Because of these inadequacies,
erroneous decision-making procedures commonly occur [29].
When it comes to medical matters, there is always an element of unpredictability
surrounding the consequences of decisions. Incomplete information can be gleaned
from clinical examinations and other diagnostic methods. The indications for the
majority of therapeutic approaches, in addition to the risks and benefits associated
with those approaches, are ill-defined or unknown. When it comes to the vast majority
of clinical issues, information gleaned from randomized clinical trials is either inac-
cessible or cannot be generalized to the patient in question. Probability theory, the
threshold model of decision-making, and expected value decision analysis are all
examples of useful tools that can be utilized to assist in navigating this uncertainty
[26].
With a few significant exceptions, teenagers do not have the legal ability to offer
permission for or refuse medical procedures. These exceptions are rare, though.
However, there are some circumstances in which the question arises over whether
or not a mature child should be allowed to make a life-altering medical choice that
would be contested if the decision were made by the minor’s parent [13].
6 K. G. Nalbant and S. Aydin

5 Organizational Decision-Making in Healthcare

Global healthcare spending will exceed $8.7 trillion by 2020. Many healthcare insti-
tutions lack the funds to replace their antiquated infrastructure and legacy technolo-
gies. Decision-makers are strategically shifting their focus toward population health
management, including analyses of health, quality, and cost trends; understanding
and better aligning healthcare providers’ financial incentives to bear financial risk;
and adopting innovative delivery models to improve processes and coordination of
care. This is a move toward value-based care [33].
Making “excellent judgments” is a significant part of the job that clinicians do
on a daily basis in the field of medicine. They need to properly diagnose illnesses
based on little data, and they need to do so in a short amount of time. Additionally,
they need to choose the optimal treatment approach among several options for the
patient they are now working with. Clinicians are highly competent professionals
in the aforementioned endeavors. They have been trained for a number of years,
and during the course of their careers, many of them have diagnosed and cared for
patients whose numbers are in the five-digit range. In addition, clinicians use a variety
of diagnostic instruments, such as medical imaging technologies, which enable them
to evaluate a patient’s physical health with a great deal of anatomical specificity. In
conclusion, making a diagnosis in the medical field is frequently a group effort since
the patient herself as well as other clinical professionals are consulted along the way
[17].
The framework in which healthcare managers function is becoming increasingly
complicated. The impact of management decisions on employees in the workplace
and on the company’s performance is direct, and these decisions are impacted by
a variety of other elements that might result in financial success, customer happi-
ness, and long-term sustainability for the firm. People are directly affected by these
decisions because of the many changes happening in the economy, the law, ethics,
organizations, and technology. Making decisions is an essential component of all
management activities, and it has a tight relationship with the function of planning.
Every manager, regardless of their rank, is responsible for making choices. However,
the judgments made by senior managers have a wider scope, involve more people,
and have a higher impact than the ones made by first-line supervisors. The process
of making decisions involves selecting the most advantageous alternative in order to
accomplish certain personal and organizational goals. Not every manager is familiar
with the steps involved in decision-making. Managers can make better decisions
when they know and follow the steps of the decision-making process [18].
Methodologies for Decision-Making in the Health and Medicine Sector 7

6 Healthcare Marketing

Hospitals, pharmaceutical companies, medical device manufacturers, telemedicine

companies, medical tourism companies, health insurance companies, and compa-
nies that conduct clinical trials are all included in the healthcare industry. Lifestyle
diseases, rising demand for affordable healthcare, technological advances, more
people having health insurance, and government programs like increasing spending
on public health and e-health, as well as private spending on new ways to deliver
healthcare services, have all helped the sector grow quickly. Additionally, customers
of medical services have developed a heightened awareness of the rights and
responsibilities associated with their own healthcare maintenance [28].
The marketing of healthcare services is an example of an interdisciplinary field
due to the fact that it makes use of ideas, approaches, and strategies that are distinctive
to both traditional and social marketing. The marketing of healthcare products and
services is unique in that there is neither a product nor a market that can be monetized.
This indicates that the effectiveness of its application can be found in the image of
a healthy population, the detection of a category of people who are chronically ill,
the ensuring of the treatment of sick people by going through the entirety of the
rehabilitation process, the professional the social reintegration of sick people, etc.
Because of the state of the population’s health, there was no choice but to implement
marketing strategies in the healthcare industry. The following are some advantages
to putting marketing strategies into action [31]:
. to gain a competitive advantage,
. to gain visibility,
. to establish a solid reputation among patients,
. to comprehend consumer needs and expectations,
. to comprehend patients’ perceptions of the quality and outcomes of their
experience within the medical organization,
. to provide memorable experiences to patients, and, of course, to establish a strong,
effective, and dominant brand in the health services market.

7 Conclusion

As is the case with every industry that is undergoing development and transfor-
mation, the use of artificial intelligence has resulted in a number of developments
and enhancements in the disciplines of medicine and healthcare. The accuracy of
the diagnosis is improved by AI, and this allows doctors to begin treatment earlier,
before the problem even manifests itself. As a result, the patient can begin treat-
ment sooner, which boosts the patient’s chances of making a full recovery. Patients
are becoming increasingly open to the use of AI and robots in the medical industry
as they search for more effective healthcare; the application of AI to the fields of
medicine and health will lead to the discovery of new medications. By investigating
8 K. G. Nalbant and S. Aydin

the diseases that individuals have, which is done via gene research, it will be possible
to forestall the conditions that would result in the appearance of an infection. As a
direct result of this, it is possible to avoid contracting the condition, which in turn
enables people to enjoy a long and healthy life. The use of robots in the health
and medical fields contributes greatly to the reduction of the amount of time spent
on treatment as well as the amount of work done by doctors. In addition to that, the
circumstances will provide a new perspective on the practice of medicine. Health and
medical technology have come a long way, and both people and robots have helped
a lot. The advancement of artificial intelligence gadgets will allow future genera-
tions to live lives that are both healthier and longer than their predecessors. The
use of artificial intelligence technologies has made it feasible for clinicians to treat
patients remotely during the pandemic (COVID-19). Interaction between medical
staff and patients should be kept to a minimum. Healthcare experts made an effort to
protect the patient’s health by utilizing equipment that utilizes artificial intelligence.
Clinicians are able to diagnose patients, prescribe medications, and provide treat-
ments without ever having to communicate directly with the patients. Additionally,
positive identification of COVID-19 cases is achievable through the utilization of arti-
ficial intelligence. When it comes to the education of medical professionals, artificial
intelligence, robots, and other forms of technology are used rather frequently. Virtual
reality (VR) technology that shows human anatomy has made it simpler for students
of medicine and health to learn new information. The use of digital technology in
the process of decision-making is of crucial relevance in the field of medicine and
health.

References

1. Abbasi, F., Allahviranloo, T.: Conception and implementation of a new data-driven fuzzy
method for reliability and safety analysis. New Math. Nat. Comput. 16(02), 339–361 (2020).
https://doi.org/10.1142/s1793005720500210
2. Abbasi, F., Allahviranloo, T.: The fuzzy arithmetic operations of transmission average on
Pseudo-Hexagonal fuzzy numbers and its application in fuzzy system reliability analysis. Fuzzy
Inf. Eng. 13(1), 58–78 (2021). https://doi.org/10.1080/16168658.2021.1915449
3. Abbasi, F., Allahviranloo, T.: Realistic solution of fuzzy critical path problems, case study: the
airport’s cargo ground operation systems. Granul. Comput. 8(3), 617–632 (2022). https://doi.
org/10.1007/s41066-022-00347-w
4. Akram, M., Shahzadi, S., Shah, S.M.U., Allahviranloo, T.: A fully Fermatean fuzzy multi-
objective transportation model using an extended DEA technique. Granul. Comput. (2023).
https://doi.org/10.1007/s41066-023-00399-6
5. Allahviranloo, T., Abbasi, F.: A new estimation of failure analysis in fuzzy environment, case
study: the electrical model failure for the football stadium. New Math. Nat. Comput. 18(03),
791–817 (2022). https://doi.org/10.1142/s1793005722500387
6. Amirteimoori, A., Allahviranloo, T., Kordrostami, S., Bagheri, S.F.: Improving decision-
making units in performance analysis methods: a data envelopment analysis approach. Math.
Sci. (2023). https://doi.org/10.1007/s40096-023-00512-5
7. Amirteimoori, A., Allahviranloo, T., Zadmirzaei, M.: Scale elasticity and technical efficiency
analysis in the European forest sector: a stochastic value-based approach. Eur. J. Forest Res.
(2023). https://doi.org/10.1007/s10342-023-01589-2
Methodologies for Decision-Making in the Health and Medicine Sector 9

8. Amirteimoori, A., Allahviranloo, T., Zadmirzaei, M., Hasanzadeh, F.: On the environmental
performance analysis: a combined fuzzy data envelopment analysis and artificial intelli-
gence algorithms. Expert Syst. Appl. 224, 119953 (2023). https://doi.org/10.1016/j.eswa.2023.
119953
9. Baek, J.W., Kim, J.C., Chun, J., Chung, K.: Hybrid clustering based health decision-making
for improving dietary habits. Technol. Health Care 27(5), 459–472 (2019)
10. Cai, C.J., Reif, E., Hegde, N., Hipp, J., Kim, B., Smilkov, D., Terry, M.: Human-centered tools
for coping with imperfect algorithms during medical decision-making. In: Proceedings of the
2019 Chi Conference on Human Factors in Computing Systems, pp. 1–14 (2019)
11. Chatterjee, P., Cymberknop, L.J., Armentano, R.L.: IoT-based decision support system for intel-
ligent healthcare—applied to cardiovascular diseases. In: 2017 7th International Conference
on Communication Systems and Network Technologies (CSNT), pp. 362–366. IEEE (2017)
12. Chowdhury, N.K., Kabir, M.A., Rahman, M.M.: An ensemble-based multi-criteria decision
making method for COVID-19 cough classification. arXiv:2110.00508 (2021)
13. Diekema, D.S.: Adolescent brain development and medical decision-making. Pediatrics
146(Supplement_1), S18–S24 (2020)
14. Domínguez, S., Carnero, M.C.: Fuzzy multicriteria modelling of decision making in the renewal
of healthcare technologies. Mathematics 8(6), 944 (2020)
15. Drake, J.I., de Hart, J.C.T., Monleón, C., Toro, W., Valentim, J.: Utilization of multiple-criteria
decision analysis (MCDA) to support healthcare decision-making FIFARMA, 2016. J. Mark.
Access Health Policy 5(1), 1360545 (2017)
16. Glaize, A., Duenas, A., Di Martinelly, C., Fagnot, I.: Healthcare decision-making applications
using multicriteria decision analysis: a scoping review. J. Multi-Criteria Decis. Anal. 26(1–2),
62–83 (2019)
17. Grote, T., Berens, P.: On the ethics of algorithmic decision-making in healthcare. J. Med. Ethics
46(3), 205–211 (2020)
18. Guo, K.L.: DECIDE: a decision-making model for more effective decision making by health
care managers. Health News 39(3), 133–141 (2020)
19. Hansson, S.O.: Decision theory. A Brief Introduction. Department of Philosophy and the
History of technology. Royal Institute of Technology, Stockholm (1994)
20. Harputlugil, T.I.M.U.C.I.N., Prins, M.A.T.T.H.I.J.S., Gültekin, A.T., Topçu, Y.I.: Conceptual
framework for potential implementations of multi criteria decision making (MCDM) methods
for design quality assessment. In: Management and Innovation for a Sustainable Built Environ-
ment MISBE 2011, Amsterdam, The Netherlands, June 20–23. CIB, Working Commissions
W55, W65, W89, W112; ENHR and AESP (2011)
21. Henig, M.I., Buchanan, J.T.: Solving MCDM problems: process concepts. J.Multi-Criteria
Decis. Anal. 5(1), 3–21 (1996)
22. Jain, K., Prajapati, V.: NLP/Deep learning techniques in healthcare for decision making. Prim.
Health Care Open Access 11(3), 1–4 (2021)
23. Jiang, F., Jiang, Y., Zhi, H., Dong, Y., Li, H., Ma, S., Wang, Y.: Artificial intelligence in
healthcare: past, present and future. Stroke Vasc. Neurol. 2(4) (2017)
24. Li, M., Chapman, G.B.: Medical decision making. The Wiley Encyclopedia of Health
Psychology, pp. 347–353 (2020)
25. Loftus, T.J., Tighe, P.J., Filiberto, A.C., Efron, P.A., Brakenridge, S.C., Mohr, A.M., Bihorac,
A.: Artificial intelligence and surgical decision-making. JAMA Surg. 155(2), 148–158 (2020)
26. Lurie, J.D., Sox, H.C.: Principles of medical decision making. Spine 24(5), 493–498 (1999)
27. Majumder, M.: Impact of urbanization on ưater shortage in face of climatic aberrations. Springer
(2015)
28. Mehta, S.: Healthcare marketing. In: Healthcare System Management: Methods and Tech-
niques, pp. 239–260. Springer Nature Singapore, Singapore (2022)
29. Patel, V.L., Kaufman, D.R., Arocha, J.F.: Emerging paradigms of cognition in medical decision-
making. J. Biomed. Inform. 35(1), 52–75 (2002)
30. Pohekar, S.D., Ramachandran, M.: Application of multi-criteria decision making to sustainable
energy planning—a review. Renew. Sustain. Energy Rev. 8(4), 365–381 (2004)
10 K. G. Nalbant and S. Aydin

31. Purcarea, E.V.L.: The impact of marketing strategies in healthcare systems. J. Med. Life 12(2),
93 (2019)
32. Seyyedabbasi, A.: A reinforcement learning-based metaheuristic algorithm for solving global
optimization problems. Adv. Eng. Softw. 178, 103411 (2023). https://doi.org/10.1016/j.adveng
soft.2023.10341
33. Shahid, N., Rappon, T., Berta, W.: Applications of artificial neural networks in health care
organizational decision-making: a scoping review. PLoS ONE 14(2), e0212356 (2019)
The Application of System Simulation
in the Health Sector: A Rapid Review

Mohammadreza Mobinizadeh, Marita Mohammadshahi, Parisa Aboee,

Zeinab Fakoorfard, Alireza Olyaeemanesh, and Efat Mohamadi

Abstract Introduction: Implementation of applied mathematics aligned with

evidence informed decision-making methods can be used for improving of health
system performance. This research intends to help the policymakers decide on the
use of simulation methods (with operations research techniques) in the health care
by reviewing the available evidence. Methods: By such rapid-review, Cochrane,
PubMed and Google Scholar databases have been searched by April 2023. The
inclusion criteria were studies that investigated on the use of various kind of simula-
tion (with operations research techniques) on the health policy context. Results: On
the base of retrieved data, system dynamics (SD), discrete event simulation (DES),
and agent-based modeling (ABM) were the most commonly used methods. SD is a
way of creating computer models that show how complex systems work and change
over time. DES refers to how things happen to one person, and what they experience
in those situations. The ABM model looks at separate things and shows how they
behave by following simple rules. Conclusion: These models (DS) can help combine
feedback information with real-time data to create useful tools for managing health
care and making health system policies. This new way of collecting information can
help decision makers make better decisions.

M. Mobinizadeh · M. Mohammadshahi · P. Aboee · Z. Fakoorfard · A. Olyaeemanesh (B)

National Institute for Health Research, Tehran University of Medical Sciences, Tehran, Iran
e-mail: [email protected]
M. Mobinizadeh
e-mail: [email protected]
M. Mohammadshahi
e-mail: [email protected]
P. Aboee
e-mail: [email protected]
Z. Fakoorfard
e-mail: [email protected]
A. Olyaeemanesh · E. Mohamadi
Health Equity research center (HERC), Tehran University of Medical Sciences, Tehran, Iran
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 11

T. Allahviranloo et al. (eds.), Decision Making in Healthcare Systems,
Studies in Systems, Decision and Control 513,
https://doi.org/10.1007/978-3-031-46735-6_2
12 M. Mobinizadeh et al.

