Optimization of Certainty Factor Method to Detect Preeclampsia in

Women Pregnant

Linda Perdana Wanti

, Nur Wachid Adi Prasetya

, Laura Sari

and Lina Puspitasari

Department of Informatics, Politeknik Negeri Cilacap, Cilacap, Indonesia

Department of Midwifery, STIKES Graha Mandiri, Cilacap, Indonesia

Keywords: Certainty Factor Method, Optimization, Preeclampsia, Pregnant Women, Expert System.

Abstract: Preeclampsia is a hypertensive disorder in pregnant women that significantly affects morbidity and one of the

causes of death in pregnant women and fetuses. According to WHO data, the prevalence of preeclampsia is

1.8-18% in developing countries, while in developed countries it is 1.3-6%. This value indicates that in

developing countries the case of pregnant women with preeclampsia is higher than in developed countries

because preventive treatment for pregnant women with preeclampsia is handled faster in developed countries

than in developing countries. In Indonesia alone, the Maternal Mortality Ratio (MMR) for the last 10 years

amounted to 459 maternal and fetal deaths from 100,000 births with the frequency of occurrence of

preeclampsia around 3% to 10% of the total number of pregnancies. The purpose of this study is the early

detection of preeclampsia in pregnant women through diagnosis carried out using the Expert system so that

pregnant women get preventive measures as an early prevention of preeclampsia disease that attacks and

reduces the MMR rate. Another goal is to provide recommendations for therapy that can be given to pregnant

women with preeclampsia. This research uses an optimized certainty factor method so as to provide a certainty

value about the Expert's statement through the Expert's confidence level which is symbolized by a number in

the range -1 to 1 in diagnosing preeclampsia. The output of this study is the result of expert diagnosis by

optimizing the certainty factor method which provides recommendations for preventive actions to be taken to

pregnant women with preeclampsia.

1 INTRODUCTION

Maternal Mortality Ratio (MMR) according to the

World Health Organization (WHO) is the incidence

of death in pregnant women during the period around

delivery, which is the period of 42 days after the end

of pregnancy, which is caused by all causes related to

pregnancy or incorrect handling and is not caused by

injury or accident. (Macedo et al., 2020). Maternal

Mortality Ratio (MMR) and Infant Mortality Ratio

(IMR) are benchmarks for the health and welfare of

the people in a country. WHO reports from various

sources that the direct cause of maternal death occurs

during and after childbirth caused by bleeding,

infection or high blood pressure during pregnancy by

75% (Wang et al., 2020). In Indonesia itself as a

https://orcid.org/0000-0002-6679-2560

https://orcid.org/0000-0002-4598-4336

https://orcid.org/0000-0002-3495-2558

https://orcid.org/0000-0001-7348-6564

developing country, the MMR is still quite high, data

from the Inter-Census Population Survey (SUPAS)

recorded MMR as many as 305 cases during the

period 2011 to 2014, which means that there are 305

cases of maternal death caused by pregnancy until

delivery. 42 days after delivery per 100,000 live

births (Aini, Fajaria Nur; Widyawati, Melyana Nurul,

Santor, 2019). In Cilacap Regency, according to data

from the Cilacap Regency Health Office, it shows that

during the 2019 period of MMR there were 15 cases

while for IMR there were 155 cases. As for the

maximum target of the Cilacap District Medium-

Term Development Plan (RPJMD), MMR is 19 cases

and IMR is 139 cases (Tri Budiarti, Dhiah Dwi

Kusumawati., Nikmah Nuur Rochmah, 2019).

Wanti, L., Prasetya, N., Sari, L. and Puspitasari, L.

Optimization of Certainty Factor Method to Detect Preeclampsia in Women Pregnant.

DOI: 10.5220/0010941400003260

In Proceedings of the 4th International Conference on Applied Science and Technology on Engineering Science (iCAST-ES 2021), pages 147-155

ISBN: 978-989-758-615-6; ISSN: 2975-8246

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

147

Based on this target, the MMR in Cilacap Regency is

still quite high even though it is still below the

maximum standard set. This has become the concern

of relevant institutions in Cilacap Regency to

continue to suppress MMR and IMR so that the level

of community welfare increases.

