Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 12

PREECLAMPSIA EXPERT SYSTEM MODEL

USING HILL CLIMBING METHOD

Agung Setiawan1, 2, Fauzi Erwis1, 2, N Jalinus2, Ambiyar2 , M Giatman2


1
Lecturer in Information Technology Education, STKIP. Rokania, Riau
2
Technical and Vocational Education and Training, Universitas Negeri Padang, Sumatera Barat

ABSTRACT
One of the causes of maternal failure is the presence of blood pressure that suddenly rises. This can be caused
by several factors, one of which is preeclampsia and eclampsia, where a mother's blood is poisoned by the baby she
is carrying. The diagnosis of preeclampsia is based on several symptoms, namely excessive weight gain, edema,
hypertension, and proteinuria. Edema is seen as an increase in body weight, swelling of the feet, fingers and face.
Blood pressure> = 140/90 mmHg or systolic pressure increases> 30 mmHg or diastolic pressure> 15 mmHg which
is measured after the patient has rested for 30 minutes. Diastolic pressure in the second trimester of more than 85
mmHg is suspected as a talent for preeclampsia. Proteinuria is a condition when there is protein as much as 0.3 g / l
in urine 24 hours or a qualitative examination shows +1 or 2; or protein> 1g / l in urine excreted with a catheter or
middle portion, taken at least 2 x with an interval time of 6 hours. It is with the Hill Climbing method, the accuracy
being produced above with pregnancy is more than or equal to /> = 20 weeks.

Keywords: Expert System, Preeclampsia, Eclampsia, Hill Climbing

1. Introduction
Pregnancy health problems must be considered well, for the safety of the mother and baby born. There are
many problems faced by pregnant women, one of which is the increase in blood pressure due to pregnancy. This is
mostly not realized by pregnant women, because of the absence of high blood pressure in pregnant women. One
increase in blood pressure in pregnant women can be caused by the poisoning of blood by the baby they contain.
The increase in blood pressure in pregnant women is also called preeclampsia and eclampsia.
Various interventions using information technology (IT) have been developed to improve medication safety,
and IT-based interventions such as clinical decision support system(CDSS) plays an integral role in this field (Pengli
Jia, 2016). According to the definition of the US Food and Drug Administration, medication safety means risk
management, medication errors, and surveillance for adverse drug reactions (Pengli Jia, 2016).
Medication errors are recognized as the single most preventable cause of patient harm, and their reduction is of
increasing importance (Pengli Jia, 2016).
Clinical decision support systems (CDSSs) can be defined as software that is designed to be a direct aid to
clinical decision-making in which the characteristics of an individual patient are matched to a computerized clinical
knowledge base (KB), and patient-specific assessments or recommendations are then presented to the clinician
and/or the patient for a decision (Filip Velickovski, 2014).
CDSS reduces medication error by obviously improving process of care and inconsistently improving patient
outcomes. Larger samples and longer-term studies are required to ensure more reliable evidence base on the effects
of CDSS on patient outcomes. The methodological and reporting quality were varied and some realms need to be
improved (Pengli Jia, 2016). One of the main challenges here is how to realistically handle the possible uncertainties
so that a CDSS can support clinical experts to make correct and reliable diagnosis and treatment decisions (Imran
Sarwar Bajwa, 2017).
Besides the lack of well-described success factors, an often mentioned barrier to implementation is the low
computer skills among physicians. This must be carefully taken into account within the design of the CDSS alerts.
New generation physicians, like medical students and junior physicians, may bring a higher level of computer
literacy to clinical practice and stimulate implementation of a CDSS in practice (Garg, 2005). Another barrier,
identified by Bates (Bates, 2003) is the loss of physician’s autonomy with the use of CDSS. However, CDSS are
able to present the best evidence-based practice automatically without requiring extra thought or work. This allows
the health professionals to focus on those areas of special need and adjust care to each individual patient. This not
only increases patient safety, but also physician’s safety by reducing the risk on malpractice. Moreover, the system
may improve clinical skills through a learning effect of the corrective messages, so it can improve the performance
of professionals over time (Mohamed Khalifa, 2014).
Through CDSS, it is expected to reduce errors in the detection of a woman's pregnancy and can minimize the
failure of the birth process. To make a good CDSS, an expert system thus, is needed to be able to provide problem
solving for pregnant women. In this study using the hill climbing, method for expert system methods is created.