Keywords Dynamic simulation · Health sector · System dynamic simulation ·

Discrete event simulation · Agent based modeling

1 Introduction

Applied mathematical methods can be employed to enhance the performance of the

health system by incorporating effective organizational practices. System simulation
is a prevalent operational research methodology that has been applied in healthcare
systems due to its adaptable nature, capability to handle uncertainty and variability,
and utilization of visual interfaces to enhance communication and comprehension
among healthcare personnel. System simulation can be categorized in two main
forms, Static Simulation (SS) and Dynamic Simulation (DS) [1–3]
The static model considered the problem structural view, which is not varied over
time [2]. This models can be considered as a ‘snapshot’ of a system’s response to a
specific input conditions [4] while DS modeling approaches can develop mathemat-
ical representations of the system operation to test interventions and scenarios and
the resulting outcomes over time to promote the understanding of the system [1].
System dynamics (SD), discrete event simulation (DES), and agent-based modeling
(ABM) were the most commonly used DS modeling methods. Simulation modeling
called SD can show how complicated things work. The concept was created by Jay
Forrester in the 1950s and is based on industrial patterns [1].
The term DES can be used to show how things happen to people over time. DES
is a way to study queues and how resources are used. The important ideas in DES are
events (things that happen), entities (things that do something), resources (things that
are used), attributes (characteristics of things), and queues (lines of things waiting)
[1].
The ABM model considered individual objects and introduces their local behavior
with local rules. The ABM method works from the bottom, while SD and DES
start at the top. Thomas Schelling’s segregation model is an old agent-based models
developed in 1971. An ABM model works with things called agents, which are
objects that can move and interact on their own [1]. For instance, how individuals
tend to gather in certain places because of expected actions [4, 5].
According to the mentioned contents, this paper intends to help the policymakers
decide on the use of simulation methods (with operations research techniques) in the
health care by reviewing the available comprehensive evidence.

2 Method

This study was a rapid review on the studies that investigated on the use of various
kind of simulation on the health policymaking which was performed in four phases:
The Application of System Simulation in the Health Sector: A Rapid … 13

Table 1 The list of included papers

No Title Publication References
year
1 Selecting a dynamic simulation modeling method for 2015 [1]
healthcare delivery research—Part 2: report of the ISPOR
dynamic simulation modeling emerging good practices task
force
2 Static modeling. In software modeling and design: UML, Use 2012 [2]
Cases, Patterns, and Software architectures (pp. 94–114)
3 Modeling the transmission of community-associated 2014 [5]
methicillin-resistant Staphylococcus aureus: a dynamic
agent-based simulation
4 Applying dynamic simulation modeling methods in health 2015 [6]
care delivery research—the SIMULATE checklist: report of
the ISPOR simulation modeling emerging good practices task
force
5 Improving health care management through the use of 2012 [7]
dynamic simulation modeling and health information systems
6 Analyzing national health reform strategies with a dynamic 2010 [8]
simulation model
7 A system dynamics simulation applied to healthcare: A 2020 [9]
systematic review
8 A system dynamics simulation applied to healthcare: A 2020 [10]
systematic review
9 Discrete-event simulation in healthcare settings: A review 2022 [11]

(1) Searching the electronic library on PubMed, Cochrane, and Google Scholar
using keywords, including DS, Health Sector, System Dynamic Simulation,
Discrete Event Simulation, and Agent based Modeling
(2) Screening the obtained papers using inclusion criteria
(3) Data extraction by an organized data extraction form
(4) Data analysis thematically
On the base of this process 9 papers were selected for reviewing (Table 1), retrieved
data was analyzed via thematic analysis by three main themes (System Dynamic
Simulation, Discrete Event Simulation, Agent based Modeling).

3 Result

DS modeling methods can design mathematical representations of the process oper-

ation to assess scenarios and interventions as well as their consequences with the
passage of time to promote the understanding about the process, for policy design.
14 M. Mobinizadeh et al.

From retrieved data which was extracted from retrieved paper three main
System Dynamic Simulation, Discrete Event Simulation, Agent based Modeling
methods were discussed.

4 System Dynamic Simulation (SD)

Forrester at the Massachusetts Institute of Technology designed the first SD

simulation which was called DYNAMO [6].
In SD models, people are studied in groups called states, instead of looking at each
person individually. This means that SD models can show which people have certain
traits or behaviors. Stock, flow, or auxiliary are different things we can measure [1].
Stocks are an accumulation of something, such as, people, beds, etc. The flow
variables alter the stock accumulation. Flows feed out and in of stocks with the same
units of stocks for each time unit, for instance, beds per year. The auxiliary variable
is calculated values that can affect inflows and outflows or [1].
SD can help solve problems and analyze policies in complicated systems. This
means that dynamic systems work together, depend on each other, and give each
other feedback. This simulation has main important features as finding patterns in
how something works; figuring out what happens when a certain choice is made;
finding point where changes can be made to improve the whole system; and copying
how something has been done before [1].
Generally, SD models are developed in specific phases, such as problem descrip-
tion, generating a system structure diagram, conversion of the qualitative hypoth-
esis to a quantitated simulation model, model assessment, and making policy-
makers informed about the model implications [10]. This model can be designed
via softwares like Vensim or Anylogic.
Milstein et al. using SD evaluated the US health system reform with three major
strategies: “Expanding health insurance coverage”, “delivering better chronic
and preventive care”, “protection of health through improving environmental
conditions and enabling healthier behavior”. This model respond questions on
the effect of these strategies at the national level, which is a good example of
a problem with wide implications requiring a holistic view and attention toward
dynamic processes in the system [8].
The developers anticipated the combined effects of the three strategies between
2000 and 2010 and asked about the events making the United States take decisive
action in these three areas to reduce avoidable deaths and decrease health care costs.
Simulated scenarios indicate that these three strategies can save millions of deaths
and also offer good economic value [8].
Such simulations report the cost-effectiveness of a strategy to expand insurance
coverage and improve health care quality, however, if it is performed with no other
interventions can provide modest improvements in health status and increase costs
and worsen health inequities [8].
The Application of System Simulation in the Health Sector: A Rapid … 15

The system dynamics focusing on dynamic problems caused by complex systems,

is often applied to healthcare. The appeal of using the methodology in healthcare
focuses on information feedback, interdependence, and generating actionable model-
based insights [7].

5 Discrete Event Simulation (DES)

DES has been applied since the 1950s in military operations, supply chain
management, manufacturing, and computer and network design [11].
In healthcare system, DES analyzes the effects on health outcomes and is effective
for problems where it is particularly relevant to be able to account for variable
properties of entities and where processes are characterized by events [6]. This model
can be designed via softwares like Arena or Anylogic.
Patients are different from one another and how we take care of them takes a lot
of time and resources. This happens when they first come in to visit and when they
need to be admitted to the hospital. Patients stand in queues for these treatments, go
through them, and finally leave the hospital [1].
DES can facilitate decision-making for a health system to invest in emergency
department (ED) and/or intensive care unit (ICU) expansion according to variable
patient flow [9].
Patient flow to the hospital is often limited by ED capacity; ICUs limit flow when
there is high rate of referrals, or when patient flow raises from the other health system
parts [9].
Therefore, next patients who require critical care are kept in the ED, and cases that
may have high-revenue appointments like surgery should be cancelled and rebooked
[9].
When the hospital emergency room doesn’t have any beds available, they can’t
accept more patients for treatment. Many hospitals are thinking about making the
emergency room bigger or expanding other units, like the ICU, to help patients move
through more easily [9].

6 Agent Based Modeling (AB)

Schelling employed ABM to suggest a theory to describe the racial segregation persis-
tence; however, the cultural and legal environment was one of growing tolerance. A
basic ABM model with if–then statements indicated that using colored squares on a
matrix, segregation is the equilibrium situation [6].
In the realm of interactive agents, individuals consistently engage in environ-
mental interaction, perceiving and reacting to stimuli in accordance with behav-
ioral decision-making principles. The behaviors of the agent are determined by
mathematical logic operators [1].
16 M. Mobinizadeh et al.

The principal notions underpinning agent-based modeling (ABM) consist of

agency, dynamics, and structure. The concept of agency entails the capacity of
agents, namely patients or providers, to engage in active behavior and interact with
one another, resulting in the acquisition and sharing of innovative knowledge within
the social network. The concept of dynamics pertains to the temporal fluctuations
observed in the entities interacting within their respective milieu. The emergence of
structure is a result of the interaction among agents [1]. This model can be designed
via softwares like Netlogo or Anylogic.
Macal et al. used ABM to identify target interventions to reduce transmission
methicillin-resistant Staphylococcus aureus (CA-MRSA) in Chicago [5].
They designed a model based on agent to represent heterogeneity in population
behavior, locations, and contact patterns. The Chicago CA-MRSA ABM included
locations, like workplaces, households, hospitals and schools. The agents in the
ABM possesses a “daily activity profile” showing the times he/she can occupy each
location. Social contact between agents happens when several agents occupy one
location at the same time [5].

7 Discussion

DS modeling is often used to study complex issues in healthcare systems, such

as how policies and strategies affect health outcomes. It is used to plan for the
future of healthcare systems. DS has some advantages like collecting data without
spending too much money, making complicated problems easy to understand, and
finding suitable solutions. These models (DS) can help combine real-time data with
feedback modeling to make better healthcare policy decisions. Intervention planners
can use it to learn how the health system can react to different situations and make
it better. Simulation models show how a system changes over time. They illustrate
the different states the system can be in and how it transitions from one state to
another. But these models have problems. They can’t fully explain the small actions
of healthcare workers and it can be hard to check if they’re right when dealing with
many things at once.
The people in charge of hospitals were thinking about ways to make them
work even better. Using a computer program that shows how different parts of a
system affect each other over time. The managers’ skill in managing how patients,
doctors, pharmacists, and nurses interact with each other was crucial to how well
the whole system worked. DES has been applied in medical emergency system and
AB modeling can by agents which can disseminate novel learning to other agents in
the social network. Using discrete event simulation in healthcare is still growing.
These models cause process efficiency and resource allocation. Further expan-
sion of discrete event simulation into clinical simulations indicates its flexibility,
adaptability, and utility.
ABM is a helpful way of planning for public health because it is getting better
and better. This plan includes goals that focus on being efficient, improving people’s
The Application of System Simulation in the Health Sector: A Rapid … 17

health, making profits, and saving money. However, the ABM has been used occa-
sionally in population health, it is discussed that the ABM can be the most effective in
the field and it can be used as a tools for answering questions which didn’t normally
have access to the traditional epidemiological toolkit.
Clear models of data and systems can help people in charge better understand how
things work, figure out what might happen if they try different things, and decide
what to do to make things better. Complicated models can be used to combine
information about health systems and diseases to help policymakers and healthcare
providers understand how to work together. This will make it easier to create strong
public health programs and reduce risks without putting too much stress on healthcare
services. Using simulation modeling and new data gathering methods can really help
clinicians make better decisions about medical treatments.

References

1. Marshall, D.A. et al.: Selecting a dynamic simulation modeling method for health care delivery
research—Part 2: report of the ISPOR dynamic simulation modeling emerging good practices
task force. Value Health 18(2), 147–160 (2015)
2. Gomaa, H.: Static modeling. In Software Modeling and Design: UML, Use Cases, Patterns,
and Software Architectures, pp. 94–114. Cambridge University Press, Cambridge (2011).
doi:https://doi.org/10.1017/CBO9780511779183.009
3. Brailsford, S.C., Hilton, N.A.: A comparison of discrete event simulation and system dynamics
for modelling health care systems (2001)
4. Anynoumus. Access at URL: http://www.edscave.com/static-vs.-dynamic-models
5. Macal, C.M., North, M.J., Collier, N., Dukic, V.M., Wegener, D.T., David, M.Z., Lauderdale,
D.S.: Modeling the transmission of community-associated methicillin-resistant Staphylococcus
aureus: a dynamic agent-based simulation. J. Trans. Med. 12(1), 1–12 (2014)
6. Marshall, D.A., Burgos-Liz, L., IJzerman, M.J., Osgood, N.D., Padula, W.V., Higashi, M. K.,
Crown, W.: Applying dynamic simulation modeling methods in health care delivery research—
the SIMULATE checklist: report of the ISPOR simulation modeling emerging good practices
task force. Value Health 18(1), 5–16 (2015)
7. Goldsmith, D., Siegel, M.: Improving health care management through the use of dynamic
simulation modeling and health information systems. Int. J. Inf. Technol. Syst. Approach
(IJITSA) 5(1), 19–36 (2012)
8. Milstein, B., Homer, J., Hirsch, G.: Analyzing national health reform strategies with a dynamic
simulation model. Am. J. Public Health 100(5), 811–819 (2010)
9. Troy, P.M., Rosenberg, L.: Using simulation to determine the need for ICU beds for surgery
patients. Surgery 146(4), 608–620 (2009)
10. Davahli, M.R., Karwowski, W., Taiar, R.: A system dynamics simulation applied to healthcare:
a systematic review. Int. J. Environ. Res. Public Health 17(16), 5741 (2020)
11. Forbus, J.J., Berleant, D.: Discrete-event simulation in healthcare settings: a review. Modelling
3(4), 417–433 (2022)
Data Science in the Field of Health

Handan Kulan and Ezgi Özer

Abstract Data science in healthcare has made great progress using data analysis and
machine learning methods that have the potential to detect and help solve healthcare
problems. After mortality and during morbidity, relevant data about a health problem
have been gathered. This massive amount of data in various forms needs to be handled
for any healthcare issues are significant. With the growth of big data in healthcare
communities, accurate analysis of medical data has the benefits of early disease
detection, improved patient care, and effective community services. Because of its
significance, there is a need to develop efficient and better-performing data analytics
techniques and tools to analyze medical big data from the gene level to the clinical
level. The purpose of data analytics in healthcare is to find new insights in data, at least
partially automate tasks such as diagnosing, and to facilitate clinical decision-making.
Also, healthcare analytics has the potential to reduce costs of treatment, predict
outbreaks of disease, avoid preventable diseases, and improve the quality of life in
general. The average human lifespan is increasing across the world population and
the application of big data analytics in healthcare are increasing day by day. Different
format of health data are used for different types of analyses. For example, IoT gadgets
are used by certain patients and clinicians as wearable monitors to track heartbeat and
temperature. This signal generated data should be carefully analyzed over time. Also,
scans such as X-rays, magnetic resonance images (MRIs), and computed tomography
(CAT) scans can be studied with different data analysis techniques and machine
learning algorithms to visualize the insides of the body in depth. In addition, data
on individual cases of disease are analyzed; data received as text must be sorted,
categorized, and coded for statistical analysis; and data from surveys might need
to be weighted to produce valid estimates for sampled populations. The number
of resources healthcare professionals can obtain from their patients continues to
increase. Since these data are normally in different formats and sizes, it can be
difficult for analysis. However, the current focus is no longer on how big the data is,

H. Kulan (B) · E. Özer

Istinye University, Computer Engineering, Istanbul, Turkey
e-mail: [email protected]
E. Özer
e-mail: [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 19

T. Allahviranloo et al. (eds.), Decision Making in Healthcare Systems,
Studies in Systems, Decision and Control 513,
https://doi.org/10.1007/978-3-031-46735-6_3
20 H. Kulan and E. Özer

but how intelligently it is managed by data analysis techniques and machine learning
algorithms. This study examines analytical techniques for different forms of health-
related data to generate comprehensive healthcare reports and transform them into
relevant critical insights that can then be used to provide better care.