MMR can be identified based on the general

condition of the mother during a gestation period

lasting 40 weeks (Qiao et al., 2020). One of the

identifications can be done through the health

examination of pregnant women in available health

facilities (Gustri et al., 2016). This identification

serves to reduce the risk of death of pregnant women

and fetuses that can be predicted based on the

symptoms experienced by pregnant women during

pregnancy through prompt and correct handling in the

most dangerous period, namely the period around

delivery (Saraswati & Mardina, 2016). An expert

system can be interpreted simply as a transfer of

knowledge from an expert to a computer through an

information system that can be utilized without the

time and place restrictions (Mathew et al., 2020). The

Expert System asks for facts that will later be used as

knowledge inference which is then processed to be

able to provide conclusions or decisions that are

conical to a result of these facts (Dai et al., 2019). The

conclusion is considered as the result of consultation

with the Expert, who provides non-expert advice, and

explains the possible solutions for the consequences

(Zieschang et al., 2019). In research conducted, the

Expert system is used to provide recommendations

for therapy that can be carried out by pregnant women

with preeclampsia during pregnancy as a decision or

conclusion based on the symptoms that are inputted

into the Expert system which is processed through a

knowledge base. The conclusions/decisions given are

non-experts, which can later be consulted with real

experts if there are doubts about the results given by

the Expert system. The therapy recommended by a

machine learning-based intelligent system is in the

form of an active solution for the decisions/

conclusions given to pregnant women with

preeclampsia.

Research on preeclampsia has been conducted by

Macedo et al who examined the prevalence of

preeclampsia and eclampsia in pregnancy on 291,247

adolescents worldwide since 1969. The results

showed 70 studies starting from 1969 to 2019 and

consisting of 30 countries with 291,247 adolescents

with prevalence rates the overall

preeclampsia/eclampsia was 6.7%. Subgroup

analysis revealed that the preeclampsia/eclampsia

relationship was influenced by country income and

the highest prevalence was found in the low- and

middle-income countries group with prevalence

values of 11.5% and 10.6% (Macedo et al., 2020).

Furthermore, research conducted by Zain et al

analyzed the certainty factor method and

implemented it into an Expert system for early

detection of disease in beef cattle. In this study, the

certainty factor method was used to determine the

level of confidence in the disease found in beef cattle

and the forward chaining method was used to

determine the search on the knowledge base that

determined the conclusion of the disease in beef

cattle. The output of this study is a mobile-based

Expert system for early detection of disease in beef

cattle so that control measures can be taken quickly

and precisely in determining the quality of good beef

cattle (Zain & Astutik, 2015).

Further research by Khairina et al applied the

certainty factor to the Expert system for diagnosing

ENT diseases. The expert in this study is an ENT

specialist who provides complete and detailed

information about the causes and symptoms

experienced by patients who have problems with their

ears, nose and throat. The results of this study are a

website-based information system that can diagnose

ENT diseases by selecting the symptoms experienced

by patients and the system provides search results in

the form of information about ENT diseases suffered

based on the selected symptoms (Setyaputri et al.,

2018).

Furthermore, research conducted by Yudia et all

which examines the determinants of preeclampsia in

RSUP Dr. Moch. Husein Palembang. This research

used a case-control study design. The data used

secondary data from mothers medical record with

preeclampsia and without preeclampsia in the period

1 January 2015-31 December 2015. The number of

samples was 85 cases and controls. Analysis of data

used univariate, bivariate analysis used chi square

test, and multivariate used multiple logistic regression

prediction model. Multivariate analysis showed that

factors associated with preeclampsia were age> 35

years (OR: 4.120; 95% CI: 1.715 to 9.897) obesity

(OR: 2.134; 95% CI: 1.093 to 4.167) and a history of

hypertension (OR: 12.143; 95% CI: 1.368 to

107.792). The most dominant factor related to the

incidence of preeclampsia in pregnant women is a

history of hypertension. The advice can be given that

the relevant agencies to improve promotive and

preventive efforts by providing socialization of the

factors which may be a risk of preeclampsia so that

cases of preeclampsia can be prevented at an early

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

148

stage (Gustri et al., 2016).