2. Related Work

2.1 Expert System


In the realm of Natural Language Processing much of the work in deep learning has been oriented towards
methods involving learning word vector representations using neural language models. (Oscar Araque, 2017) The
expert system’s role is in the preparation to capture the knowledge of the experts and the data from the customer’s
requirements. The system has the capability to compile the collected data and form the appropriate rules for
choosing fragrance notes for the products. In order to identify the hidden pattern of the customer’s needs, the
artificial neural networks technique has been applied to classify the fragrance notes based upon a list of selected
information (Oscar Araque, 2017).
An expert system, also known as a knowledge-based system, is a computer-based system that uses captured
human knowledge to solve problems that ordinarily require human experts (Tomei, 2008). An expert system is a
computer program that uses encoded knowledge and rules of reasoning to draw conclusions or solve problems.
Since reasoning (as opposed to mechanical calculation) is a form of intelligent behavior, the field of expert systems
(also called knowledge representation or knowledge engineering) is part of the broader field of AI (artificial
intelligence) (Henderson, 2009).
Furthermore, the expert system’s role is in the preparation to capture the data from the customer’s requirements
and predict appropriate perfume. For this end, factors of perfume costumers’ decision were recognized using Fuzzy
Delphi method and a back propagation neural network classification model was developed and trained with 2303
data of customers (P.Isakki alias Devi., 2011).
Expert systems aroused to perform a wide variety of complicated tasks that can only be performed by highly
trained human experts. Although the problem domain for robot selection may be considered as narrow, which fits
the expert system structure. It is also complex and it requires a multi criteria decision making (MCDM) procedure.
An expert system needs to access the database of a decision-making system to gather factual knowledge.
Furthermore, the judgmental data obtained from experts can be incorporated into a decision-making system through
an expert system (Wah, 2008).
In addition to that, the goal of expert systems is to ensure that scarce expertise can be utilized when a human
expert is not available (e.g., due to cost, other commitments, illness, and retirement) and when efficiency or
consistency of an expert needs to be enhanced. Expert systems can be utilized to alleviate these needs from simple to
complex situations, such as when a decision maker needs some expert advice about a problem, a human expert may
not be available. Instead of waiting or paying to consult another human expert, the decision maker could
immediately consult an expert system to get comparable advice (Tomei, 2008).
Figure 1. Expert System Structure

The inference engine examines the knowledge base and produces reasoning. The knowledge engineering tool
allows for changing or enlarging the knowledge base by adding further rules, cases, or models. There may also be an
explaining component, which illustrates the diagnostic process and which gives a rationale. A knowledge-based
expert system with an empty knowledge base is called shell. It can be used for the development of other expert
systems by adding a new knowledge base. (Hannes Alder, 2014)
An expert system (ES) known as knowledge based system, is a computer program that uses knowledge and
inference procedures to solve problems that are ordinarily solved through human expertise. The main components of
an ES are: a) knowledge base, b) inference engine, c) user interface. (R.A. Soltan, 2013)

Figure 2. Typical Structure of a Knowledge-Based Expert System (Hannes Alder, 2014)

There are many applications of expert systems such as diagnosis, design, planning, financial decision making
etc. Most applications of expert systems in medicine involve predicting, diagnosing and treating a particular disease
(R.A. Soltan, 2013). Currently, expert systems has many other roles in clinical care such as disease prevention,
therapy, rehabilitation of the patient after therapy etc. In medicine, an expert system is used to train medical students
about various medical tasks. In certain situations, where either the case is quite complex or there is no medical
experts readily available for patients medical, expert systems are useful. From the very beginning, the main obstacle
of using expert systems in medicine has been the accuracy of such systems (Karagiannis S., 2006).
The development of an expert system requires medical data of specialized doctor. This data is collected in two
phases. Firstly, the creation of personal interview between doctor and patient record the medical background of heart
disease. Secondly, medical data is turned into rules (IF THEN). Rules for diagnosis contain in IF part the symptoms
and in THEN part the disease. Rules for treatment contain in IF part the disease and in THEN part the treatment. The
inference engine (forward reasoning) is the mechanism through which rules are selected to be fired. It is based on a
pattern matching algorithm whose main purpose is to associate the facts (input data) with applicable rules form the
rule base. Finally, the heart diseases are produced by the inference engine (Jackson, 1999).