Keywords Healthcare · Data analytics · Signal processing · Image processing ·

Machine learning

1 Introduction and Motivation

Today, the increasing volume of digital data has created new problem areas. The
main difficulties are developing methods or systems to handle large amounts of
multidimensional and complex data, new types of data and distributed data, and to
use and security of data. The enormous amount of health-related data has made it
difficult to process these data by traditional data processing methods and has led
to the introduction of the concept of big data into health services. They can be
analyzed completely and quickly using data science to transform high-volume, fast,
and diverse data sets that cannot be stored, managed, and analyzed into meaningful
and value-creating results.
Data Science in healthcare collects data and analyzes it with the help of data
analytic techniques. The purpose of healthcare data analysis is to predict and solve
problem using data-driven findings in a quick way. From the genet level, molecular
level, to the clinical level in major healthcare areas such as electronic health record
maintenance, disease diagnosis, and prediction of emergency conditions of patients,
data are accumulated. This big-style health data refers to the vast quantities of infor-
mation created by the digitization of everything, that gets consolidated and analyzed
by specific analytic techniques. Collected data can be any format; text document,
image file or time-stamped data. Doctors can monitor a patient’s blood pressure,
circadian cycle, and calorie intake using laboratory equipment or analytical tools. If
the disease will be tested on animals, an experiment will be conducted in the labo-
ratory and then the data will be evaluated. Documentation output from laboratory
tests or analytic tools can be evaluated to identify critical factors in disease. In the
preprocessing processes of the document, the missing values must be filled and then
normalized. Also, text values need to be converted to numeric values in order to
process them with analytical techniques or machine learning algorithms.
Early diagnosis and disease classification are the most important elements of
disease management. Investigating disease findings and further examination not only
reduces the risk of death of patients but also increases the quality of life of the patients.
Depending on the nature of the disease to be diagnosed, both the syndromes seen
in the patients and the imaging and/or by analyzing activities recorded as time or
frequency series, diseases can be diagnosed. The stage or severity of the disease can be
classified, with the help of the findings obtained. Statistical and visual data is obtained
from past patients in order to make diagnoses and predictions. In addition, in order
Data Science in the Field of Health 21

to make these diagnoses and inferences, in general terms decision support-based

software programs based on statistical and mathematical methods are used.
In recent years, parallel to the development of technology, computers, and
computer-aided systems have had an important place in human life. The studies
carried out in order to summarize the data, establish a relationship between the data,
and interpret the results as a result of these relationships has developed the concept
of artificial intelligence. A computer system created using artificial intelligence tech-
niques establishes relationships between events using the available data and has the
ability to predict new events to come.
There are different methods for all processes in preprocessing steps. After prepro-
cessing steps, the feature selection methods can be applied to determine critical
factors in disease. By evaluating accuracy result of machine learning algorithms, the
reliability of text data evaluation is determined. In addition to numeric data evalu-
ation, to characterize trends and detect changes in disease incidence, data must be
evaluated either over time span or across image area. Evaluation of numeric data
alone pinpoints key factors and provides an overview of the disease process. Time
span data show functional change during the incidence and image analysis high-
lights structural abnormality. Therefore, health data in different formats should be
evaluated together in order to evaluate the disease process comprehensively. Thus, a
systematic evaluation of the disease process can be made and the underlying factors
can be understood.

2 Literature Review

2.1 Numeric Data Evaluation

Healthcare data analytics refers to the collection and analysis of patient data to
improve medical care and patient experience. Patients go through a continuum of
caregiving from diagnosis to recovery. The data can be obtained from laboratory
output, customer service calls, online forms and other digital methods. If the data
is in text form, natural language processing (NLP) techniques can be applied for
conducting proper text mining. The text data is converted into numeric form by
applying different filter methods and algorithms so that it can be understood by
computers. The type of analysis is divided into semantic search, sentiment analysis,
and named entity recognition (NER). Artificial intelligence, in the form of machine
learning, can be applied to this type of analytics to make data analysis faster and more
accurate. Combining AI and NLP in healthcare, the semantic insights are gained.
If the output is in the formatted type like excel, csv, sql file format, the missing
values must be filled in and the datasets must be normalized at the preprocessing
step. Datasets which contain missing values result to misleading predictions for the
unknowns. Thus, the missing values must be replaced with different values such as
median value, mean value and most frequent term. After replacement of missing
22 H. Kulan and E. Özer

values with appropriate value, all features must be normalized to prevent factor with
higher values influence on the classification result erroneously. There are different
normalization methods such as Z-score normalization, Gaussian Normalization [1].
After preprocessing step, the proper machine learning can be applied to select crit-
ical features in the dataset [2]. Using grid search method in the algorithm, different
combination of hyperparameters can be determined for each model. Then, classi-
fication algorithm is applied and best model is selected by taking into account of
classification accuracy. There are many classification algorithms such as Artificial
Neural Network (ANN), Gradient Boosting Tree (GBM), Support Vector Machine
(SVM) [2–5]. The common goal of all classification algorithms is to minimize loss
function and make the prediction more accurate. For example, in boosting algorithm,
a strong learner is created by adding a new weak learner to each iteration. With each
iteration, new models are built to overcome errors made in previous iterations and
at the end, the loss function is minimized. In the classification algorithm, the grid
search technique can also be used to create a model for each possible combination
of all hyperparameter values such as the maximum depth of each tree, the number
of trees, the learning rate, and the minimum number of observations at the terminal
nodes of the trees in GBM. The accuracy of the grid search method is found by
comparing the results according to different hyperparameter values. Each model in
a different combination of parameters must be evaluated and the model that gives
the most accurate result must be selected. Furthermore, k-fold cross-validation must
be applied to evaluate the performance of the model. In k-fold cross validation, part
of the data that is not used to train the model is then used to test this sample. Thus,
every observation in the dataset has the opportunity to appear in the training and test
set. After classification process, with the help of PCA cluster analysis, the selected
features can be clustered and projected onto reduced 2D axes. (Bontempi et al. 2008;
Hotelling 1936) [6, 7] PCA is a method for finding the projection of data onto a
low-dimensional axis in such a way as to maximize variance. With the Kmeans
clustering algorithm, the data points are divided into k sets, where each data point
belongs to the set with the nearest average, and the data can be effectively clustered
with decreasing axis size. This PCA cluster method provides virtual evidence for
accuracy of classification result.

2.2 Time Spanned Health Data Evaluation

On the time-spanned health data evaluation, signal processing is one of the data
collection methods, solving problems by making data meaningful for the correct
interpretation, reducing noise, reconstructing the signal, data compression, and
obtaining important information. They examine the activities of the organism, from
gene-protein sequences to nerve and heart rhythms, tissue, and organ images. Signal
processing includes a set of mathematical and statistical techniques used in the anal-
ysis, processing, and interpretation of biomedical signals. The methods of filter design
of signal processing are the discrete-time filter design based on amplitude response,
Data Science in the Field of Health 23

based on the length of the impulse response, and based on phase response to convert
the signal from analog to digital [8–10]. These transformed signals are classified
as continuous and discrete time signals according to the time variable, periodic and
non-periodic according to their periodicity, and deterministic and random according
to their statistical characteristics [8].
Biological signals are examined in two separate groups, electrical or non-electrical
origin, which are detected by electrodes or transducers from the living body. These
signals provide information on any pathological condition in the human body. Some
of these signs and their sources are Electromyogram (EMG), Electrocardiogram
(ECG), Electroencephalogram (EEG), Electrogastrogram (EGG), and Electrodermal
activity (EDA) [11]. They are often corrupted by artifacts, noise, and missing
data. Artifacts are generally divided into two groups which are physiological and
non-physiological artifacts. Physiological artifacts come from a source other than
the signal to be recorded from eye movement artifacts, muscle activity artifacts,
motion artifacts, pulse artifacts, sweating artifacts, and respiratory artifacts. Non-
physiological artifacts are produced by factors other than human anatomy, including
environmental artifacts, recorder-induced artifacts, cable, and electrode-induced arti-
facts. In order to obtain important information to be used in the analysis of the
signal, the effect of noises, and artifacts should be reduced or eliminated by different
methods according to their source or characteristics, including Kalman filtering,
median filtering Butterworth filter [12–15].
Biomedical diagnostic systems contain enormous data for analysis of signal
processing. Time–frequency analysis methods allow the capture of hidden features
that cannot be detected visually and numerically. The morphological features of
the signals are extracted using mathematical functions on both the time domain
and frequency domain, by carrying information from the time domain form to the
frequency domain, or vice ver, using Fourier transform, wavelet transform, Hilbert
transform, and periodicity transform [16–21].
At the feature extraction stage, it is ensured that the features that can represent that
data are obtained from the raw data. In the feature selection phase, effective feature
values are obtained from the existing feature set by using various feature selection
algorithms. In this way, effective features are detected and the size of the feature
vector is reduced. These features can be obtained both using the entire time series at
the same time and splitting into appropriate window widths and each window width
processes the data recorded at a certain time separately. The time and frequency
domain features are extracted using statistical methods, energy, and entropy. As a
morphological feature vector, features are obtained using common spatial patterns,
local binary patterns as well as statistical time and frequency domain, and intrinsic
energy features including variance, number of zero crossings, mean, median, mode,
waveform length, root mean square, average absolute value, average frequency,
median frequency, peak frequency, average power, total power, sample entropy,
Shannon entropy, fuzzy entropy, permutation entropy, spectral entropy, Kolmogorov-
Sinai entropy, approximate entropy, Renyi’s Entropy, permutation entropy, Tsalli’s
entropy, wavelet entropy and Phase Entropy (PhEn), distribution entropy, Kraskov
entropy (Millan et al. 2018; Kang et al. 2018; Xiaolin et al. 2018) [19, 22–25].
24 H. Kulan and E. Özer

As well as the feature extraction methods, different feature selection algorithms

are performed. The purpose of feature selection is to reduce the number of features
by selecting the most qualified features for the related problem within the feature set
under consideration. Thus, feature selection helps to reduce the number of dimen-
sions. Some feature selection methods are principal component analysis (PCA), inde-
pendent component analyses (ICA), mutual information, particle swarm optimization
(PSO), and genetic algorithms [26–30].
In the modeling of nonlinear systems, other than the determination of the source
of uncertainty in the data or related physical events, another factor we encounter is
which techniques should be used to model depending on the problem of interest. The
emergence of many situations such as class imbalance, class noise, outlier observa-
tion, and irrelevant features cause a decrease in model reliability and performance.
In this context, the hybridization of artificial intelligence methods with appropriate
signal transformations for the solution of different problems in the field of health
allows flexible modeling and more effective solutions to be obtained.
Machine learning is a subset of artificial intelligence, to solve different types
of problems, which are classification, clustering, and regression. Classification is a
method used to categorize data into predetermined classes, by assigning data points to
specific categories or classes. Clustering is a method that aims to form homogeneous
groups by bringing together data points with similar characteristics, by discovering
hidden structures and relationships in the data set. Regression is a method used to
understand the relationship between variables and to estimate the value of a dependent
variable, by determining the dependence of a dependent variable on independent
variables and to predict future values. Some machine learning methods are support
vector machine, decision trees, random forest, ensemble learning, long short-term
memory, bidirectional long short-term memory, recurrent neural network, logistic
regression, k-mean clustering, the naïve-Bayes classifier [16, 19, 31–36].
The effectiveness of the solutions obtained from the models determined by the
use of model selection criteria should be determined by reliability analysis. For the
methods used for classification, clustering, and regression problems, model selection
criteria should be used to determine the best model, and reliability and effective-
ness analyses should not be neglected. The artificial intelligence techniques used
increase the complexity of the predicted models depending on both the number of
inputs and the hyperparameters of the models. Effective approaches to control the
complexity of models are feature selection, cross-validation, early stopping, training
with noise, mixtures of networks, branching and pruning techniques (Tibshirani
1994) [16, 19, 30, 33, 35–38].

2.3 Health Image Data Evaluation

For analysis image data, different algorithms have convolutional neural network
(CNN) structures, such as You Only Look Once (YOLO), RESNET, VGGNET,
GOOGLENET(INCEPTION), Dense Convolutional Network (DenseNet). [39–42].
Data Science in the Field of Health 25

Imaged data must be preprocessed before the analysis. The preprocessed step
contains standardization of images, adjustment of morphological conversions and
pixel luminance conversions, normalization, taking average and standard deviation
of input data and data augmentation.
Standardization scales images to the same size. Morphological conversions and
pixel luminance conversions include thresholding, erosion and dilation, opening
and closing steps. At thresholding step, all pixels with intensities above a certain
threshold are taken and transformed. Pixels whose value is less than the threshold
are converted to zero. This process results in binary image. In erosion process, bright
areas are reduced and dark areas are enlarged. Dilation is the opposite, dark areas
are reduced and bright areas are magnified. Opening step can remove small bright
spots, connect small dark cracks. Closing process removes small dark spots in the
image and connects small shiny cracks.
Normalization is one of the most important step in the preprocessing part. Pixel
density are normalized so that their values fall within a limited range. By taking the
mean and standard deviation of the input data, the average values for each pixel in all
training samples are obtained. This process helps to get an idea of the basic structures
in the image. Data augmentation is to increase data diversity without collecting new
data and making a change. Especially if the amount of data we have is not enough to
perform the classification task well enough, we should perform data augmentation.
It should only be done on training data.
Convolutional learning, especially image data based on deep learning algorithms
allows operations to be performed on it. By making transformations on the image
data, the data is vectorized as shown in many figures. For this, some special convolu-
tion layers are used. Convolutional neural network architecture includes three layers
in addition to the input and output layer. The first is the convolution layer which
come after input layer. This layer is followed by the sampling layer and end with
the full connection layer. Convolution is a conversion process which is filter-assisted
moving of each moment of an image in a slider window. After convolution layer, a
nonlinear function is used to check output of each layer. The purpose of the pooling
layer is to reduce the size and resolution of the inputs from the convolution layer.
Thus, the number of parameters in the network and the amount of computation can be
reduced. After sampling of the convolutional network, the full link layer is activated.
The output of the last convolution layer is flattened and converted to a vector here.
There are different algorithms which are used CNN structures. One of this algo-
rithm is VGGNet. It consists of 16 convolutional layers and has a very smooth
architecture. It is the most preferred architecture for feature extraction from images.
It contains 3 × 3 convolution layer and many filters. VGGNet network consists of
138 million parameters.
ResNet architecture is called residual neural network. It has algorithm which
transfer One ResNet module to another module. These transferred links together
also known as ported gates or coupled repeating units. Although it has 152 layers, it
has a lower complexity than VGGNet [41].
The GoogleNet architecture differs from other architectures by adding a new
element called the starter module. Stack normalization, image distortion and
26 H. Kulan and E. Özer

RMSprop optimization that is designated for neural networks are used in this module.
This architecture uses very small size and large quantity convolution layer and greatly
reduces the number of parameters.
In DenseNet architecture, which is called dense network, each layer is an addi-
tional layer from all previous layers. It transfers its own feature maps to all subsequent
layers. Each layer receives collective information from all layers. Because each layer
receives feature maps from all previous layers, the network can be thinner and more
compact. Thus, it has higher computation efficiency and memory efficiency.
YOLO algorithm can do object tracking quickly and it is an effective algorithm
in terms of accuracy. The reason why the YOLO algorithm is so fast is by passing
the image through the neural network at one time, the class and coordinates of all
the objects in the image can be guessed. The YOLO architecture divides the image
into regions, then divides the objects in these regions and draws the boxes called
bounding boxes. After that, the probability of the object being found in each region
and the trust score are calculated.

3 Conclusion

The average human lifespan is increasing across the world. The importance of big
data analytics application in healthcare is increasing day by day. Based on different
format of health data, the analytics techniques may vary. In this study, the critical
points in the analysis of different formatted health data are emphasized. In order to
gain robust and accurate prediction or prevention of healthcare situations, all types
of health data must be analyzed together. After understanding of advantages and
pitfalls, in the future, it will be very useful to develop the big software platforms
which contribute to evaluation of different data type together.