Further research by Retno et al which analyzed the

risk factors of severe preeclampsia in pregnant

women in RSUD Dr. Moewardi Surakarta. Causes of

maternal mortality include hemorrhage, eclampsia,

and infection. Pre-eclampsia is a unique disorder that

is only found in human pregnancy. Pre-eclampsia

usually occurs in the third trimester. Pre-eclampsia at

Dr. Moewardi hospital in 2010-2011 had increased.

The aim of this study was to determine the risk factors

of preeclampsia in pregnant women in Dr.Moewardi

hospital. This study was an observational study with

case-control design. Technique sampling used was

consecutive sampling. The data were analyzed

through Chi Square test. The results of this study

showed that there were correlations between

gestational age (p = 0.001; OR = 16.125, 95% CI =

1.993 to 130.459), occupational status (p = 0.001; OR

= 4.173, 95% CI = 1.709 to 10.188) and the incidence

of severe preeclampsia in pregnant women, and there

was not any relationship between history of diabetes

mel- litus (p = 1.000; OR = 1.000, 95% CI = 0.061 to

16.508) and the incidence of severe preeclampsia in

pregnant women (Wulandari, 2012).

The novelty of the research carried out with

previous research is the optimization of the certainty

factor method adapted to provide a confidence value

to the results of the Expert system search provided by

the Expert. The search results are used to detect early

and diagnose symptoms of preeclampsia in pregnant

women by providing recommendations for therapy

that should be carried out by the family and pregnant

women in order to prevent unwanted things such as

the death of pregnant women and fetuses. The output

of this study is a conclusion of preeclampsia by

optimizing the certainty factor method which

describes the certainty value of an expert's conclusion

on the diagnosis of preeclampsia in pregnant women

with details of the stage of preeclampsia and

recommendations for therapy that can be taken as a

preventive measure to suppress the Maternal

Mortality Ratio so that the results of the

recommendations can prevent the death of pregnant

women as early as possible.

2 RESEARCH METHOD

2.1 Preeclampsia

Preeclampsia is a disease in pregnant women

characterized by hypertension, oedema, and

proteinuria that arise due to pregnancy (Bracken et

al., 2021). Preeclampsia is generally detected in the

third trimester of pregnancy or in pregnant women

who have a history of hypertension, so this disease

can occur in the previous trimester (Aguilar-Cordero

et al., 2020). The emergence of hypertension usually

precedes other symptoms and in cases of

preeclampsia, there are two types of hypertensive

disorders, namely systolic and diastolic (Qiao et al.,

2020). Excessive accumulation of body fluids in body

tissues that can be identified through swelling of

certain body parts such as fingers, curries and face

and weight gain is called oedema (Macedo et al.,

2020). Proteinuria is the level of protein

concentration in urine that reaches 0.3 g/liter of urine

within 24 hours and usually appears last compared to

the other two symptoms (Wang et al., 2020), (Gustri

et al., 2016).

For pregnant women with preeclampsia, mild

preeclampsia syndrome, and symptoms as above are

often missed and not monitored by the pregnant

woman and her family, which results in severe

preeclampsia and leads to the death of pregnant

women (Saraswati & Mardina, 2016). Therefore,

early detection of preeclampsia and appropriate

therapy as a preventive measure need to be

implemented immediately (Wulandari, 2012).

Preeclampsia is grouped into 2, namely mild and

severe preeclampsia, whigich is shown through the

following symptoms (Kurniasari et al., 2015):

a. For mild preeclampsia, an examination every 6

hours showed a systolic blood pressure of 140 mm

Hg or an increase of 30 mm Hg and a diastolic

blood pressure result of 90 mm Hg or an increase

of 15 mm Hg. The weight of pregnant women

increased significantly by 1 kg in a period of 1

week continuously and the protein concentration

of 0.3 g or more in 1 litre of urine.

b. Severe preeclampsia blood pressure in pregnant

women of 160/110 mm Hg or more. The increase

in protein concentration reaches 3 g / litre of urine

or more. Pregnant women experience pain in the

epigastrium, decreased visual function,

headaches, decreased conscious function,

accumulation of fluid in the lungs and cyanosis.

After further examination, the results of the

examination showed increased liver enzyme

levels accompanied by an increase in bilirubin

levels, retinal bleeding occurred, and platelet

levels <100,000/mm.