2.2 Preeclampsia and Eclampsia


Preeclampsia is a pregnancy which is related to disorder, where there is a development of hypertension and
proteinuria with or without edema in a normotensive and non-protein uric women after 20 weeks of gestation. The
changes usually subsided after delivery of placenta. The exact cause of these maternal changes in relation to
pregnancy is not clear. There is generalized maternal vascular endothelial dysfunction and leukocyte activation (Das
Brahmarshi, 2012).
Ruptured sub capsular hematoma of the liver (RSHL) in pregnancy is uncommon and occurs in 1:45,000 to
1:225,000 deliveries. It usually develops as a complication of the hemolysis, elevated liver enzymes, and low
platelet (HELLP) syndrome, associated with pre-eclampsia. Patients with RSHL present with life-threatening
intra-abdominal hemorrhage, leading to hypotension. This clinical picture mimics a ruptured interstitial pregnancy
that also results in hypotension, and may present after the first trimester (even at 30 weeks’ gestation), which is also
the same gestational period when the clinical signs of pre-eclampsia develop (NC Ngene, 2015).
Preeclampsia (PE) is a serious complication of pregnancy with potentially life threatening consequences for
both mother and baby. Presently there is no test with the required performance to predict which healthy first-time
mothers will go on to develop PE. The high specificity, sensitivity, and multiplexed nature of selected reaction
monitoring holds great potential as a tool for the verification and validation of putative candidate biomarkers for
disease states. (Richard T. Blankley, 2013).
Preeclampsia is a pregnancy related disorder, where there is development of hypertension and proteinuria with
or without edema in a normotensive and non-protein uric women after 20 weeks of gestation. The changes usually
subsided after delivery of placenta. The exact cause of these maternal changes in relation to pregnancy is not clear.
There is generalized maternal vascular endothelial dysfunction and leukocyte activation (Das Brahmarshi, 2012).
The leading cause of fetal–maternal morbidity and mortality is preeclampsia (PE), which manifests as proteinuria
and hypertension in 5–8% of pregnancies. Distinguishing PE from chronic kidney disease (CKD) is important since
earlier delivery is warranted in PE. (A Rolfo, 2012)
Preeclampsia (PE) is characterized by increased blood pressure, proteinuria and/or edema developing after 20
weeks of gestation. Though the exact etiology is unknown, early vascular dysfunction manifested as defective
placentation and later systemic endothelial dysfunction have been documented to be a common feature of PE. P-
selectin, a cell adhesion molecule present in platelets and endothelial cells is expressed on the surface when cells are
activated. It mediates interactions between leucocytes, platelets and endothelium. The soluble form of P-selectin (sP-
selectin) is derived mainly from platelets after it has been cleaved from the cell surface (Sujata Wangkheimayum,
2011).
Subsequently, blood pressure that suddenly rises at 20 weeks' gestation can be an early indication of the
presence of preeclampsia or eclampsia, even though pregnant women do not have medical records with high blood
pressure. At the time it is not treated quickly, it can endanger the lives of pregnant women and their babies.
The parameters of the success of maternal health services can be seen from the maternal mortality rate and the
mortality rate of newborns. However, some developed countries consider the success of reducing maternal mortality
better than the rate of death of babies born. This is because the health and safety of the fetus in the womb is very
dependent on the condition and perfection of the workings of the system in the mother's body, which has a function
to grow the results of conception from mudigah to a month-long fetus. One of the causes of death of newborns is
preeclampsia and eclampsia.

