Page 246 - Data Science Algorithms in a Week

230 Khaled Alshareef, Ahmad Rahal and Mohammed Basingab

INTRODUCTION

The gap between healthcare spending and economic growth in many nations around
the world including the United States, has been widening at a faster rate requiring scares
resources be allocated to mitigate the impact of the steep rise of healthcare cost instead of
being devoted to economic growth. This uncontrolled phenomenon can be attributed to
many factors including population growth, population aging (Thorwarth & Arisha, 2009),
the development cost of new technologies (Aboueljinane, Sahin, & Jemai, 2013), and the
use of expensive new diagnostic tests and treatments. Furthermore, the limited
availability and the over-utilization of healthcare facilities and providers such as
physicians, nurses, and others (Tien & Goldschmidt-Clermont, 2009), have also
attributed to the deterioration of the efficiency and effectiveness of healthcare processes,
and the degradation of the proper delivery of healthcare services (Faezipour & Ferreira,
2013).
Discrete Events Simulation (DES) has been used by many healthcare organizations as
a tool to analyze and improve their healthcare processes such as delivery systems, patient
flow, resources optimization, and patient admission (Gosavi, Cudney, Murray, & Masek,
2016; Hamrock, Paige, Parks, Scheulen, & Levin, 2013; Katsaliaki & Mustafee, 2010;
Parks, Engblom, Hamrock, Satjapot, & Levin, 2011). However, the use of DES poses
many challenges including the modeling complexity of the healthcare environment, the
lack of real data, and the difficulty in the implementation of the proposed solutions and
recommendations. Furthermore, the need to understand the stochastic nature of the
decision-making modeling process has limited the involvement of many healthcare
decision makers, and has reduced the effectiveness of the use of simulation in the
healthcare field as compared to other fields (Roberts, 2011).

CASE-BASED REASONING (CBR)

The advancement in artificial intelligence (AI) technologies have led to the
development of many technologies including genetic algorithms, fuzzy logic, logic
programing, neural networks, constraint-based programing, rule-based reasoning, and
case base reasoning (CBR). CBR is a computerized method that reuses and if necessary
adapts solutions of previously solved problems. “CBR basically packages well-
understood statistical and inductive techniques with lower-level knowledge acquisition
and representational schemes to affect efficient processing and retrieval of past cases (or
experiences) for comparison against newly input cases (or problems)” (Mott, 1993). It
uses database management and machine learning techniques to perform the retrieval
process (Bichindaritz & Marling, 2006; Watson, 1999).