In severe preeclampsia, if these symptoms continue

to increase, it will result in maternal death, namely

Optimization of Certainty Factor Method to Detect Preeclampsia in Women Pregnant

149

Figure 1: Expert System Environment.

pregnant women and fetuses (Aini, Fajaria Nur;

Widyawati, Melyana Nurul, Santor, 2019).

2.2 Expert System

An expert system can be interpreted as a computer

program that functions as a problem solver whose use

can reach a performance level equivalent to a human

expert/expert in a special field to narrow the

discussion area and provide decision recommen-

dations/conclusions from the problems encountered

(Rada, 2008), (Wanti & Ramadan, 2020). Expert

systems require the input of facts that occur to be

processed in an inference engine and produce output

in the form of a special conclusion (Zieschang et al.,

2019). The structure of the Expert system is divided

into two groups of environments, namely the

development environment which is used for the

development of the Expert system from the

knowledge base and the needs of both functional and

non-functional systems and the consulting

environment used by a non-expert or end-user to

conduct consultations (Castelli et al., 2017), (Zasada

et al., 2017).

The Expert System consists of several

components, including:

a. Knowledge base and knowledge acquisition are

two different things. The knowledge base contains

a collection of knowledge and information from a

human expert on a particular skill, while

knowledge acquisition is a change in the expertise

of an Expert/expert from a knowledge base into a

computer system (Zhang et al., 2013).

b. The user interface is used to interact between the

intelligent system and the end-user. This

component is found in the consulting environment

where the inference machine on the system gets

input in the form of facts/symptoms from the user,

and the system produces output in the form of

information on the conclusion of a problem for the

user (Liao, 2005).

c. The inference machine is the control centre for a

machine learning-based intelligent system in

which there is a mechanism for the Expert

thinking function and the intelligent system

reasoning role used by the expert/Expert and can

draw conclusions and control the mechanism of

the intelligent system being built (Mathew et al.,

2020).

d. Workplace is located in the development

environment or can be called an area in computer

memory and serves as a database for the process

of recording the results of a temporary diagnosis

which will later be used as a

decision/recommendation on a problem (Ooi &

Tan, 2016).

e. Explanation expert system serves to improve the

performance of intelligent systems by describing

the flow of reasoning carried out to end-users (Liu

et al., 2020).

f. Improved knowledge is used by Experts who have

the expertise to solve problems through analysis

of facts/symptoms that occur and improve the

performance of intelligent systems through

improving the performance of Expert system

performance (Wanti et al., 2019).

The advantage of expert systems compared to

conventional systems is that the algorithms in the

inference machine are not written in the program

source code, only stored as a knowledge base for the

intelligent system that was built (Dweiri et al., 2017),

(Dai et al., 2019).

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

150

2.3 Certainty Factor Method

Certainty theory underlies the use of certainty factors

that cover the value of certainty in a problem based

on facts/symptoms that occur in the field based on the

perception of an expert (Aji et al., 2018). The range

of values to describe the level of confidence of an

expert is from a negative one (sure it doesn't happen)

to one (sure it happens) (Zain & Astutik, 2015). Here

is a table of confidence levels to provide a certainty

factor value for a fact/symptom:

Table 1: Certainty Factor Uncertain Term.

Uncertain Term Certainty Factor Value

Definitely Not 0.2

Almost Certainty Not 0.3

Probably Not 0.4

Maybe Not 0.5

Unknown 0.6

Maybe 0.7

Probably 0.8

Almost Certainty 0.9

Definitely 1.0

Users who do not experience symptoms are

indicated by a CF value of 0, while for users who

experience symptoms it can be predicted according to

the selected CF value which shows the percentage for

a symptom with a high confidence value experienced

by the user (Riadi, 2017). Determination of uncertain

terms is based on expert knowledge after consultation

to describe each symptom/actual fact that will be

given an uncertain term value (Setyaputri et al.,

2018). Certainty Factor (CF) accumulates the value

of the degree of confidence and the value of the

degree of distrust which is denoted by NP(P|E) to

describe the confidence value of hypothesis P,

evidence E and NK (P|E) to describe the value of

distrust of hypothesis P, evidence E, where the sum

of the two is not always 1 (Arifin et al., 2017).