2.3. Hill Climbing


The word Heuristic comes from a Greek verb, heuriskein, which means 'seek' or find. In the world of
programming, some people use the word heuristic as opposed to algorithmic words, where the word heuristic is
interpreted as a process that might solve a problem but there is no guarantee that the solution sought can always be
found. There are several heuristic search methods, namely:
a. Generate and Test
In principle this method is a combination of depth-first search with backtracking, which is moving backwards
towards an initial state. Algorithm:
• Generate a possible solution (generating a certain point or certain path from the initial state).
• Test to see whether the node is really the solution by comparing the node or the end node of a selected path with
a collection of expected goals.
• If a solution is found, exit. If not, repeat the first step.
Weakness:
• Need to raise all possibilities before testing
• Requires a long time in the search
b. Hill Climbing
1. Simple Hill Climbing
• This method is almost the same as the generation & testing method, except that the testing process is
carried out using the heuristic function.
• The next generation of generation is very dependent on feedback from the testing procedure.
• The test in the form of a heuristic function will show how well the guessed value is taken against other
possible conditions.
Algorithm
 Starting from the initial state, do the test: if it is a destination, stop; and if not, continue with the current
state as a starting state. Work on the following steps until the solution is found, or until no new operator
will be applied to the current situation:
 Search for operators that have never been used; use this operator to get a new state.
 Evaluate the new condition.
 If the new condition is a destination, exit.
 If it is not the goal, the value is better than the present situation, so make the new situation into the
present situation.
 If the new situation is no better than the current state, then continue the iteration.

2. Steepest Ascent Hill Climbing


Algorithm
 Starting from the initial state, do the test: if it is a destination, stop; and if not, continue with the current
situation as the initial state.
 Work until the goal is reached or until the iteration does not change the current situation.
 Determine SUCC as the best heuristic value from successors.
 Work for each operator used by the current state:
 Use the operator and form new conditions.
 Evaluate the new condition. If it is a destination, exit. If not, compare the heuristic value with the
SUCC. If it's better, make the heuristic value of the new condition a SUCC. But if it's not better, the
SUCC value doesn't change.
 If SUCC is better than the current heuristic value, change the SUCC node to the current state.

c. Best First Search


 This best-first search method is a combination of depth-first search method and breadth-first search method
by taking advantage of the two methods.
 If the search for hill climbing method is not allowed to return to the node at a lower level even though the
node at the lower level has a better heuristic value, it is different from the best-first search method.
 In the best-first search method, search is allowed to visit existing nodes at a lower level, if it turns out that
nodes on higher ones have worse heuristic values.
 To implement this method using graph conditions, it takes 2 queues containing nodes, namely:
o OPEN, contains nodes, nodes that have been raised, but have not been tested. Generally in the form of a
priority queue containing elements with the highest heuristic value
o CLOSED contains nodes that have been tested
Algorithm
• Place the initial node A in the OPEN queue.
• Work on the following steps until the destination is found or the OPEN queue is empty:
• Take the best node from OPEN;
• Awaken all of its successors;
• For each successor to do:
o If the node has never been raised before, evaluate the node and enter it into OPEN;
o If the node has been previously raised, change the parent if the new path is more promising. Remove the
node from the OPEN queue.

Hill climbing (HC) technique is used to find the factors of an integer number. Hill Climbing is a local search
technique. It starts with an initial solution and steadily and gradually generates neighboring successor solutions. If
the neighboring state is better than the current state then the neighboring state is considered as the current state.
There are different variants of hill climbing namely simple hill climbing, steepest hill climbing, stochastic hill
climbing and random restart hill climbing (B. Choudhury, 2015).
The HC uses the iterative improvement technique, which is applied to a single point in the search space. A new
point is selected from the neighborhood of the current point. If the new point provides a better value of the objective
function, the new point becomes the current point. It terminates if no further improvement is possible. On the other
hand, the Genetic Algorithms (GA) performs a multi-directional search by maintaining a population of potential
solutions and encourages information exchange between these directions. (Ceylan, 2006).
Generalized hill climbing algorithms provide a framework to describe and analyze metaheuristics for
addressing intractable discrete optimization problems. The performance of such algorithms can be assessed
asymptotically, either through convergence results or by comparison to other algorithms (Jacobson, 2004).
The Hill climbing algorithm adaptively determines the initial cluster centers and the number of clusters
according to the characteristics of the image. Using hill climbing algorithm as a preliminary stage with clustering
algorithms reduces the number of iterations for classification and costs less execution time (B. Sai Chandana, 2014).
In the biological world, ants construct a foraging path using a volatile substance called a pheromone, which has
been widely studied and whose characteristics have been used in a transportation network model. When a navigation
path is constructed by autonomous agents using this pheromone model, the created potential field can be very noisy,
with many local peaks due to the unsynchronized updates of the field (Piljae Kim, 2011).