References

1. Abdi, H., Williams, L.J.: Normalizing data. Encycl. Res. Des. 1 (2010)
2. Kulan, H., Dag, T.: In silico identification of critical proteins associated with learning process
and immune system for Down syndrome. Plos One. 14 (2019)
3. Natekin, A., Knoll, A.: Gradient boosting machines, a tutorial. Front. Neurorobot. 7, 21 (2013)
4. Chicco, D.: Ten quick tips for machine learning in computational biology. BioData mining
10(1), 35 (2017)
5. Alpaydin, E. (2010). Introduction to machine learning. MIT press.
6. Hotelling, H.: Analysis of a complex of statistical variables into principal components. J. Educ.
Psychol. 24, 498–520 (1933)
7. Leznik, M., Tofallis, C.: Estimating invariant principal components using diagonal regression
(2005)
8. Haykin, S., Veen, B.V.: Signals and Systems. John Willey & Sons. Inc., New York (1999)
9. Piskorowski, J.: Time-efficient removal of power-line noise from EMG signals using IIR notch
filters with non-zero initial conditions. Biocybern. Biomed. Eng. 33(3), 171–178 (2013)
Data Science in the Field of Health 27

10. Sreelekha, K.R., Bindiya, T.S.: Design of cost effective variable bandwidth 2D low-pass, high-
pass and band-pass filters with improved circularity. Digital Signal Process. 133, 103842 (2023)
11. Shortlii, E.H., Cimino, J.J.: Biomedical Informatics Computer Applications in Health Care and
Biomedicine, (Third Edition). Springer (2006)
12. Ahmad, M., Jung, L.T., Bhuiyan, A.A.: A biological inspired fuzzy adaptive window median
filter (FAWMF) for enhancing DNA signal processing. Comput. Methods Progr. Biomed. 149,
11–17 (2017)
13. Bougerol, P.: Kalman filtering with random coefficients and contractions. SIAM J. Control.
Optim. 31(4), 942–959 (1993)
14. Mutanen, T.P., Metsomaa, J., Makkonen, M., Varone, G., Marzetti, L., Ilmoniemi, R.J.: Source-
based artifact-rejection techniques for TMS–EEG. J. Neurosci. Methods 382, 109693 (2022)
15. Robertson, D.G.E., Dowling, J.J.: Design and responses of Butterworth and critically damped
digital filters. J. Electromyogr. Kinesiol. 13(6), 569–573 (2003)
16. Kocadagli, O., Ozer, E., Batista, A.G.: Preictal phase detection on EEG signals using hybridized
machine learning classifiers with a novel feature selection procedure based GAs and ICOMP.
Expert Syst. Appl. 212, 118825 (2023)
17. Louis, A.K., Maass, D., Rieder, A.: Wavelets: Theory and Applications. John Wiley & Sons
Ltd (1997)
18. Mallat, S.G.: A theory for multiresolution signal decomposition: the wavelet representation.
IEEE Trans. Pattern Anal. Mach. Intell. 11(7), 674–693 (1989)
19. Raghu, S., Sriraam, N., Temel, Y., Rao, S.V., Hegde, A.S., Kubben, P.L.: Performance evaluation
of DWT based sigmoid entropy in time and frequency domains for automated detection of
epileptic seizures using SVM classifier. Comput. Biol. Med. 110, 127–143 (2019)
20. Sethares, W.A., Staley, T.W.: Periodicity transforms. IEEE Trans. Signal Process. 47(11), 2953–
2964 (1999)
21. Szcz˛esna, A., Augustyn, D.R., Josiński, H., Har˛eżlak, K., Świtoński, A., Kasprowski, P.:
Chaotic biomedical time signal analysis via wavelet scattering transform. J. Comput. Sci.
72, 102080 (2023)
22. Zhang, T., Chen, W., Li, M.: Fuzzy distribution entropy and its application in automated seizure
detection technique. Biomed. Signal Process. Control 39, 360–377 (2018)
23. Luo, H., Qiu, T., Liu, C., Huang, P.: Research on fatigue driving detection using forehead EEG
based on adaptive multi-scale entropy. Biomed. Signal Process. Control 51, 50–58 (2019)
24. Luo, T.J.: Parallel genetic algorithm based common spatial patterns selection on time–frequency
decomposed EEG signals for motor imagery brain-computer interface. Biomed. Signal Process.
Control 80, 104397 (2023)
25. Ozer, E., Kocadagli, O., Batista, A.G.: Time-frequency analysis of the EEG signals: visual
identification of epileptic patterns. In: y-BIS 2019 Conference Book: Recent Advances in Data
Science and Business Analytics (2019). ISBN- 978–605–5005–95–5
26. Buzzell, G.A., Niu, Y., Aviyente, S., Bernat, E.: A practical introduction to EEG time-frequency
principal components analysis (TF-PCA). Dev. Cogn. Neurosci. 55, 101114 (2022)
27. Cui, D., Pu, W., Liu, J., Bian, Z., Li, Q., Wang, L., Gu, G.: A new EEG synchronization strength
analysis method: S-estimator based normalized weighted-permutation mutual information.
Neural Netw. 82, 30–38 (2016)
28. Judith, A.M., Priya, S.B., Mahendran, R.K.: Artifact removal from EEG signals using regen-
erative multi-dimensional singular value decomposition and independent component analysis.
Biomed. Signal Process. Control 74, 103452 (2022)
29. Li, H., Guo, W., Zhang, R., Xiu, C.: Variable length particle swarm optimization and multi-
feature deep fusion for motor imagery EEG classification. Biochem. Biophys. Res. Commun.
571, 131–136 (2021)
30. James, G., Witten, D., Hastie, T., Tibshirani, I.R.: An introduction to statistical learning with
applications in R (2017)
31. Cortes, C., Vapnik, V.: Support-vector networks. Mach. Learn. 20, 273–297 (1995)
32. Efron, B., Tibshirani, R.J.: An introduction to the bootstrap (Chapman & Hall/CRC monographs
on statistics & applied probability). Chapman and Hall/CRC (1994)
28 H. Kulan and E. Özer

33. Bishop, C.M., Nasrabadi, N.M.: Pattern recognition and machine learning, vol. 4, no. 4, p. 738.
Springer, New York (2006)
34. Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural Comput. 9(8), 1735–1780
(1997)
35. Tsiouris, KM, Pezoulas, V.C., Zervakis, M., Konitsiotis, S., Koutsouris, D.D., Fotiadis, D.I.:
A long short-term memory deep learning network for the prediction of epileptic seizures using
EEG signals. Comput. Biol. Med. 99, 24–37 (2018)
36. Unanyan, N.N., Belov, A.A.: Design of upper limb prosthesis using real-time motion detection
method based on EMG signal processing. Biomed. Signal Process. Control 70, 103062 (2021)
37. Aggarwal, C.C.: Neural networks and deep learning. Springer 10(978), 3 (2018)
38. Li, M., Chen, W., Zhang, T.: Classification of epilepsy EEG signals using DWT-based envelope
analysis and neural network ensemble. Biomed. Signal Process. Control 31, 357–365 (2017)
39. Sagar, A.: Convolutional neural network for breast cancer classification: deep learning for
solving the most commonly diagnosed cancer in women. Towards Data Sci. Medium Publ.
Shar. Concept. Ideas Codes
40. Hamed, G.: YOLO based breast masses detection and classification in full-field digital mammo-
grams. Comput. Methods Progr. Biomed. 200 (2020). https://doi.org/10.1016/j.cmpb.2020.
105823
41. Yılmaz, A., Kaya, U.: Derin Öğrenme. Kodlab, İstanbul (2021)
42. Özkan, Y.: Uygulamalı Derin Öğrenme. Papatya Yayıncılık, İstanbul (2021)
43. Hotelling, H.: Relations between two sets of variates. In Breakthroughs in Statistics:
Methodology and Distribution, pp. 162–190. Springer New York, New York, NY (1992)
44. Mason, L., Baxter, J., Bartlett, P., Frean, M.: Boosting algorithms as gradient descent. Adv.
Neural Inf. Process. Syst. 12 (1999)
45. Lotter, W., Diab, A.R., Haslam, B., Kim, J.G., Grisot, G., Wu, E. Gregory Sorensen, A.:
Robust breast cancer detection in mammography and digital breast tomosynthesis using an
annotation-efficient deep learning approach. Nat. Med. 27(2), 244–249 (2021)
46. Millán, P.C., García-Ferrero, M.Á., Llanes-Estrada, F.J., Riojano, A.P., García, E.M.S.:
Shannon entropy and particle decays. Nucl. Phys. B 930, 583–596 (2018)
47. Kulan, H., Dag, T., : Using machine learning classifiers to identify the critical proteins in down
syndrome. Proceedings of the 2018 2nd International Conference on Computational Biology
and Bioinformatics. 51–54 (2018)
48. Mahmoodirad, A., Allahviranloo, T., Niroomand, S.: A new effective solution method for fully
intuitionistic fuzzy transportation problem. Soft. Comput. 23(12), 4521–4530 (2019)
49. Moloudzadeh, S., Allahviranloo, T., Darabi, P.: A new method for solving an arbitrary fully
fuzzy linear system. Soft. Comput. 17(9), 1725–1731 (2013)
50. Rahmani, A., Lotfi, F.H., Rostamy-Malkhalifeh, M., Allahviranloo, T.: A new method for
defuzzification and ranking of fuzzy numbers based on the statistical beta distribution. Adv.
Fuzzy Syst. 2016, 1–8 (2016)
Evaluation of Hospitals and Health Care
Centers with Ratio Data

Mehdi Soltanifar

Abstract One of the most important topics in healthcare management is the perfor-
mance evaluation of hospitals and healthcare centers. This work is done with different
models of decision analysis and after determining the evaluation criteria. One of the
most up-to-date and widely used ways to evaluate performance is the use of non-
parametric models such as Data Envelopment Analysis (DEA). In many evaluations,
after determining the criteria and specifying the inputs and outputs for using DEA
models, we are faced with ratio data. Traditional DEA models are not suitable models
for handling this type of data, and it is necessary to use DEA models to handle this
type of data. In this research, after examining some ratio criteria for evaluating public
hospitals and health centers, DEA-R models are presented to handle these data both
in cases of non-negative data and negative data.

Keywords Data envelopment analysis (DEA) · Ratio data · DEA-R models ·

Negative data · Public hospitals · Health care centres

1 Introduction and Motivation

Public health is actually the science of protecting and improving the health of people
and their living places. Improving the level of public health is achieved by promoting
a healthy lifestyle, research on the prevention of diseases and injuries and their
timely diagnosis, prevention and treatment of infectious diseases. One of the tools to
improve the quality of public health is public health centers and public hospitals that
provide good health services. Health centers are community-based, patient-centered
organizations that provide affordable, accessible, and high-quality primary health
care services to individuals and families, including those experiencing homeless-
ness, living in poverty, residents of public housing, the disabled, and Veterans offer.
Public hospitals (public or government-supervised hospitals) provide free or low-cost

M. Soltanifar (B)
Department of Mathematics, Semnan Branch, Islamic Azad University, Semnan, Iran
e-mail: [email protected]; [email protected]

© The Author(s), under exclusive license to Springer Nature Switzerland AG 2024 29

T. Allahviranloo et al. (eds.), Decision Making in Healthcare Systems,
Studies in Systems, Decision and Control 513,
https://doi.org/10.1007/978-3-031-46735-6_4
30 M. Soltanifar

health care to every citizen. These hospitals are usually run with government funds.
However, there are also government-funded and government-supervised hospitals.
Therefore, continuous monitoring of health centers and public hospitals is essential
and one of the duties of the government. This monitoring requires the evaluation and
ranking of public hospitals and health centers. In order to achieve this, it is neces-
sary to determine evaluation criteria and indicators and then use decision analysis
models. So far, researchers have presented and used many methods to evaluate public
hospitals and health care centers. Many of these methods can be found in [1–3]. In
this research, the focus is on the methods of decision analysis, which uses mathemat-
ical models and after determining the indicators and evaluation criteria, calculates
the center’s efficiency score or ranks them. Among these methods, non-parametric
methods such as Data Envelopment Analysis can be mentioned.
Data Envelopment Analysis (DEA) is a technique based on linear programming to
evaluate the performance of a set of homogeneous Decision-Making Units (DMUs)
[4]. DEA is a suitable and efficient tool in the field of measuring and evaluating
productivity, which is used as a non-parametric method to calculate the efficiency
of DMUs. Today, the use of DEA technique is expanding rapidly and is used in the
evaluation of various organizations and industries such as the banking industry, post
office, hospitals, educational centers, power plants, refineries, etc. [5–18]. There have
been many developments in theoretical and practical aspects in DEA models, which
makes it indispensable to know its various aspects for more precise application [19].
The many advantages of the DEA technique, such as insensitivity to the units of
measurement of inputs and outputs, calculating the relative efficiency of units, high
generalizability and expansion, the possibility of hybridizing it with other decision-
making methods such as Multi-Criteria Decision-Making (MCDM) methods and
the existence of efficient methods for ranking the units prompted researchers to use
this technique in healthcare management widely. Linna [20] measured hospital cost
efficiency with panel data models. Banker et al. [21] conducted an illustrative study
of hospital production using DEA and translog methods. Stefko et al. [22] applied
the DEA technique to evaluate the efficiency of health care in the Slovak Republic.
A network-DEA model to evaluate the impact of quality and access on hospital
performance was presented by Afonso et al. [23]. Chiu et al. [24] used the DEA
model to study the performance of hospitals through medical quality. Antunes et al.
[25] evaluated performance and synergy in Chinese health care and showed that
synergy has played a pivotal role in China’s health care systems. Miszczynska and
Miszczyński [26] presented a window-DEA evaluation to measure the efficiency of
the healthcare sector in Poland. Hollingsworth [27] applied DEA policy to measure
the efficiency and productivity of health care delivery. Kohl et al. [28] reviewed 262
papers on DEA applications in healthcare with a particular focus on hospitals. Arya
and Yadav [29] applied DEA to assess the health sector by considering the logic of
uncertainty. Recently, research trends from 2017 to 2022 have been presented by
Jung et al. [30] regarding efficiency measurement using DEA in public health care.
Recent research points out that most mathematical models like dynamic systems
and also linear systems get involved with real-world problems. Since their related
information has different forms like certainty and uncertainty, the uncertain version of
Evaluation of Hospitals and Health Care Centers with Ratio Data 31

these models does have more importance in the applications. In health problems, two
types of mathematical models; fuzzy dynamic systems and fuzzy linear programming
problems even transportation problems, play an important role, on the other hand, the
main and basic model of fuzzy linear programming problems is fuzzy linear systems.
In conclusion, fuzzy differential equations (as a special version of dynamic
systems) [31, 32], and fuzzy linear systems (as a basic model of fuzzy linear
programming problems) [33], have an important role in this research. Recently
several research has been done on investigating the advanced version of the uncertain
information [31, 34, 35], and moreover the above-mentioned basic models.
One of the techniques for decision making is Data Envelopment Analysis (DEA),
and many researchers have conducted research on this topic. Several recent papers
have been cited to mention their research area [4, 36, 37, 38]. Some authors have
explored the topic in a different way: safety analysis and reliability [39, 40, 41].
In the studies carried out for the evaluation of public hospitals and health centers,
first the effective evaluation criteria are extracted in the form of inputs and outputs,
and then DEA models are used for evaluation, calculation of relative efficiency,
ranking, etc. But in many cases, after determining the criteria and specifying the
inputs and outputs, we are faced with ratio data, which traditional DEA models
are not suitable models for evaluating this type of data. In this research, we intend
to study the evaluation criteria of public hospitals and health centers, which are
provided in the form of ratio data, and then use the appropriate DEA models, known
as DEA-R models, to handle these data. Therefore, the continuation of the chapter
is organized as follows. In Sect. 2, the introductions related to DEA models that
are able to handle ratio data are presented. In Sect. 3, the motivation for presenting
this research is presented by examining relative criteria in the evaluation of public
hospitals and health centers. In Sect. 4, some management perspectives are presented
with the healthcare management aspect, and finally, the conclusion of this chapter is
presented in Sect. 5.