The stages of the process on the certainty factor

method are as follows:

a. Determining the value of CF

𝐶𝐹



𝐻, 𝐸



= 𝑀𝐵



𝐻, 𝐸



−𝑀𝐷 [𝐻, 𝐸] (1)

Information :

CF [H,E] : a measure of the certainty of the

hypothesis H that affected by E

symptoms

MB [H,E] : a measure of MB's confidence in H

affected by E

MD [H,E] : measure of distrust MD to H which

is affected by E

b. Determine the value of CF Combination

determined by one premise

𝐶𝐹

[

𝐻, 𝐸

]

= 𝐶𝐹

[

𝐸

]

∗𝐶𝐹

[

𝑅𝑈𝐿𝐸

]

= 𝐶𝐹

[

𝑈𝑆𝐸𝑅

]

∗

𝐶𝐹 [𝐸𝑋𝑃𝐸𝑅𝑇] (2)

c. Determine the value of CF Combination

determined by more than one premise

𝐶

𝐹

[

𝑋∧𝑌

]

= 𝑀𝑖𝑛



𝐶𝐹

[

𝑥

]

, 𝐶𝐹

[

𝑦

]



∗𝐶𝐹

[

𝑅𝑈𝐿𝐸



(3)

𝐶𝐹

[

𝑋∨𝑌

]

= 𝑀𝑎𝑥



𝐶𝐹

[

𝑥

]

, 𝐶𝐹

[

𝑦

]



∗𝐶𝐹 [𝑅𝑈𝐿𝐸}

(4)

d. Determine the CF value for the same conclusion

𝐶𝐹 𝐶𝑜𝑚𝑏𝑖𝑛𝑒𝑑

[

𝐶𝐹1, 𝐶𝐹2

]

= 𝐶𝐹1+𝐶𝐹2 ∗ (1 −

𝐶𝐹1) (5)

The certainty factor method is widely adapted for a

machine learning-based intelligent system because

the measurement of a definite or uncertain

hypothesis, such as early detection of a disease based

on the symptoms shown and the calculation for the

certainty factor method is only one calculation

process and the minimum data that can be obtained

processed as much as two data to ensure the accuracy

of the results obtained (Aji et al., 2018).

3 RESULT AND ANALYSIS

The data used to diagnose preeclampsia in pregnant

women is data on factors that influence the cause of

the emergence of preeclampsia obtained from the

results of consultations with experts to determine the

value of MB (Measure of Believe) and MD (Measure

of Disbelieve) which were formulated with

developers to build a knowledge base system.

Experts. Determination of the classification class to

raise the chance of preeclampsia disease which is

classified into three categories, namely mild

preeclampsia, moderate preeclampsia and severe

preeclampsia that attacks pregnant women. The stage

begins with collecting data and starting to analyze all

the resources needed for the preeclampsia diagnosis

process. The parity variable data is shown in table 2

and the data will be used to classify preeclampsia into

three categories. The experts in this study were a

midwife and an obstetrician specialist.

Optimization of Certainty Factor Method to Detect Preeclampsia in Women Pregnant

151

Table 2: Data on Preeclampsia Causing Factors.

Factor

Code

Information Factor

Description

F01 A

e U1, U2, U3

F02 Parit

P1, P2

F03 Pre

nanc

Distance JK1, JK2

F04 Multiple Pregnancy KG1, KG2

F05 History of Preeclampsia RP1, RP2

F06 History of Hypertension RH1, RH2

F07 Descendants Histor

RK1, RK2

F08 Histor

of DM RD1, RD2

F09 Nutritional status SG1, SG2

F10 Antenatal Care AC1, AC2

F11 Family Planning Acceptor Histor

RA1, RA2

F12 Educational status SP1, SP2

F13 Knowled

e P1, P2, P3

F14 Economic Status SE1, SE2

F15 Wor

PK1, PK2

F16 Health Service Distance J1, J2, J3

Table 3: Expert Interpretation.