3. Research Methodology
. This study uses a framework consisting of steps from analyzing the problem to the model produced. The
framework used in this study is as follows:

Figure 3. Research Methodology

1. Analysis
System analysis is the decomposition of an intact information system into its component parts with the intention
of identifying and evaluating problems, opportunities, constraints that occur and expected needs so that
improvements can be proposed.
A System Analysis & Design (SAD) methodology can also be referred to as a Systems Development Life Cycle
(SDLC) that includes the development process as well as the ongoing maintenance process. The classic SAD
methodology is the waterfall model which was originally conceived for the software development; hence the
focus is on programming. The key phases of the waterfall model are the analysis and design phases. It is obvious
that there will always be an implementation phase and an operations phase. The analysis phase focuses on
understanding the needs of the organization. The design phase focuses on designing the physical aspects of a
system to support the needs of the organization (Dobing B, 2000).
The analysis in this study is used to carry out a needs analysis to develop a preeclampsia expert system model.
From the results of the analysis, it is found that many researchers or end-level students have difficulty in getting
preeclampsia references for their research, so that more research questioned experts on preeclampsia, through
interviews and observations.
2. Collection Source
The collection of sources in this study uses several journals and books that serve to increase knowledge and
support this research. In collecting resources, it is done by downloading from the internet and conducting
research from the sciences. In addition to using journals and books, this research is also supported by sources
from analysis experts and experts, obstetricians and other experts in the field of preeclampsia and information
technology through observation and interviews.
3. Design
Information Systems Analysis and Design is pivotal for both the success of IT projects and the IS graduates
skills
and knowledge portfolio. All the ACM/AIS curricula guidelines for both undergraduate and graduate IS
programmers place analysis and design as a main pillar in graduates formation. It is with many differences in
number and/or length, European and worldwide IS programmers have a number of courses which treat
analysis/design topics. Apart from IS Analysis, IS Design, there are courses with parts and chapters dedicated to
IS A&D subjects, such as Business Process Management, Databases, Object-Oriented Programming, Software
Engineering, etc. (Marin Fotache, 2015)
The system design definition is the stage after analysis of the system development cycle: defining functional
requirements and preparation for design implementation; describe how a system is formed. The results of the
above analysis are used to design the prototype that will be produced in this study. The design of the research
uses data flow diagrams (DFD) and unification modeling language (UML), so that a design for an android java
application is produced.
4. Schema Making
Schema making in this study is intended to make an initial design in modeling, so it is known that the flow of the
model is made. In this study the sketcher is illustrated by the making of the groove from the beginning of Hill
Climbing making, namely:

Figure 4. Model Flow Scheme


The picture above shows that the scheme starts from the patients who have problems accompanied by input from
experts and other sources, such as research papers, journals, etc. Inputs from experts and other sources are
grouped into a knowledge data, continued into the knowledge base to be processed in inference machines.
Knowledge base besides getting input from knowledge data, can also be input from patients through knowledge
engineering tools as problem renewal.

Further, in the next step, the patient enters the problem into the user interface as the media asks into the inference
machine to find information about the problem at hand. The inference engine then works by processing the input
results through the user interface and is connected to the knowledge base. The results of processing machine
information are then displayed in the user interface as information as solving patient problems.