2 Literature Review

In the study of many organizations, especially public hospitals and health centers,
inputs and outputs are presented in a ratio form. Therefore, it is necessary to provide
DEA models to handle this form of data. In this section, ratio-based DEA (DEA-R)
models are discussed in the presence of non-negative data as well as negative data.

2.1 Non-negative Data

Consider m-dimension positive input vector of (x1 j , x2 j , ..., xm j ) > 0 for DMUj ,
1 ≤ j ≤ n that is used for producing s-dimension positive output vector of
32 M. Soltanifar

(y1 j , y2 j , ..., ys j ) > 0. Despić et al. [42] introduced the following models for evalu-
ating DMUp assuming Constant Returns to Scale (CRS), called the DEA-R max–min
efficiency [43].
 
 xi j yr p
e p =   max min wri (1)
wri =1,wri ≥0 j
r i
xi p yr j
r i

1
e p =   max min   (2)
wri =1,wri ≥0 j  y x
r i wri yrr pj xii pj
r i

Assuming Variable Returns to Scale (VRS), the DEA-R max–min models for
DMUp can be presented as (3) and (4).

1
ep =  max
 min   (3)
wi0 + wri =1,wri ≥0 j  x  y xi p
i r i wi0 xii pj + wri yrr pj xi j
i r i
 
 yr p   xi j yr p
ep =  max
 min wr 0 + wri (4)
wr 0 + wri =1,wri ≥0 j
r
yr j r i
xi p yr j
r r i

Models (5) and (6) are linear programming models obtained from models (3) and
(4), respectively.

e p = min ϕ
s.t.
 yr j 
 xi p   xi j
wi0 + wri yr p ≤ ϕ, ∀ j
x ij r xi p
i
 
i (5)
wi0 + wri = 1
i r i
wri ≥ 0, ∀r, ∀i
wi0 is free, ∀i.
Evaluation of Hospitals and Health Care Centers with Ratio Data 33

e p = max ϕ
s.t.
 xi j 
 yr p   yr j
wr 0 + wri xi p ≥ ϕ, ∀ j
r
yr j r yr p
 
i (6)
wr 0 + wri = 1
r r i
wri ≥ 0, ∀r, ∀i
wr 0 is free, ∀r,

Models (5) and (6) are in the multiplier form. By using the rules of duality, the
corresponding envelopment form models, can be presented as models (7) and (8),
respectively.

e p = max θ
s.t.
 yr j 
 xi j
λ j yr p ≥ θ, ∀r, ∀i
j xi p
  
xi p
λj = θ, ∀i (7)
j
xi j

λj = 1
j

λ j ≥ 0, ∀ j
θ is free.

e p = min θ
s.t.
 xi j 
 yr j
λj xi p ≤ θ, ∀r, ∀i
j yr p
  
yr p
λj = θ, ∀r (8)
j
yr j

λj = 1
j

λ j ≥ 0, ∀ j
θ is free.

The above models were studied by Mozaffari et al. [44, Olesen et al. 45] and
Mozaffari et al. [46] and are able to handle DEA policy for ratio data. These models
34 M. Soltanifar

can only be used for non-negative data, and this is a limitation for their application
in various problems. In the next section, DEA-R models for handling negative data
are presented.

2.2 Negative Data

In the evaluation of many organizations and units, the possibility of negative data
is a probable hypothesis. The DEA-R models mentioned in the previous section are
not able to handle negative data. Therefore, in the continuation of the models of the
previous part, models are rewritten in such a way that they are able to handle negative
data. The models are presented for two cases of negative outputs or negative inputs.
It should be noted that the simultaneous negative assumption of inputs and outputs
is not a correct assumption due to the structure of the models.
First suppose the outputs of the problem are divided into two categories. The
first category of outputs that are positive, and the second category of outputs can be
negative (Yr ). Variables YrP and YrN are defined as follows [47, 48]:
 
yr j i f yr j ≥ 0 −yr j i f yr j < 0
yrPj = yrNj = , ∀j
0 Other wise 0 Other wise

In fact, yr j = yrPj − yrNj and yrPj , yrNj ≥ 0. DEA-R input-oriented in the presence
of negative outputs is presented in the model (9).

max θ
s.t.
 yr j yr p
λj ≥θ , ∀i, ∀r ∈ O
j
xi j xi p
 yrPj yrPp 
λj ≥θ , ∀i, ∀r ∈ O
j
xi j xi p
 yrNj yrNp 
λj ≤θ
, ∀i, ∀r ∈ O (9)
j
xi j xi p
  xi p 
λj = θ, ∀i
j
xi j

λj = 1
j

λ j ≥ 0, ∀ j
θ is free.
Evaluation of Hospitals and Health Care Centers with Ratio Data 35

where the index set associated with the variables are specified with O and the index

set associated with the variables YP are specified with O . By presenting this model,
Soltanifar et al. [16] showed that their model is always feasible.
In a similar way, Soltanifar et al. [16] assumed that inputs are divided into two
categories. A positive group and another group that can be negative. Note that the
second category can be positive or negative. The following variables can be defined
for second category inputs.
 
xi j i f xi j ≥ 0 −xi j i f xi j < 0
xiPj = xiNj = , ∀j
0 Other wise 0 Other wise

In fact, xi j = xiPj − xiNj and xiPj , xiNj ≥ 0. DEA-R output-oriented in the presence
of negative inputs is presented in model (10).

min θ
s.t.
 xi j xi p
λj ≤θ , ∀r, ∀i ∈ I 
j
yr j yr p
 xiPj xiPp 
λj ≤θ , ∀r, ∀i ∈ I
j
yr j yr p
 xiNj xiNp 
λj ≥θ
, ∀r, ∀i ∈ I (10)
j
yr j yr p
  yr p 
λj = θ, ∀r
j
yr j

λj = 1
j

λ j ≥ 0, ∀ j
θ is free.

where the index set associated with the variables IP are specified with I  and the

index set associated with the variables IP are specified with I . By presenting this
model, Soltanifar et al. [16] proved its feasibility.
Soltanifar et al. [16] also rewrote their DEA-R models with the modifications
made by Kaffash et al. [49] on the models presented by Emrouznejad et al. [47, 48].
This rewriting is presented in models (11) and (12).
36 M. Soltanifar

max θ
s.t.
 yr j yr p
λj ≥ (1 + θ ) , ∀i, ∀r ∈ O
j
x i j x ip

 yrPj yrPp 
λj ≥ (1 + θ ) , ∀i, ∀r ∈ O
j
xi j xi p
 yrNj yrNp 
λj ≤ (1 − θ ) , ∀i, ∀r ∈ O (11)
j
xi j xi p
  
xi p
λj = (1 + θ ), ∀i
j
xi j

λj = 1
j

λ j ≥ 0, ∀ j
θ is free.

min θ
s.t.
 xi j xi p
λj ≤ (1 + θ ) , ∀r, ∀i ∈ I 
j
yr j yr p
 xiPj xiPp 
λj ≤ (1 + θ ) , ∀r, ∀i ∈ I
j
yr j yr p
 xiNj xiNp 
λj ≥ (1 − θ ) , ∀r, ∀i ∈ I (12)
j
yr j yr p
  
yr p
λj = (1 + θ ), ∀r
j
yr j

λj = 1
j

λ j ≥ 0, ∀ j
θ is free.

The feasibility of these models was also proved by Soltanifar et al. [16]. In the
next section, the motivation and necessity of using these models in the evaluation of
public hospitals and health centers is presented.
Evaluation of Hospitals and Health Care Centers with Ratio Data 37

3 Ratio Data in Healthcare Management and Motivation

to Use DEA-R Models

In most of the performance evaluation and ranking methods, it is necessary to first

determine the criteria affecting the quality and quantity of the services provided,
and then determine the efficiency of each unit in each criterion. The criteria can be
profit or cost type or presented in qualitative or quantitative form. Data related to
these criteria can also be collected in different forms: in the form of a tree structure
and pairwise comparison matrices, in the form of a decision matrix, in the form of
an input/output table, etc. In fact, depending on what model is going to be used to
evaluate the performance, the data collection form is different. To use non-parametric
models such as DEA models, data are set in the form of input and output tables. Basic
DEA models can be used for non-negative data, but these models are not suitable for
ratio or negative ratio data [47, 48, 50]. In such cases, the DEA-R models presented in
the previous section should be used. In this section, first, the criteria after presenting
and constructing the inputs and outputs are presented in the form of non-negative
ratio data and show the necessity of using DEA-R models for non-negative data.
Then the case where ratio criteria are based on healthcare standards is discussed. In
this case, ratio criterion is provided that can be negative and this shows the necessity
of using DEA-R models for negative data.
Ghiyasi et al. [51] used the criteria of Table 1 in the evaluation of public hospitals.
They intended to use DEA models to evaluate public hospitals. Therefore, they
presented the input/output table in the form of Table 2.
Therefore, they faced ratio inputs and outputs, which must be handled by DEA-R
models.
[13] used the standard criteria of the World Health Organization to evaluate public
hospitals and health centers, which were presented in a radio format (Table 3). “Bed
occupancy rate risk” is a criterion that can have a negative value. Therefore, they
inevitably used DEA-R models that have the ability to handle negative data for
evaluation. In this way, the motivation and necessity of using the models presented in
the previous section for the management of healthcare services is clear, and therefore,
while using the many advantages of the DEA policy, the evaluation can be done
considering the type of data.
Further managerial implications and applications regarding the use of DEA policy
in healthcare management are presented in the next section.

4 Further Managerial Implications and Applications

In many real-world problems, especially those related to health care management,

decision makers are faced with data on evaluation criteria that are only available in
the form of ratios or converted to ratio form after conversion to inputs and outputs.
This may have various reasons, including the confidentiality of criteria values in a
38 M. Soltanifar

Table 1 Criteria for assessing hospitals [51]

Criterion Criterion title Definition
number
C1 Number of physicians Total number of medical doctors (physicians) in the
hospital
C2 Number of surgeons Total number of surgical specialists in the hospital
(surgery is an invasive technique with the fundamental
principle of physical intervention on organs/organ
systems/tissues for diagnostic or therapeutic reasons)
C3 Number of emergency A physician who directs emergency medical technicians
physicians in the emergency department and focuses on immediate
decision-making and necessary actions to prevent death
or any further disability in pre-hospital settings
C4 Number of nurses Persons who have completed a program of basic nursing
education and are qualified and registered or authorized to
provide responsible and competent service for the
promotion of health, prevention of illness, care of the
sick, and rehabilitation, and are actually working in the
hospital. Nursing personnel includes professional nurses,
auxiliary nurses, enrolled nurses and related occupations
such as dental nurses and primary care nurses
C5 Number of hospital The number of hospital beds available in public and
beds private hospitals. Hospital beds are regularly maintained
and staffed for the accommodation and full-time care of a
succession of inpatients and situated in the wards or a part
of a hospital where continuous medical care for inpatients
is provided. The total number of such beds constitutes the
normally available bed complement of the hospital. Cribs
and bassinets maintained for use by healthy newborn
babies who do not require special care are not included
C6 Number of active Hospital beds that are actively available
hospital beds
C7 Number of special The number of beds regularly maintained and staffed for
hospital beds the accommodation and full-time care of a succession of
inpatients and which are situated in the wards or a part of
the hospital where continuous medical care for inpatients
is provided. The total number of such beds constitutes the
normally available bed complement of the hospital. Cribs
and bassinets maintained for use by healthy newborn
babies who do not require special care are not included
C8 Number of emergency Patients admitted to the hospital when admission is
admission unpredictable and at short notice because of clinical needs
C9 Number of hospital An operating room, also called a surgery center, is the unit
operating rooms of a hospital where surgical procedures are performed
C10 Number of outpatients The number of patients who are not hospitalized
overnight but visit a hospital, clinic or associated facility
for diagnosis or treatment
(continued)
Evaluation of Hospitals and Health Care Centers with Ratio Data 39

Table 1 (continued)
Criterion Criterion title Definition
number
C11 Number of The number of patients who are hospitalized overnight
hospitalized patients
C12 Number of surgical Total number of surgical operations (surgery is an
operations invasive technique with the fundamental principle of a
physical intervention on organs/organ systems/tissues for
diagnostic or therapeutic reasons)

Table 2 Inputs and outputs

Inputs/outputs Calculation formula
for evaluating hospitals [51]
Input 1 C1
C5
Input 2 C4
C5
Input 3 C7
C5
Output 1 C10
C1
Output 2 C11
C6
Output 3 C12
C9
Output 4 C8
C3
Output 5 C12
C2

public hospital or the standard definitions provided in the ratio form by competent
organizations. In such cases, models capable of handling ratio data such as DEA-R
models should be used to evaluate units. Also, sometimes some criteria are supplied
with negative data and these models should have the ability to handle negative data
as well. In the study of such public hospitals or health centers, managers sometimes
face issues such as resource allocation, merging two or more centers, and similar
issues for which the use of inverse DEA-R models can be very helpful [13, 16,
51–53]. In fact, the use of the models presented in this chapter is not limited to
performance evaluation and ranking, and most DEA policy applications such as
resource allocation, merger analysis, benchmarking, progress review, and the like
are possible. Ratio data are important and very common in public hospital and health
center management literature. Any kind of evaluation and planning of such centers
is involved with such data. Due to the nature of ratio data, it is important to treat
them accurately and logically.
40 M. Soltanifar

Table 3 Criteria in the study of hospitals [13]

Criterion Title Criterion formula Type of Description
criterion
Nursing staff number
I1 Number of Total hospital beds Input This criterion is
nurses to used to supply
beds nursing staff
compared to the
total hospital beds
and is used in
planning for proper
human resource
allocation. Total
hospital beds
include curative (or
acute) care,
rehabilitative,
long-term, and other
hospital beds
Generalist physicians number
I2 Number of Total hospital beds Input General physicians
generalist are medical doctors
physicians who treat acute and
to beds chronic illnesses
and provide
preventive care and
health education to
patients of all ages.
They are
responsible for the
provision of
continuing care to
individuals and
families. The ratio
of generalist
physicians to the
total hospital beds is
a criterion for
allocating hospital
resources
(continued)
Evaluation of Hospitals and Health Care Centers with Ratio Data 41

Table 3 (continued)
Criterion Title Criterion formula Type of Description
criterion
Specialist physicians number
I3 Number of Total hospital beds Input Specialist
specialists physicians are
physicians pediatricians,
to beds anesthesiologists,
cardiologists,
dermatologists,
hematologists,
internists,
pathologists,
orthopedists,
ophthalmologists,
neurologists,
obstetricians/
gynecologists,
psychiatrists,
medical specialists,
surgical specialists,
and so on
Equipment score
I4 Equipment Total population covered by the hospital Input This criterion shows
to the access to hospital
population equipment,
covered by including Gamma
hospital Camera,
Radiotherapy,
Angiography, CT
scan, MRI, etc.
Geographical
distribution and
waiting time,
particularly for
these devices, are
essential. In other
words, this criterion
is the score of each
hospital due to the
existence or absence
of the above devices
at that hospital
(continued)
42 M. Soltanifar