Factor

Description

Code

Information CF

User

Expert

U1 <= 18

ears 0.8 0.8

U2 18 - 38

ears 0.6 0.6

U3 >= 38

ears 0.8 0.9

P1 First 0.7 0.8

P2 Second/more 0.6 0.6

JK1 < 24 months 0.8 0.9

JK2 >/ 24 months 0.7 0.7

KG1 Double 0.8 0.9

KG2 Single 0.7 0.7

RP1 There is 0.8 0.9

RP2 No 0.7 0.6

RH1 There is 0.8 0.9

RH2 No 0.7 0.6

RK1 There is 0.8 0.9

RK2 No 0.7 0.6

RD1 There is 0.8 0.9

RD2 No 0.7 0.6

SG1 Obesit

0.8 0.9

SG2 No 0.7 0.6

AC1 </= 3 times 0.8 0.9

AC2 > 3 times 0.7 0.6

RA1 There is 0.7 0.6

RA2 No 0.8 0.9

SP1 Elementary/junior high

school

0.8 0.8

SP2 Senior Hi

h School

PT 0.7 0.7

P1 Not enou

h 0.8 0.9

P2 Currentl

0.7 0.8

P3 Well 0.7 0.7

SE1 <500

0.8 0.9

SE2 >/= 500

0.7 0.6

PK1 Unem

ment 0.9 0.9

PK2 Wor

0.7 0.6

J1 >1000 meters 0.8 0.9

J2 </= 1000 meters 0.6 0.6

Table 4: Preeclampsia Category Data.

Category Code Information

P01 Mild Preeclam

sia

P02 Moderate Preeclam

sia

P03 Severe Preeclampsia

Table 5: Rule.

No Rule

1 If U1 and P1 and JK2

2 If U2 and RP1 and RH2

3 If U3 and RH1 and RD1

4 If U2 and SG1 and RH1

5 If U3 and SG2 and RD2

An example of the calculation process using the

certainty factor method by determining the CF value

using equation (1), namely for CF Users and CF

Experts is available in table 3. Then determining the

CF Combination value determined by one premise

using equation (2), is as follows:

Rule : If U3 and RH1 and RD1

CF S1 : CF User * CF Expert

: 0.8 * 0.9

: 0.72

CF S2 : CF User * CF Expert

: 0.8 * 0.9

: 0.72

CF S3 : CF User * CF Expert

: 0.8 * 0.9

: 0.72

The next step is to determine the CF for the same

conclusion. Because there is more than one symptom

experienced by patient A, we use equation (5), the

calculation is as follows:

CF C1 : CF S1 + CF S2 * (1-CF 1)

: 0.72 + 0.72 * (1-0.72)

: 0.72 + 0.72 * 0.28

: 0.72 + 0.2019

: 0.92

CF C2 : CF C1 + CF S3 * (1-CF C1)

: 0.92 + 0.72 * (1-0.92)

: 0.72 + 0.72 * 0.08

: 0.72 + 0.0576

: 0.9776

The CF C2 value of 0.9776 was obtained which is

the CF diagnosis of preeclampsia experienced by

patient A. Then to determine the percentage of

confidence in preeclampsia which is included in the

category of severe preeclampsia, using the equation:

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

152

Percentage : CF Disease * 100

: 0.9776*100

: 97.76%

Based on the calculation by taking a sample of patient

A, the information on the level of confidence based

on the final percentage is very possible.

For the second example, the calculation process

will be carried out using the certainty factor method

by determining the CF value in patient B using

equation (1). Values for CF User and CF Expert are

available in table 3. The first step is to determine the

CF Combination value determined by one premise

using equation (2), as follows:

Rule : If U2 and SG1 and RH1

CF S1 : CF User * CF Expert

: 0.6 * 0.6

: 0.36

CF S2 : CF User * CF Expert

: 0.8 * 0.9

: 0.72

CF S3 : CF User * CF Expert

: 0.8 * 0.9

: 0.72

The next step is to determine the CF for the same

conclusion. Because there is more than one symptom

experienced by patient B, we use equation (5). The

calculation is as follows:

CF C1 : CF S1 + CF S2 * (1-CF S1)

: 0.36 + 0.72 * (1-0.36)

: 0.72 + 0.72 * 0.64

: 0.72 + 0.461

: 0.821

CF C2 : CF C1 + CF S3 * (1-CF C1)

: 0.821 + 0.72 * (1-0.821)

: 0.72 + 0.72 * 0.179

: 0.72 + 0.13

: 0.951

The CF C2 value of 0.9776 was obtained which is

the CF diagnosis of preeclampsia experienced by

patient B. Then to determine the percentage of

confidence in preeclampsia which is included in the

category of severe preeclampsia, using the equation:

Percentage : CF Disease * 100

: 0.951*100

: 95.1 %

Based on the calculation by taking a sample of

patient B, the information on the level of confidence

based on the final percentage is very possible.