Figure 5. Hill Climbing Inference Machine

5. Making Model
The process of detecting preeclampsia and eclampsia begins with checking the patient's blood pressure (BP) or
checking the patient's medical record related to the patient's blood pressure before becoming pregnant.
After examining the patient or checking the patient's medical record, it is known that the patient's blood pressure
rises to BP> = 140/90 mmHg, so additional examinations need to be carried out as follows:
a. Headache, impaired vision, proteinuria, hypereflexia, coma, and gestational age less than / <20 weeks, the
result is chronic hypertension or superimposed preeclampsia (Rule 3/R3).
b. Headache, impaired vision, proteinuria, hyperflexia, coma, gestational age more than or equal to / <= 20
weeks and decreased body spasms (-), the result is hypertension or mild preeclampsia or severe preeclampsia
(Rule 1/R1).
c. Headache, impaired vision, proteinuria, hypereflexia, coma, gestational age more than or equal to / <= 20
weeks and increased body spasms (+), the result is eclampsia (Rule 2/R2)..
If the blood pressure is normal and further examination is needed, then another step can be taken to examine
other symptoms, then proceed with 3 checks as follows:
a. Rising seizures (+), decreased fever (-) and decreased stiffness (-), the result is epileptic hypertension
b. Headache, ascending fever, raised (+) curduk, disorientation, the result is malaria meningitis.
c. Headache, vision problems and vomiting, the result is migraine.
Figure 6. Research Method with Hill Climbing

After discovering the flow of the research method with Hill Climbing, the next step is to make the system model
flowchart created. To facilitate the making of the system model created, it is carried out beginning with the age
of pregnancy.

Figure 7. Flowchart of Application Model for Preeclampsia and Eclampsia


The next step is to make a system model created, in this study using a java application program. However, for
making models, other application programs can be used, such as visual basic, PHP or others.

Figure 8. Application Model of Preeclampsia and Eclampsia

6. Testing Model
Stage of Testing models in this study is done by entering data from experts and other sources into the knowledge
base. Expert results and other sources as indicators are the basis for the characteristics of preeclampsia and
eclampsia, namely the age of the womb> = 20 weeks, blood pressure> = 140/90 mmHg, proteinuria,
hyperreflection and coma. All data is put into the inference engine to be processed and information will be
generated for pregnant women. So that the results obtained from this application model are in accordance with
the AWA design, namely with the above characteristics, pregnant women suffer from preeclampsia or eclamsia.

7. Test Result
The test results found that the condition of pregnant women with womb age conditions> = 20 weeks, blood
pressure> = 140/90 mmHg, proteinuria, hyper reflection and coma have been confirmed that the mother is
suffering from preeclampsia or eclampsia, so it needs to be handled specifically. Apart from these cases, it is
ensured that the mother does not contain preeclampsia or eclampsia.

4. Results and Discussion


The diagnosis of preeclampsia is based on the presence of two of three symptoms, namely excessive weight
gain, edema, hypertension, and proteinuria. Excessive weight gain if there is a 1 kg increase a week and occurs
several times. Edema is seen as an increase in body weight, swelling of the feet, fingers and face. Blood pressure> =
140/90 mmHg or systolic pressure increases> 30 mmHg or diastolic pressure> 15 mmHg which is measured after
the patient has rested for 30 minutes. Diastolic pressure in the second trimester of more than 85 mmHg is suspected
as a talent for preeclampsia. Proteinuria if there is protein as much as 0.3 g / l in urine 24 hours or a qualitative
examination shows +1 or 2; or protein> 1g / l in urine excreted with a catheter or middle portion, taken at least 2 x
with a time interval of 6 hours.
It is with the application that has been made it can be concluded that the accuracy produced by this application
is above 80%, in conditions that are in accordance with the characteristics of preeclampsia and eclampsia. For other
diseases, this study is not discussed in detail and must refer from experts or other sources related to other diseases
other than preeclampsia or eclampsia.