Table 3 (continued)
Criterion Title Criterion formula Type of Description
criterion
Utilized bed -days
O1 Bed 85 − Available bed -days × 100 Output The bed occupancy
occupancy rate is the average
rate risk number of days
when a hospital bed
was occupied as %
of available
365 days. Bed
occupancy rate is
one of the hospital
productivity criteria.
This rate is also
used to interpret the
sources of service
providers and
guides for planning
and managing
hospital beds.
Studies have shown
that a high
occupancy rate
(usually above
85%) can indicate
bed deficiency and
is generally
associated with an
increased risk of
hospital infections
Available bed days−Utilized bed days
O2 Bed Inpatient discharges Output This criterion shows
turnover the level of
interval exploitation of the
hospital beds. Short
turnover intervals
have been linked to
increased
methicillin-resistant
Staphylococcus
aureus (MRSA)
infections
(continued)
Evaluation of Hospitals and Health Care Centers with Ratio Data 43

Table 3 (continued)
Criterion Title Criterion formula Type of Description
criterion
The number of hospital admissions
O3 Inpatient Total population covered by the hospital × 1000 Output This criterion shows
admission the number of
rate hospital admissions
per person per year.
This criterion
represents inpatient
care and utilization.
Hospital records are
the basis for
statistics on
performance related
to inpatient
activities, including
the number of beds,
admissions,
discharges, deaths,
and stay duration
Number of discharges (including deaths)
O4 Bed Total hospital beds Output The hospital bed
turnover turnover rate
rate measures the extent
of hospital
utilization. It is the
number of times
there is a change of
occupant for a bed
during a given
period

5 Conclusion

The focus of this chapter is on the use of DEA policy in health care management
and specifically the performance evaluation and ranking of public hospitals and
health centers. Where evaluation criteria are available in the form of ratio data and
sometimes, they take negative values. DEA-R models can provide researchers with
DEA policy for handling relative data. Also, these models were presented in a form
that has the ability to handle negative ratio data. Although the basic models provided
for performance evaluation and ranking are provided, but by placing these models
as a basis, it is possible to apply other management such as resource allocation,
merging, benchmarking, progress review, and the like.
44 M. Soltanifar

References

1. Diehr, P., Yanez, D., Ash, A., Hornbrook, M., Lin, D.Y.: Methods for analyzing health care
utilization and costs. Annu. Rev. Public Health 20(1), 125–144 (1999)
2. Fronczek-Munter, A.: Evaluation methods for hospital facilities. Int. J. Facil. Manag. 12, 215–
226 (2013)
3. Mbau, R., Musiega, A., Nyawira, L., Tsofa, B., Mulwa, A., Molyneux, S., Barasa, E.: Analysing
the efficiency of health systems: a systematic review of the literature. Appl. Health Econ. Health
Policy 21(2), 205–224 (2023)
4. Akram, M., Shahzadi, S., Shah, S.M.U., Allahviranloo, T.: A fully Fermatean fuzzy multi-
objective transportation model using an extended DEA technique. Granul. Comput. (2023).
https://doi.org/10.1007/s41066-023-00399-6
5. Sharafi, H., Hosseinzadeh Lotfi, F., Jahanshahloo, G., Rostamy-malkhalifeh, M., Soltanifar,
M., Razipour-GhalehJough, S.: Ranking of petrochemical companies using preferential voting
at unequal levels of voting power through data envelopment analysis. Math. Sci. 13, 287–297
(2019)
6. Sharafi, H., Soltanifar, M., Lotfi, F.H.: Selecting a green supplier utilizing the new fuzzy voting
model and the fuzzy combinative distance-based assessment method. EURO J. Decis. Process.
10, 100010 (2022)
7. Soltanifar, M.: A new interval for ranking alternatives in multi attribute decision making
problems. J. Appl. Res. Indus. Eng. (2022)
8. Soltanifar, M.: A new group voting analytical hierarchy process method. J. Operat. Res. Appl.
Appl. Math. Lahijan Azad Univ. 14(3), 1–13 (2017)
9. Soltanifar, M.: The voting linear assignment method for determining priority and weights in
solving MADM problems. J. Appl. Res. Indus. Eng. 8(Special Issue), 1–17 (2021)
10. Soltanifar, M., Farhadi, F.: An application of data envelopment analysis for measuring the
relative efficiency in banking industry. Manag. Sci. Lett. 4(5), 1021–1026 (2014)
11. Soltanifar, M., Shahghobadi, S.: Selecting a benevolent secondary goal model in data envel-
opment analysis cross-efficiency evaluation by a voting model. Socioecon. Plann. Sci. 47(1),
65–74 (2013)
12. Soltanifar, M., Sharafi, H.: Preferential voting in the presence of undesirable voters. Int. J.
Appl. Decis. Sci. 15(6), 753–766 (2022)
13. Soltanifar, M., Tavana, M., Ghiyasi M., Sharafi, H.: A new method for solving inverse data
envelopment analysis problems with negative ratio data. OR Spectrum (vol. Submitted 2023)
14. Soltanifar, M., Jahanshahloo, G.R., Lotfi, F.H., Mansourzadeh, S.M.: On efficiency in convex
hull of DMUs. Appl. Math. Model. 37(4), 2267–2278 (2013)
15. Soltanifar, M., Shahghobadi, S.: Classifying inputs and outputs in data envelopment analysis
based on TOPSIS method and a voting model. Int. J. Bus. Anal. (IJBAN) 1(2), 48–63 (2014)
16. Soltanifar, M., Hosseinzadeh Lotfi, F., Sharafi, H., Lozano, S.: Resource allocation and target
setting: a CSW–DEA based approach. Ann. Oper. Res. 318(1), 557–589 (2022)
17. Lotfi, F.H., Jahanshahloo, G.R., Ebrahimnejad, A., Soltanifar, M., Mansourzadeh, S.M.: Target
setting in the general combined-oriented CCR model using an interactive MOLP method. J.
Comput. Appl. Math. 234(1), 1–9 (2010)
18. Soltanifar, M., Lotfi, F.H.: The voting analytic hierarchy process method for discriminating
among efficient decision-making units in data envelopment analysis. Comput. Ind. Eng. 60(4),
585–592 (2011)
19. Emrouznejad, A., Yang, G.L.: A survey and analysis of the first 40 years of scholarly literature
in DEA: 1978–2016. Socioecon. Plann. Sci. 61, 4–8 (2018)
20. Linna, M.: Measuring hospital cost efficiency with panel data models. Health Econ. 7(5),
415–427 (1998)
21. Banker, R.D., Conrad, R.F., Strauss, R.P.: A comparative application of data envelopment
analysis and translog methods: an illustrative study of hospital production. Manage. Sci. 32(1),
30–44 (1986)
Random documents with unrelated
content Scribd suggests to you:
The Project Gutenberg eBook of Harper's Young People,
November 15, 1881
This ebook is for the use of anyone anywhere in the United States and most other
parts of the world at no cost and with almost no restrictions whatsoever. You may
copy it, give it away or re-use it under the terms of the Project Gutenberg License
included with this ebook or online at www.gutenberg.org. If you are not located in
the United States, you will have to check the laws of the country where you are
located before using this eBook.

Title: Harper's Young People, November 15, 1881

Author: Various

Release date: September 19, 2015 [eBook #50013]

Most recently updated: October 22, 2024

Language: English

Credits: Produced by Annie R. McGuire

*** START OF THE PROJECT GUTENBERG EBOOK HARPER'S YOUNG PEOPLE,

NOVEMBER 15, 1881 ***
 FAREWELL.
"LUCK."
THE TALKING LEAVES.
CHARLOTTE CORDAY.
HOW "THE BABY" WENT NUTTING, AND WHAT CAME OF IT.
A PRIVATE CIRCUS.
A LAWN TENNIS TOURNAMENT.
AN ASIATIC STORY.
THE MERRY WIND.
A DANGEROUS PLAYTHING.
NURSERY RHYMES.
OUR POST-OFFICE BOX.
THE PERFORMING ELEPHANT.

Published by HARPER & BROTHERS,

Vol. III.—No. 107. price four cents.
New York.
Tuesday, November 15, $1.50 per Year, in
Copyright, 1881, by Harper & Brothers.
1881. Advance.
WASHING DOLLY.
FAREWELL.
"It's growing late," said the honey-bee;
"Winter's no sort of weather for me;
I'll hurry away to the hive."
"It's growing cold," said the bustling fly;
"There's going to be plenty of snow by-and-by,
And how will a poor fly thrive?"

The cricket piped, "The season is old,

Leaves and grasses are turning to gold;
It's a queer world that changes so;
My chirp has lost its musical tones,
And the north wind bites to my very bones;
I think I had better go."

The squirrel said, "It is growing chill;

The windfalls have gone to the cider-mill;
But there's many a chestnut burr
Ready to burst at the frost's first touch.
If snow flies soon, I sha'n't mind much,
Wrapped in my thickening fur."

"The best of the year," trilled the lingering thrush,

"Has left us behind; there's a tender hush
Brooding o'er meadow and dell;
Our nests are all empty, our birdlings have flown;
There is nothing to keep us at home, I must own;
There's nothing to sing but 'Farewell.'"
 "LUCK."
"Just like his luck!" half of the boys said, when Charlie Foster won the State Scholarship.
They had made the same remark when his name had been sent in by the principal of the
school to the superintendent as his best scholar. In all likelihood these same old school-
fellows will keep on saying, "Just like his luck!" if Charlie ever becomes a Judge, or a
Senator, or if he marries happily, or makes a fortune. Every step upward is attributed by
some men and boys to that unknown quantity called "luck." And curiously enough, just
as "Like his luck" is used to account for the success of one's friends, so "Just like my
luck" is used to explain our own failures.
"It is just my luck! There was not a single question about anything I knew. I had
crammed up the capitals of the States, square root, and the conjugations, and I was
asked about mountain ranges, compound interest, and the fifth declension. I always was
unlucky!"
In all this talk about "luck" is there not a good deal of inconsistency? We never employ
the word to account for our own successes or somebody else's failures. When the said
Charlie Foster misses a catch at base-ball, or catches a crab in a race, we do not cry,
"How unlucky he is!" but, "What a muff that Charlie Foster is!" and when we ourselves
manage to get on the roll of honor, we resent with virtuous indignation any
congratulations on our luck. "Luck, indeed!" we growl; "there was no luck at all. It was
just hard work, and nothing else."
Moreover, this talk about luck is, in the first place, somewhat unmanly, not to say
cowardly. To trust to luck is a confession that one can not do anything by one's own labor
or one's own intellect. It is really, my boy, an acknowledgment that you have no
independence of character, no strength of will, no patience, and no perseverance. It is a
sure confession of carelessness and idleness. "I'll study this thing or that thing, and trust
to luck for the rest," you say, and the result is you are nowhere in the examination.
So in everything we undertake. If we neglect to take ordinary pains, if we omit ordinary
prudence, no luck ever saves us from disaster.
Trusting in luck is a very different thing from trusting in Providence. Providence aids
those who aid themselves, and just in proportion as they do their work honestly and
conscientiously. Luck is a kind of capricious spirit which is expected to set at naught all
the laws of nature for our advantage, or to our disadvantage, without the slightest
apparent reason why it should intervene at all. If there is such a thing, that can either
make or mar us, our first duty is not to be its slave, but to make ourselves its master.
We must not stand like beggars at a street corner until luck drops a few coppers into our
hats. We must be a law unto ourselves, and not mere playthings of chance. Let us be
honest enough to acknowledge our own mistakes. The grumbler who laments,
 "I never had a slice of bread,
Cut nice and smooth and long and wide,
But fell upon the sanded floor,
And always on the buttered side,"
fancies himself unlucky. If he were honest, he would blame himself for not keeping good
hold of his bread and butter, and if he thought about it, he would see that falling on the
buttered side was a natural result of the way in which he was holding it.
This notion of luck very often arises from a mixture of conceit and jealousy. We do not
like to allow that another has more talent than we have, and has used his faculties better.
He has, however, if we examine his career, been more studious, more careful, more
observant. It would be much more noble of us, instead of repeating like parrots the word
"luck," meaning thereby that he has got a reward which he does not deserve, to candidly
say, "He has deserved all he has won; he is the better fellow."
Another evil arising from this talk about luck is that at last we actually believe in it. Once
under the influence of this notion, we exercise no caution or foresight. "Luck," we say,
"will bring us through." Fortunately for our future and permanent success, luck does
nothing of the sort. In the long-run, luck is nowhere. You may have heard of games of
chance—gambling games, as they are styled—and of lotteries and the like. You have
heard of people being lucky at them. The professional gambler and lottery-keeper know
better than that; they know that even in throwing dice there is very little luck. The man
who is lucky to-day is unlucky to-morrow: it is in reality skill or trickery and not luck that
enables the professional gambler to pursue his career.
Lucky people, in fact, are people who have thoroughly trained themselves for the battle
of life. They have eyes open to perceive a coming danger, and have learned how to avoid
it; they recognize a difficulty, and know how to overcome it; they see an opportunity, and
know how to make use of it; and they are ready, with all their faculties alert, to seize it
before it has gone forever. Their success is visible to every eye, and arrests our attention
at once. What we do not see, very often what we will not see, but deliberately shut our
eyes to, is the foresight they exercise, the careful training they have undergone, the long
practice which has made them perfect.
There is nothing brilliant or showy about this practice and training, and therefore we have
not noticed them. But they are there, nevertheless. To all of us, every day of our lives,
opportunities present themselves which pass without our heeding them, or, if we see
them, without our having the courage and skill to avail ourselves of them. We let them
fly, never to return, because we are not ready, and then we cry, "Just like our luck!" As
Shakspeare says,
"The fault, dear Brutus, is not in our stars,
But in ourselves, that we are underlings."
Away with your notions of luck. Be manly, and trust to work. Do your duty, and let luck
do its worst.
 THE TALKING LEAVES.[1]
An Indian Story.

BY WILLIAM O. STODDARD.