Therapeutic recommendations that can be done to

patient A and patient B who are diagnosed with

severe preeclampsia are to periodically check the

general condition of the mother and the baby in her

womb and monitor the pregnancy process until the

birth process to obtain measures that prevent maternal

and infant mortality. This process can be carried out

by medical personnel either in integrated health

services such as Posyandu, Puskesmas or the nearest

hospital.

4 CONCLUSIONS

The optimization of the certainty factor method to

determine the value of expert certainty on the

condition of a patient who has the factors causing

preeclampsia has been proven. Based on the

calculation simulation using the certainty factor

method, the results showed that the percentage of

patient A was 97.76%, while patient B was 95.1% and

both patients were very likely to be included in the

category of severe preeclampsia with each of the

causative factors accompanying the two patients.

These results can be used by medical personnel and

families to determine the most likely preventive

action in the process of pregnancy to the birth of a

baby to prevent maternally and infant mortality.

REFERENCES

Aguilar-Cordero, M. J., Lasserrot-Cuadrado, A., Mur-

Villar, N., León-Ríos, X. A., Rivero-Blanco, T., &

Pérez-Castillo, I. M. (2020). Vitamin D, preeclampsia

and prematurity: A systematic review and meta-analysis

of observational and interventional studies. Midwifery,

87, 102707. https://doi.org/10.1016/j.midw.2020.

102707

Aini, Fajaria Nur; Widyawati, Melyana Nurul, Santor, B.

(2019). Diagnosa Preeklamsia Pada Ibu Hamil

Menggunakan Sistem Informasi Berbasis Web. Jurnal

Keperawatan Silampari, 2(2), 18–27.

Aji, A. H., Furqon, M. T., & Widodo, A. W. (2018). Sistem

Expert Diagnosa Penyakit Ibu Hamil Menggunakan

Metode Certainty Factor ( CF ). Jurnal Pengembangan

Teknologi Informasi Dan Ilmu Komputer, 2(5), 2127–

2134. http://j-ptiik.ub.ac.id/index.php/j-ptiik/article/v

iew/1556

Arifin, M., Slamin, S., & Retnani, W. E. Y. (2017).

Penerapan Metode Certainty Factor Untuk Sistem Expert

Diagnosis Hama Dan Penyakit Pada Tanaman

Tembakau. Berkala Sainstek, 5(1), 21. https://doi.org/

10.19184/bst.v5i1.5370

Bracken, H., Buhimschi, I. A., Rahman, A., Smith, P. R. S.,

Pervin, J., Rouf, S., Bousieguez, M., López, L. G.,

Buhimschi, C. S., Easterling, T., & Winikoff, B. (2021).

Congo red test for identification of preeclampsia:

Results of a prospective diagnostic case-control study

in Bangladesh and Mexico. EClinicalMedicine, 31.

https://doi.org/10.1016/j.eclinm.2020.100678

Optimization of Certainty Factor Method to Detect Preeclampsia in Women Pregnant

153

Castelli, M., Manzoni, L., Vanneschi, L., & Popovič, A.

(2017). An expert system for extracting knowledge

from customers’ reviews: The case of Amazon.com,

Inc. Expert Systems with Applications, 84, 117–126.

https://doi.org/10.1016/j.eswa.2017.05.008

Dai, S., Xu, B., Shi, G., Liu, J., Zhang, Z., Shi, X., & Qiao,

Y. (2019). SeDeM expert system for directly

compressed tablet formulation: A review and new

perspectives. Powder Technology, 342, 517–527.

https:// doi.org/10.1016/j.powtec.2018.10.027

Dweiri, F., Kumar, S., Ahmed, S., & Jain, V. (2017).

Corrigendum to “ Designing an integrate d AHP base d

decision support system for supplier selection in

automotive industry ” Expert Systems. Expert Systems

With Applications, 72, 467–468. https://doi.org/

10.1016/j.eswa.2016.12.025

Gustri, Y., Januar Sitorus, R., & Utama, F. (2016).

Determinants Preeclampsia in Pregnancy At Rsup Dr.