5. Conclusion
The results of the analysis and design in this study are carried out by using documentation from several sources
and asking directly from obstetricians, especially those who understood Preeclampsia and Eclampsia. Therefore, the
conclusions are as follows:
1. Accuracy of results obtained above 80% with pregnancies greater than or equal to /> = 20 weeks, indeed some
cases can occur with pregnancies of less than 20 weeks. The test results produced are generally preeclampsia.
2. This research can be developed for content below 20 weeks, so that other diseases can be known to suffer by
pregnant women and can be anticipated by giving the appropriate medication.
3. The resulting application model can be developed, to detect other diseases beyond preeclampsia and eclampsia,
such as high blood pressure, migraine in pregnant women.

6. References
A Rolfo, R Attini, AM Nuzzo, et al (2012). Chronic kidney disease may be differentially diagnosed from
preeclampsia by serum biomarkers. Kidney International 2012, published online jan 2013; 83(1):177-181 doi:
10.1038/ki.2012.348.
B. Choudhury, S. Neog (2015). Finding Factors of a Number Using Steepest Ascent Hill Climbing. International
Journal of Electronics and Applied Research (IJEAR) 2015, vol. 2, issue 1: 29-34.
B. Sai Chandana, K. Srinivas, R. Kiran Kumar (2014). Clustering Algorithm Combined with Hill Climbing for
Classification of Remote Sensing Image. International Journal of Electrical and Computer Engineering
(IJECE) Vol. 4, No. 6, December 2014: 923-993.
Bates, D. W., Kuperman, G. J., Wang, S., Gandhi, T., Kittler, A., Volk, L., ... & Middleton, B. (2003). Ten
commandments for effective clinical decision support: making the practice of evidence-based medicine a
reality. Journal of the American Medical Informatics Association, 10(6), 523-530.
Benjamin W. Wah (2008). Wiley Encyclopedia of Computer Science and Engineering. Wiley-interscience. New
York. USA. E: 1.
Das Brahmarshi, Saha-Roy Santa, Das Gupta Anindya, Lahiri Tapan Kumar, Das Harendra Nath (2012).
Assessment of Placental Oxidative Stress in Preeclampsia. The Journal of Obstetrics and Gynecology of India
(January–February 2012) 62(1):39–42. doi: 10.1007/s13224-012-0150-z.
Dobing B, Parsons J (2000). Understanding the Role of Use Cases in UML: A Review and Research Agenda.
Journal of Database Management 11(4): 28-36; doi: 10.4018/jdm.2000100103.
Filip Velickovski, Luigi Ceccaroni, Josep Roca, Felip Burgos, Juan B Galdiz, Nuria Marina (2014). Clinical
Decision Support Systems (CDSS) for preventive management of COPD patients. Journal of Translational
Medicine 2014, 12(Suppl 2).1-10.
Garg, A. X., Adhikari, N. K. J., McDonald, H., Rosas-Arellano, M. P., Devereaux, P. J., Beyene, J., Sam, J., et al.
(2005). Effects of computerized clinical decision support systems on practitioner performance and patient
outcomes: a systematic review. JAMA. The Journal Of The American Medical Association, 293(10), 1223-
1238.
Halim Ceylan (2006). Developing Combined Genetic Algorithm – Hill Climbing Optimization Method for Area
Traffic Control. Journal of Transportation Engineering (2006)132:8(663): 1-40. doi: 10.1061/(ASCE)0733-
947X.
Hannes Alder, Beat A. Michel, Christian Marx, Giorgio Tamborrini, Thomas Langenegger, Pius Bruehlmann,
Johann Steurer, Lukas M. Wildi (2014). Computer-Based Diagnostic Expert Systems in Rheumatology:
Where Do We Stand in 2014?. Hindawi Publishing Corporation International Journal of Rheumatology
Volume 2014, Article ID 672714: 1-10. doi.10.1155/2014/672714.
Harry Henderson (2009). Encyclopedia of Computer science and technology. Revised Edition. Facts On File, Inc.
New York. USA: 187.