Chapter VII.
efore Steve Harrison and his friend left the ruins of the ancient town
behind them, they had decided that they were going away from a
complete solitude—a place where even wild Indians did not very often
come.
It looked desolate enough, with its scattered inclosures of rough stone,
not one of them with any roof on, or any sign that people had lived there
for a hundred years at least. The windows in the tumbling walls had
probably never had either sash or glass in them, and the furniture, used
by the people who built the village, whatever it may have been, had long
since disappeared.
It could never have been a very large or populous town, but it could hardly at any time
have had a wilder-looking set of inhabitants than were the party of men who drew near it
at about the time when Steve and Murray were killing their cougar.
Two tilted wagons, a good deal the worse for wear, apparently pretty heavily laden, and
drawn by six mules each, were accompanied by about two dozen men on horseback.
Their portraits would have made the fortune of any picture-gallery in the world.
Everybody would have gone to look at such a collection of bearded desperadoes.
They were not Indians, nor were they dressed as such. They were attired in every
fashion except well and cleanly. If the odds and ends of several clothing stores had been
picked up after a fire, and then worn about out, and patched and mended with bits of
blankets and greasy buckskin, something like those twenty odd suits of clothes might
have been produced; that is, if the man who tried to do it could have had these for a
pattern. If not, he would have failed.
The men themselves were as much out of the common way as were the clothes they
wore, but they had somehow managed to keep their horses and mules in pretty good
condition.
Horses and mules are of more importance than clothing to men who are as far away from
tailors and civilization as were these new-comers in the neighborhood of Steve's mine.
If Steve had seen them he would probably have trembled for the "Buckhorn," for Murray
would at once have told him that these men were miners.
That was nothing against them, certainly, and they must have been daring fellows to
push their hunt for gold so far beyond any region known to such hunters.
One look at their hard, reckless faces would have convinced anybody about their
"daring." They looked as if they were ready for anything.
So they were, indeed, and it is quite probable a man of Murray's experience would have
guessed at once that they were ready for a good many other things besides mining.
Just now certainly they were thinking of something else.
"Bill," said the foremost rider to a man a little behind him, "we were wrong to leave the
trail of them army fellers. We're stuck and lost in here among the mountains."
"It looks like it. We'll hev to go into camp, and scout around till we find a pass. But it
wasn't any use follerin' the cavalry arter we found they was bound west."
"That's so. It won't do for us to come out on the Pacific slope. It's Mexico or Texas for
us."
"We'd better say Santa Fe."
"They'd make us give too close an account of ourselves there. Some of the boys might let
out somethin'."
"Guess it's Mexico, then. That isn't far away now. But I wish I knew the way down out of
this."
The ruins, strange and wonderful as they were, did not seem to excite any great degree
of curiosity among those men. They talked about them, to be sure, but in a way which
showed that they had all seen the same sort of thing before during their wild rovings
among the mountains and valleys of the great Southwest.
Just such ruins are to be found in a great many places. We do not even know how many,
and nobody has been able yet to more than guess by whom they were built or when.
Mere ravines and gorges and cañons would not do for this party. They must find a regular
"pass," down which they could manage to take their horses and mules and wagons. Even
before they halted, several of them had been looking and pointing toward what Murray
had spoken of as "the western gap."
That was the opening through the ranges which had been for a moment such a
temptation to Steve Harrison.
"It's west'ard, Bill, but it may hev to do for us."
"It may take us down, to some lower level, or it may show us a way south."
"The great Southern Pass is down hereaway somewhar."
"Further east than this. We ought to strike it, though, before we cross the border."
"Mexico ain't a country I'd choose to go inter, ef I hed my own way, but we've got to go
for it this time."
But whatever may have been their reason for seeking Mexico, they were just now a good
deal puzzled as to the precise path by means of which they might reach it. It was getting
late in the day, too, for any kind of exploration, and the mule-teams looked as if they had
done about enough.
So it came to pass that the ruined village of the forgotten people was once more
occupied.
Did they go into the houses? No. It was the man called Bill who said it, but all the rest of
them seemed to feel just as he did, when he remarked:
"Sleep in one of them things? No, I guess not—not even if it was roofed in. They were
set up too long ago to suit me."
That stamped him as an American, for there is no other people in the world that hate old
houses. No real American was ever known to use an old building of any kind a day longer
than he could help. He would as soon think of wearing old clothes just because they were
old.
The ground near the ruins was covered with fragments of stone and fallen masonry, but
there was a good camping ground between that and the trees from which Murray and
Steve had fired at the buck.
"It's the loneliest kind of a place, Captain Skinner," said Bill, just after he had helped turn
the mules loose on the grass.
"I wish I knew just how lonely it is. I kind o' smell something."
"Do ye, Cap?"
Every such band of men has its "Captain," of some kind, and sometimes very good
discipline and order is kept up. But Captain Skinner was hardly the man anybody would
have picked out for a leader, before seeing how the rest listened to what he said, and
how readily they seemed to obey him.
He was the shortest, thinnest, ugliest, and most ragged man in the whole party; and just
at this moment he did not appear to be carrying any arms except the knife and pistol in
his belt.
"If I don't smell it, I can see it. Look yonder, Bill."
"That's so!—blood!"
It was the spot on which the buck had fallen, and in a moment more than half a dozen
men were looking around in all directions.
They understood all they saw, too, as well as any Indians in the world, for in less than
five minutes Captain Skinner said: "That'll do, boys. We must follow that trail. Two white
hunters. They killed the buck. Both wore moccasins; so they ain't fresh from the
settlements. There's something queer about it. They were on foot, and they carried off
their game."
It was indeed very queer, and it would not do to let any such puzzle as that go by
unsolved. So, while several men were ordered out after game, and several more were left
to guard the camp, Captain Skinner himself, with Bill and five others, armed to the teeth,
set out at once on the trail of Murray and Steve Harrison.
It was easy enough to follow those two pairs of footprints as long as they were made in
the grass. After they got upon rocky ground, it was not so easy, and the miners did not
get ahead so fast; but they did not lose the trail for a moment. Indeed, it was about as
straight in one direction as the nature of the ground would permit.
"Two fellers out yer among these ere mountains, all by themselves," growled Bill, as they
drew near the ledge at the head of the deep cañon.
"We don't know that they're all alone yet," said Captain Skinner. "They carried that deer
somewhere."
"Right down yonder, Captain. They stopped here to rest from kerryin' of it, and I don't
blame 'em, if they'd got to tote it down through that thar cañon."
"It's a deep one, no mistake."
"Captain, look yer!" suddenly exclaimed one of the men. "We've lit on it this time."
"The ledge? I wasn't looking at that."
A perfect storm of exclamations followed from every pair of lips in the party. Such a ledge
as that they had never seen before, old mine-hunters as they were. But each one
seemed inclined to ask, just as Murray had asked of Steve, what could be done with it.
Gold enough, but nothing to get it out of the rock with, and no where to carry it to. It
was a sad problem for men who cared for nothing in the wide world but just such ledges
and just such gold. What was the use of it?
Steve Harrison never knew it, but his mine was of a good deal of use to him and Murray
just then. It kept Captain Skinner and his men looking at it long enough for them to get
nearly back to the camp of the Lipans.
"It won't do, boys," said Captain Skinner at last. "We're wasting time. Come on."
They followed him, every man turning his head as he did so to take another look at the
yellow spots that shone here and there in the quartz. Their way down the ravine was
made with care and circumspection, for they did not know at what moment they might
come in sight of "those two fellers and their deer."
It was well for them, probably, that they were cautious, for, after a good deal of steep
climbing, just as they were about to clamber down one of the rocky "stairs," the man
called Bill exclaimed, "Captain, thar it is!"
"The deer? They've left it. I see it."
"More'n that further down."
"A big-horn! And there's a painter lying beside it!"
"More'n that, Cap. They didn't give up that thar game for nothin'."
"Lay low, boys. Git to cover right away. Red-skins!"
There was no difficulty in hiding among the rocks and bowlders, and the miners were out
of sight in a moment.
They could see, though, even if they were not seen,
and they were soon able to count a dozen Indian
warriors leading three pack-ponies as far up the ravine
as four-footed beasts could go.
"Wonder if they've wiped out the two fellers," said Bill.
"Looks like it. Or they may have captured 'em. Lost
their game, if they haven't lost their scalps. Wonder
what tribe of red-skins they are, anyhow."
There was a better reason than that why No Tongue
and Yellow Head did not come back with their friends,
but it was just as well that Captain Skinner and his
miners did not understand it.
"Captain," whispered one of the men, "shall we let
drive at 'em? We could pick off half of 'em first fire."
"Not a shot. All we want jest now is to be let alone. I
don't mind killing a few red-skins."
THE LIPANS SECURING THE
"Mebbe they killed the two fellers." GAME.
"Likely as not. I'm kind o' glad they did. That there
ledge is ours now. Let 'em carry off their game, and then we'll climb back. I reckon I
know now how we'd best work our way down to the level those Indians came from."
The Lipans made short work of loading their ponies, and the moment they were out of
sight, the miners began their climb out of that cañon. There was no good reason why
they should follow the Lipans.