Imran Sarwar Bajwa, Bushra Ramzan, Shabana Ramzan (2017). Markov Logic Based Inference Engine for CDSS.
Mehran University Research Journal of Engineering & Technology, Volume 36, No. 1. January, 2017.
Jackson, P (1999). Introduction Expert Systems. Harlow, England: Addison Wesley Longman. Third Edition.
Karagiannis S., Dounis A., Chalastras T., Tiropanis P., and Papachristos D. (2006). Design of Expert System for
Search Allergy and Selection of the Skin tests using clips. International Journal of Information Technology.
Lawrence A. Tomei (2008). Encyclopedia of Information Technology Curriculum Integration. InformatIon
ScIence reference. New York. USA: 316.
Marin Fotache, Georgiana Olaru, Viorel Iacoban (2015). Information systems analysts and
designers.Academic/research supply vs market demand. Procedia Economics and Finance 20 ( 2015 ): 232 –
242.
Mohamed Khalifa, MD (2014). Clinical Decision Support: Strategies for Success. Procedia Computer Science 37
(2014): 422 – 427.
NC Ngene, N Amin, J Moodley (2015). Ruptured subcapsular hematoma of the liver due to pre- eclampsia
presenting as interstitial pregnancy and the role of intra- abdominal packing. Nigerian Journal of Clinical
Practice • Mar-Apr 2015 • Vol 18 • Issue 2: 300-303. doi: 10.4103/1119-3077.151074.
Oscar Araque, Ignacio Corcuera-Platas, J.Fernando Sánchez-Rada, CarlosA. Iglesias (2017). Enhancing deep
learning sentiment analysis with ensemble techniques in social applications. Expert Systems With
Applications 77 (2017): 236–246.
P.Isakki alias Devi and Dr.S.P.Rajagopalan (2011), The Expert System Designed to Improve Custome Satisfaction.
Advanced Computing: An International Journal (ACIJ), Vol.2, No.6, November 2011: 69-84. Doi:
10.5121/acij.2011.2607.
Pengli Jia, Longhao Zhang, Jingjing Chen, Pujing Zhao, Mingming Zhang (2016). The Effects of Clinical Decision
Support Systems on Medication Safety: An Overview. PLOS ONE, doi:10.1371/journal.pone.0167683,
December 15, 2016. 1-17.
Piljae Kim, Satoru Nakamura, Daisuke Kurabayashi (2014). Hill-Climbing for a Noisy Potential Field Using
Information Entropy. PALADYN Journal of Behavioral Robotics, Volume 2, Issue 2: 94-99. doi:
10.2478/s13230-011-0020-x.
R.A. Soltan, M. Z. Rashad, B.El-Desouky (2013). Diagnosis of Some Diseases in Medicine via computerized
Experts System. International Journal of Computer Science & Information Technology (IJCSIT). Vol 5, No 5,
October 2013: 79-90. doi: 10.5121/ijcsit.2013.5505.
Ramakrishnan S (2012). System Analysis and Design. Journal of Information Technology & Software Engineering.
1-2. doi; 10.4172/2165-7866.S8-e001.
Richard T. Blankley, Christal Fisher, Melissa Westwood, Robyn North, Philip N. Baker, Michael J. Walker, Andrew
Williamson, Anthony D. Whetton, Wanchang Lin, Lesley McCowan_, Claire T. Roberts, Garth J. S. Cooper,
Richard D. Unwin, Jenny E. Myers (2013). A Label-free Selected Reaction Monitoring Workflow Identifies a
Subset of Pregnancy Specific Glycoproteins as Potential Predictive Markers of Early-onset Preeclampsia.
Molecular & Cellular Proteomics 12.11. The American Society for Biochemistry and Molecular Biology, Inc.
3148- 3159.
Sheldon H. Jacobson (2004). Analyzing the Performance of Generalized Hill Climbing Algorithms. Journal of
Heuristics, 10: 387–405.
Sujata Wangkheimayum, S. K. V. S., 2011. Effect of Vitamin E on sP-Selectin Levels in Pre-Eclampsia. Ind J
Clin Biochem, 26(2), p. 169–171.

Tomei, L. A., 2008. Encyclopedia of Information Technology Curriculum Integration. 1st ed. New York:
InformatIon ScIence Reference.

Wah, B. W., 2008. Wiley Encyclopedia of Computer Science and Engineering. 1st ed. New York: Wiley-
Interscience

You might also like