[to be continued.]
 CHARLOTTE CORDAY.
Charlotte Corday is remembered as the assassin of the wicked Marat. No one was ever
more cruel than Marat. He was one of the worst of the French Jacobins at Paris, who in
1793 practiced every kind of crime. They professed to be freemen, but were tyrants more
cruel than Nero. They filled Paris with murders, executions, and every kind of misery. No
one's life or property was safe, and Marat, who was now their leader, constantly urged
them to new cruelty. He seemed to the people of Paris and all the world a savage
monster who could only live amidst bloodshed and crimes, and had begun in France what
is known as the "Reign of Terror."
There lived in the country a young girl whose intended husband, it is said, had been put
to death at the suggestion of Marat. Her name was Charlotte Corday. She was about
twenty-five years old, fond of reading and study, tall and beautiful, when she resolved to
kill Marat. If she could destroy the monster, she thought she would save the republic and
revenge her lost lover. In July, 1793, Charlotte bade her father good-by in a short note,
and set out from a friend's house at Caen on her journey to Paris. She hoped to make
her way into the famous club of the Jacobins, and stab Marat in the midst of his guilty
companions.
Early on the second morning after she had reached Paris she went to the Palais Royal,
bought a knife, and drove to the house of Marat. He had been for some time unwell, and
unable to join his companions at the Jacobin Club. Charlotte was refused admittance, and
went away disappointed. She went back to her hotel, wrote a short note to Marat, telling
him that she wished to see him on business of importance to France, and once more
returned to his house. She sent up the note. Marat read it, and ordered her to be
admitted. He was in his bath; Charlotte stood alone before her victim. It was the 13th of
July, 1793, about eight in the evening.
She told him of some events at Caen. Marat asked the names of the deputies from Caen,
and began to write down a list of them to have them put to death. The guillotine was an
instrument then employed to cut off people's heads; and Marat said, "Let them all be
guillotined."
"Guillotined!" exclaimed Charlotte, with horror, and plunged the knife into Marat's heart.
"Help!" he cried; "help, my dear!"
His housekeeper and some others ran into the room. He was seen lying covered with
blood, and Charlotte standing motionless beside him.
 PAINTING THE PORTRAIT OF CHARLOTTE
CORDAY ON THE EVE OF EXECUTION.
A crowd gathered around the house; they carried her away to prison. She was brought to
trial before the Revolutionary judges, and showed no signs of emotion or fear. "It was I
that killed Marat," she said. She was condemned to death. She wrote to her father, asking
his forgiveness for having given her life to her country. On the 15th of July she was led
through the streets of Paris to the scaffold. Many of the people followed her with
applause and cries of sympathy. She smiled as her head was cut off, looking beautiful
even in death.
Marat, her victim, was buried by his fellow Jacobins with a great display. His body was
covered with flowers, and his bust or statue appeared in every part of Paris. The Reign of
Terror went on for two years longer. The murders and executions were fearful. But at last
Robespierre, Marat's successor, was killed, and the murderers were punished. Marat's
four thousand busts were thrown down, and his grave dishonored.
As for Charlotte Corday, she was a murderess roused to madness by the crimes of her
victim.
HOW "THE BABY" WENT NUTTING, AND WHAT
CAME OF IT.
BY KATE UPSON CLARK.
"Beats all," said good old Mr. Hurlbut to good old Mrs. Hurlbut, as he laid down the paper
from which he had been reading—"beats all what mizzable little fellers some o' them poor
children in the city be. It seems a good many folks on farms, like us, Sereny, have took
'em in 'n' kep' 'em a spell. Must 'a done the poor little things good. Law! makes me feel
bad."
Good Farmer Hurlbut took off his spectacles and wiped them with great thoroughness.
He was thinking not only of the little newsboys, and the other poor children of whom he
had been reading, in the city, fifty miles away, but of a certain little boy of his own and
"Sereny's," who had gladdened their home for nine short years, and then had died,
leaving them desolate indeed, but with a warm place in their hearts for all his kind.
Presently Farmer Hurlbut spoke again, and, it seemed to Aunt Sereny, rather irrelevantly:
"Lots o' nuts this year up in the north pastur. The clump o' chestnuts is fuller 'n ever—the
biggest chestnuts I ever see; 'n' up higher there's more walnuts 'n' butternuts than you
ever see in your life. Guess we'll have to go over and get George's folks 'n' Eliza Jane 'n'
the girls, 'n' have a picnic some warm day up there, and gather 'em."
"Yes, we must," assented kind Aunt Sereny.
"It would be sorter nice for them poor little fellers in the city to take a day off in the
woods so," continued Farmer Hurlbut, jerking his thumb toward the paper from which he
had been reading.
"Yes, it would," concurred Aunt Sereny.
"But," went on Farmer Hurlbut, with a puzzled expression, "how to get at 'em—that's the
question."
"I should think so," said Aunt Sereny, whose sole mission in life was to agree and to
smooth over and to dispense peace generally.
Suddenly Farmer Hurlbut seized his paper, and began to look over what he had been
reading, passing his finger patiently along the lines.
"I thought so!" he exclaimed at last, pinning a particular place with his big thumb. "I
thought I see the name of the superintendent of the society, 'n' I did. He'd know, I
s'pose."
"Know what?" asked his wife, mildly.
"Why, how to get at 'em."
"Oh!" Aunt Sereny brightened up wonderfully.
"How d'ye s'pose 'twould do to ask a whole raft on 'em to come?" asked Farmer Hurlbut,
reflectively.
"I'd be kinder afraid on 'em, so many, seems to me"—with a little deprecatory laugh.
"Thet's so," said her considerate husband. "They be wild little critters, so I've heerd.
Mebby five or six would be enough. My! how their eyes would shine to see them nuts!"
Aunt Sereny laughed—a wholesome, sunshiny laugh as ever was heard.
"'N' I know," continued Farmer Hurlbut, affectionately, "that you'd feed 'em up, 'n' pet
'em, 'n' do 'em more good 'n all the mission schools in creation."
Aunt Sereny protested modestly, but was sure she would be willing to try and see what
she could do.
There was a little time of silence, during which the clock struck nine.
"Wa'al, what say, Sereny?" said the old farmer at last.
The old lady understood him perfectly.
"I say, Josiah," she replied, with considerable emphasis—"I say, do just as you've a mind
to."
The consequence of this conversation was a letter from Farmer Hurlbut to the
superintendent, and later, the appearance of six ragged boys, equipped with bags, on a
pleasant Wednesday morning in early November, at the railroad station in the city, ready
to take the train which would reach Farmer Hurlbut's at nine o'clock in the forenoon. That
is, six boys were expected. But when the gentleman who was waiting at the station to
put the little party on the cars came to count them, behold! there was a seventh figure,
very much smaller than any of the rest, holding on tight to a bigger boy's hand.
It was a shrunken little mite, with a big coat on it that came to the floor, and a hat that
must have belonged to somebody's grandpa—a comical, pitiful, heart-breaking little
figure as ever was seen.
"Who's that, Tim?" asked the gentleman of the boy to whose hand the little creature was
desperately clinging. He didn't know Tim very well, and had never encountered this tiny
object before.
"I don't know as you'll like it," gasped Tim, apparently in great terror lest he was going to
be circumvented, "but it's the Baby, 'n' he's five years, on'y he's little, 'cause he hasn't
growed, 'n' he's been sick, 'n' mother said as how a whiff o' country'd do him good, 'n'
mebby he could go 'stead o' me. Philly here'll see to him."
"Yes, sir," said Phil Barstow, whose outfit was only less imposing than the Baby's own. "I
know the Baby, 'n' the Baby knows me, 'n' if you think it's too many for Tim to go too, we
kinder decided—Tim's mother 'n' Tim 'n' me—that mebby the Baby'd better go 'stead o'
Tim, or," added Phil, with unexpected heroism, and swallowing hard, "or 'stead o' me."
"It's all right," said the gentleman, who was sure, from the tone of Farmer Hurlbut's
letter, that he wouldn't mind having seven any more than six. "It's all right, Tim. Now
take good care of him, and sit still, all of you."
So "the Baby" was put on board, and the cars moved slowly off.
At the end of their journey, there was Farmer Hurlbut with his big lumber wagon, which
had three boards laid across it for seats. The boys, with their bags and their dreadful
costumes, filed out as soon as the train stopped, their glowing faces revealing
unmistakably their identity.
They were immediately pounced upon and conveyed to their seats in the wagon, where
Aunt Sereny was waiting for them.
Farmer Hurlbut was overflowing with joviality and good-humor. Two great suggestive
baskets and a mighty jug were packed into the front of the wagon, and behind were
various boxes and barrels to hold the surplus nuts.
"And who's this?" asked Aunt Sereny, beaming delightfully from the front seat of the
wagon, and fixing her gaze particularly upon the forlorn little straggler clinging tight to
Tim's hand.
"Please, mum," said Tim, eagerly, "it's the Baby, 'n' he's sick, 'n' mother was for havin'
him come 'stead o' me, but they said mebby you'd take us both."
"Take you both!" exclaimed the dear old lady, wiping her eyes vigorously, and kissing the
Baby's weazened little face, "I guess we will! It'll do him good, likely's not, bless his
heart! Josiah, mebby"—as the horses started off briskly—"mebby," significantly, "the boys
are hungry after their journey. Just get out the little tin cups 'n' I'll give them a drink o'
milk apiece, 'n' mebby a sandwich 'n' a turn-over as we're riding along. It's a good ways
up to the north pastur'," continued the old lady, as she dealt out the things liberally, and
watched them grasped eagerly by the half-starved little creatures.
"There's plenty, boys; eat all you want. Goodness me! Josiah Hurlbut," she whispered to
her husband, "they haven't had nothing to eat for a week—I know they haven't!"
But the chief ecstasy was on the back seat, where the Baby was ensconced between Tim
and Philly, and eagerly swallowing a cup of Aunt Sereny's rich yellow milk.
"Massy, Phil," cried Tim, admiringly, "see the Baby a-drinkin'! How does it taste, Baby?—
good?"
The Baby nodded, a grave smile settling upon his poor little visage under the big hat.
"More," he said, weakly.
"More! My gracious!" said Tim, in the wildest spirits—"more! He wants more, Philly. Hain't
et or drinked so much as this for a month, I sh'd think. Can he have some more, mum?"
reaching out a claw-like hand with the tin cup, which went back brimming full.
Pretty soon the boys began to talk.
"See there!—quick! That's a squirrel, boys—a reg'lar squirrel. Ever see one before?"
"Trout in that brook, bet you a cent, boys! Won't the rest o' the fellers stare when we tell
'em what we've seen?"
"Are there more nuts 'n that"—pointing to a heavily laden tree which they were passing
—"in the place we're going to?"
"Humph!" returned Farmer Hurlbut, the sight of whose ponderous fist had impressed his
wild little crew as much perhaps as his kindness and generosity; "there's more nuts up in
the north pastur', where we're a-goin', than you'll see all the way put together."
In about an hour the north pastur' was reached, and the boys tumbled out of the wagon
amid a jumble of sweet-fern and pennyroyal, and other sweet woodsy-smelling things.
Aunt Sereny found a comfortable seat near by, and fell to knitting as usual, and Farmer
Hurlbut, going to a thicket close at hand, pulled out two long stout poles, which he had
prepared for this very occasion, and laid away a week before.
Then Jim Bowker and Sammy Jones, two of the biggest boys, were sent up two of the
best trees, and once well up, they lay flat along the great branches, and plied the poles
vigorously. The glossy brown nuts and prickly burrs came flying "fast and furious."
The Baby crept timidly out of the wild bombardment, and sat down beside the ample
figure of Aunt Sereny. His tiny hand—the fac-simile of Tim's only less skinny—grasped her
dress firmly. Aunt Sereny put her hand into her pocket and drew forth unheard-of
treasures of peppermints, sweet-flag root, and caraway-seeds. These the Baby gravely
took and devoured.
Noon coming ever so much too soon, Aunt Sereny, amid great applause, suggested
something more in the line of refreshments. She accordingly spread a white cloth over a
great flat rock, and set forth a feast calculated to drive a hungry boy crazy with delight.
Even the Baby fairly laughed aloud.
"I tell you, boys," said Tim, springing to his feet as he heard it, and even dropping a
precious tart in his enthusiasm—"I tell you the Baby hasn't laughed like that since I can
remember. Hi! ain't it jolly?"
The meal fairly over, they lay a little while on the warm dry grass enjoying the mild
sunshine, Aunt Sereny knitting peacefully on. Two or three boys dozed a little, and the
Baby crept up to his old place beside Aunt Sereny, and gathering up his tiny figure upon
her dress, went fast asleep. She spread a light shawl over him, and drew him closer, amid
affectionate and admiring glances from Tim. Tim adored anybody who was good to the
Baby.
Pretty soon Farmer Hurlbut roused them up to go to the walnut-trees, and two other
boys were detailed for duty in the branches, which they beat and beat again with their
poles. "Shucks" were new things to them all.
"Shure enough," said Larry O'Brien, with a fine brogue, "and now I'll know what they
mane whin they say I don't know shucks—but I do, though."
This caused an uproarious laugh, and Larry kept on saying witty things, to the great
amusement of all. Not Sydney Smith himself was ever the source of more delight.
The train was to start at five, and it was nearly that time when the tired, sunburned,
happy little crowd drew up at the railroad station. Aunt Sereny had been having a
whispered consultation with Farmer Hurlbut on the way home, and when they stopped,
she took Tim and the Baby aside.
"Tim," she said, "can't you leave the Baby with us a little while—to stay a week or two,
you know? You tell me where to write, and I'll let your mother know how he gets along.
We'll take good care of him."
Tim gazed at her with open mouth and shining eyes. "The Baby?" he gasped. "Why—
mother—and—me" (slowly) "can't get along 'thout the Baby. He sleeps with me"—his lip
trembling—"every night. Seems 's if I couldn't sleep nohow 'thout his little hand hold o'
mine."
"But he says he'd like to stay," Aunt Sereny answered, coaxingly. "I asked him"—for the
mite had ridden home in Aunt Sereny's lap.
"Does he?" said Tim, brightening. "If he wants to—mebby—well—D'ye s'pose mother'd
like it?"
But Aunt Sereny settled Tim's doubts, and the train finally rolled away without the Baby.
There he staid at the farm-house, and grew so strong and well that he was allowed to
remain for many a long year. Tim and his tired, overworked widowed mother became
frequent visitors to the same hospitable spot, as well as the rest of the boys who had
formed the memorable nutting party. In fact, a nutting party in the north pastur' became
an annual institution, which continues to the present time.
 A PRIVATE CIRCUS.
BY JIMMY BROWN.
There's going to be a circus here, and I'm going to it; that is, if father will let me. Some
people think it's wrong to go to a circus, but I don't. Mr. Travers says that the mind of
man and boy requires circuses in moderation, and that the wicked boys in Sunday-school
books who steal their employers' money to buy circus tickets wouldn't steal it if their
employers, or their fathers or uncles, would give them circus tickets once in a while. I'm
sure I wouldn't want to go to a circus every night in the week. All I should want would be
to go two or three evenings, and Wednesday and Saturday afternoons. There was once a
boy who was awfully fond of going to the circus, and his employer, who was a very good
man, said he'd cure him. So he said to the boy: "Thomas, my son, I'm going to hire you
to go to the circus every night. I'll pay you three dollars a week, and give you your board
and lodging, if you'll go every night except Sunday: but if you don't go, then you won't
get any board and lodging or any money." And the boy said, "Oh, you can just bet I'll
go!" and he thought everything was lovely; but after two weeks he got so sick of the
circus that he would have given anything to be let stay away. Finally he got so wretched
that he deceived his good employer, and stole money from him to buy school-books with,
and ran away and went to school. The older he grew the more he looked back with
horror upon that awful period when he went to the circus every night. Mr. Travers says it
finally had such an effect upon him that he worked hard all day and read books all night
just to keep it out of his mind. The result was that before he knew it he became a very
learned and a very rich man. Of course it was very wrong for the boy to steal money to
stay away from the circus with, but the story teaches us that if we go to the circus too
much, we shall get tired of it, which is a very solemn thing.
We had a private circus at our house last night—at least that's what father called it, and
he seemed to enjoy it. It happened in this way. I went into the back parlor one evening,
because I wanted to see Mr. Travers. He and Sue always sit there. It was growing quite
dark when I went in, and going toward the sofa, I happened to walk against a rocking-
chair that was rocking all by itself, which, come to think of it, was an awfully curious
thing, and I'm going to ask somebody about it. I didn't mind walking into the chair, for it
didn't hurt me much, only I knocked it over, and it hit Sue, and she said, "Oh my, get me
something quick!" and then fainted away. Mr. Travers was dreadfully frightened, and said,
"Run, Jimmy, and get the cologne, or the bay-rum, or something." So I ran up to Sue's
room, and felt round in the dark for her bottle of cologne that she always keeps on her
bureau. I found a bottle after a minute or two, and ran down and gave it to Mr. Travers,
and he bathed Sue's face as well as he could in the dark, and she came to and said,
"Goodness gracious, do you want to put my eyes out?"
Just then the front-door bell rang, and Mr. Bradford (our new minister) and his wife and
three daughters and his son came in. Sue jumped up and ran into the front parlor to light
 the gas, and Mr. Travers came to help her.
They just got it lit when the visitors came in,
and father and mother came down stairs to
meet them. Mr. Bradford looked as if he had
seen a ghost, and his wife and daughters
said, "Oh my!" and father said, "What on
earth!" and mother just burst out laughing,
and said, "Susan, you and Mr. Travers seem
to have had an accident with the inkstand."
You never saw such a sight as those poor
young people were. I had made a mistake,
and brought down a bottle of liquid blacking
—the same that I blacked the baby with that
time. Mr. Travers had put it all over Sue's
face, so that she was jet black, all but a little
of one cheek and the end of her nose, and
then he had rubbed his hands on his own
"OH, MY!" face until he was like an Ethiopian leopard,
only he could change his spots if he used
soap enough.
You couldn't have any idea how angry Sue was with me—just as if it was my fault, when
all I did was to go up stairs for her, and get a bottle to bring her to with; and it would
have been all right if she hadn't left the blacking bottle on her bureau; and I don't call
that tidy, if she is a girl. Mr. Travers wasn't a bit angry; but he came up to my room and
washed his face, and laughed all the time. And Sue got awfully angry with him, and said
she would never speak to him again after disgracing her in that heartless way. So he
went home, and I could hear him laughing all the way down the street, and Mr. Bradford
and his folks thought that he and Sue had been having a minstrel show, and mother
thinks they'll never come to the house again.
As for father, he was almost as much amused as Mr. Travers, and he said it served Sue
right, and he wasn't going to punish the boy to please her. I'm going to try to have
another circus some day, though this one was all an accident, and of course I was
dreadfully sorry about it.
LAWN TENNIS TOURNAMENT AT THE ST. GEORGE'S CLUB
GROUND, HOBOKEN, NEW JERSEY, OCTOBER 27, 1881.
 A LAWN TENNIS TOURNAMENT.
BY SHERWOOD RYSE.
Although the game of lawn tennis, which was introduced to the readers of Young People
early in the summer, has made giant strides in popularity, it does not seem to carry its
character in its face, for there are still people to be found who have seen the game and
yet have not appreciated its merits. More than one person has said to me, "I don't see
much fun in knocking a ball over a net for a person on the other side to knock it back
again."
Now there is a great deal of reason in that. To knock a ball over a net for another person
to knock it back again would be very poor fun. But, as we know, the object in knocking
the ball over the net is that the other person shall not knock it back, which is quite
another thing, and which, indeed, is the essence of the game.
Should this view of the case fail to convince the ignorant persons above referred to that
lawn tennis is a game deserving of respect, and that it is not, what Dr. Johnson called
fishing, the pastime of fools, I would take them to see a lawn tennis tournament. I would
do that, however, only out of pure good nature, for it would be a great deal more
pleasant to look on at a tournament in company with some one who knows the game.
And so, if you please, I will take my readers to the tournament at the St. George's cricket
ground at Hoboken, New Jersey. The name of the club suggests that it is English in its
origin; and that is a good omen, for is not old England the home of lawn tennis, as it is
also of cricket?
Eight courts are laid out on the carefully prepared ground, which is refreshingly green
even after this long dry summer, and several games are in progress.
Our artist has chosen for the subject of his illustration on page 41 the double-handed
match between Messrs. Anderson and Henry, of the Seabright Club, and Messrs. D. and
G. F. Miller, of Utica. Though the double-handed game is very interesting, it does not
possess the same attraction, for players at least, as a single-handed contest, in which
one player has to cover the whole of his court. Not that the young player who looks
forward to taking part some day in a public tournament should neglect the double game.
It is, indeed, a very necessary part of the practice required to make a player thoroughly
at home in the game, for it teaches him how to "place" his "returns."
Watch the players carefully, and notice the quick decision required to place the ball
beyond the reach of both their antagonists. In a single-handed game there is only one
man's vigilance to outwit. In the double game there are two, and one of two partners, if
they are both good players, should always be within reach of the ball wherever it may be
placed. Thus you see that a young player who has learned to place his returns well in a
double game will find that part of his work much easier when he has only one antagonist
on the other side of the net.
But while I have been talking about "placing," the crowd has gathered around a court
where a single-handed game is being played. Let us, then, practice what we preach, and
place ourselves where we can see the game. It is between Mr. Anderson—the same
whom we saw playing in the double game—and Mr. Cairnes, a young Englishman who is
on a visit to this country, and has returned the hospitality he has received by beating the
lawn tennis champion of the United States. Ah, well, we will forgive him, for he is young
—barely twenty-one, judging from his looks—and he does not know any better. But he
can play tennis.
As we take our places, the scorer calls, "Two games all." Anderson plays up well, and
wins the next game, and still another. The doughty Englishman is getting beaten; he is
playing carelessly. But see! It is very plain that he recognizes the fact that the games are
going on too fast, for as soon as he learns that the score is four to two in Anderson's
favor, his play begins to improve. He wins the next four games in succession, and so wins
the set. And right well did he play.
It is difficult to say wherein lies his great excellence. It is not in his "service." Service is all
very well, and it is very useful to have a good service, especially when playing against
indifferent antagonists; but among the best players service does not count for much. The
"return" is of very much more importance, if for no other reason than that one has many
more balls to return than to serve. In the first place, you should make certain that your
ball is going over the net. Youth is ambitious, and ambition every now and then gets a
fall; and so the young player who tries to just skim the top of the net every time is very
apt to drive his ball into instead of over the net. It is much better to send even the
easiest kind of a ball for your adversary to return, for there is always a chance of his foot
slipping, or something of the kind; or perhaps he will be ambitious, and drive the ball
with great skill and precision into the middle of the net. The English player returned his
balls very closely over the net, but they always went over, and doubtless his accuracy in
that respect is the result of long practice.
Another point in which he excelled was the skillful manner in which he placed the ball
close to the side lines in the back court. This is very pretty work, but it is also dangerous,
for it must always be remembered that there is not a hair's-breadth between a "good"
ball and a bad one, between just in court and just out. One is success, and the other
failure. For young players there are many opportunities of placing a ball out of the
opponent's reach without playing it right up to the base line or side lines of the court. In
tennis, as in other things, a middle course is safest for beginners.
Although lawn tennis has sprung rapidly into favor, it is still but a new game in this
country. It takes several seasons' play for a person to become a first-rate player. By the
time most of my readers are old enough to take part in a public tournament, some of
them will probably play better than the best players of to-day. As time goes on, the
standard of the game grows higher. The best players to-day are men, and they did not
have the great advantage of beginning to learn tennis when they were boys.
But it is not only a boys' game; it is quite as suitable for girls, and many girls and grown
ladies play very well, in spite of the man who said in an article on the subject not long
ago that all ladies were "duffers" at tennis. If some of our lady players were to express
Welcome to our website – the perfect destination for book lovers and
knowledge seekers. We believe that every book holds a new world,
offering opportunities for learning, discovery, and personal growth.
That’s why we are dedicated to bringing you a diverse collection of
books, ranging from classic literature and specialized publications to
self-development guides and children's books.

More than just a book-buying platform, we strive to be a bridge

connecting you with timeless cultural and intellectual values. With an
elegant, user-friendly interface and a smart search system, you can
quickly find the books that best suit your interests. Additionally,
our special promotions and home delivery services help you save time
and fully enjoy the joy of reading.

Join us on a journey of knowledge exploration, passion nurturing, and

personal growth every day!

ebookbell.com