Mohammad Hoesin Palembang. Jurnal Ilmu Kesehatan

Masyarakat, 7(3), 209–217. https://doi.org/10.

26553/jikm.2016.7.3.209-217

Kurniasari, D., JURNAL, F. A.-H., & 2015, undefined.

(2015). Hubungan Usia, Paritas Dan Diabetes Mellitus

Pada Kehamilan Dengan Kejadian Preeklamsia Pada Ibu

Hamil Di Wilayah Kerja Puskesmas Rumbia Kabupaten.

Ejurnalmalahayati.Ac.Id, 9(3), 142–150. http://ejurnal

malahayati.ac.id/index.php/holistik/article/view/232

Liao, S. H. (2005). Expert system methodologies and

applications-a decade review from 1995 to 2004. Expert

Systems with Applications, 28(1), 93–103. https://doi.

org/10.1016/j.eswa.2004.08.003

Liu, Y., Eckert, C. M., & Earl, C. (2020). A review of fuzzy

AHP methods for decision-making with subjective

judgements. In Expert Systems with Applications (Vol.

161). Elsevier Ltd. https://doi.org/10.1016/j.eswa.2020.

113738

Macedo, T. C. C., Montagna, E., Trevisan, C. M., Zaia, V.,

de Oliveira, R., Barbosa, C. P., Laganà, A. S., & Bianco,

B. (2020). Prevalence of preeclampsia and eclampsia in

adolescent pregnancy: A systematic review and meta-

analysis of 291,247 adolescents worldwide since 1969.

European Journal of Obstetrics and Gynecology and

Reproductive Biology, 248(March), 177–186. https://

doi.org/10.1016/j.ejogrb.2020.03.043

Mathew, M., Chakrabortty, R. K., & Ryan, M. J. (2020). A

novel approach integrating AHP and TOPSIS under

spherical fuzzy sets for advanced manufacturing system

selection. Engineering Applications of Artificial

Intelligence, 96(October), 103988. https://doi.org/

10.1016/j.engappai.2020.103988

Ooi, K. B., & Tan, G. W. H. (2016). Mobile technology

acceptance model: An investigation using mobile users

to explore smartphone credit card. Expert Systems with

Applications, 59, 33–46. https://doi.org/10.10

16/j.eswa. 2016.04.015

Qiao, P., Zhao, Y., Jiang, X., Xu, C., Yang, Y., Bao, Y.,

Xie, H., & Ying, H. (2020). Impact of growth

discordance in twins on preeclampsia based on

chorionicity. American Journal of Obstetrics and

Gynecology, 223(4), 572.e1-572.e8. https://doi.org/

10.1016/j.ajog.2020.03.024

Rada, R. (2008). Expert systems and evolutionary

computing for financial investing: A review. Expert

Systems with Applications, 34(4), 2232–2240. https://

doi.org/10.1016/j.eswa.2007.05.012

Riadi, A. (2017). Penerapan Metode Certainty Factor Untuk

Sistem Expert Diagnosa Penyakit Diabetes Melitus

Pada Rsud Bumi Panua Kabupaten Pohuwato. ILKOM

Jurnal Ilmiah, 9(3), 309–316. https://doi.org

/10.33096/ilkom.v9i3.162.309-316

Saraswati, N., & Mardina. (2016). Unnes Journal of Public

Health Berdasarkan data World Health Organization

Berdasarkan laporan Dinas Kesehatan. Unnes Journal

of Public Health, 5(2), 90–99.

Setyaputri, K. E., Fadlil, A., & Sunardi, S. (2018). Analisis

Metode Certainty Factor pada Sistem Expert Diagnosa

Penyakit THT. Jurnal Teknik Elektro, 10(1), 30–35.

https://doi.org/10.15294/jte.v10i1.14031

Tri Budiarti, Dhiah Dwi Kusumawati., Nikmah Nuur

Rochmah. (2019). Hubungan Berat Bayi Lahir Dengan

Kematian Bayi. Jurnal Kesehatan Al-Irsyad, 12(2), 63–

70. https://doi.org/10.36746/jka.v12i2.42

Wang, Y., Wang, K., Han, T., Zhang, P., Chen, X., Wu, W.,

Feng, Y., Yang, H., Li, M., Xie, B., Guo, P., Warren, J.

L., Shi, X., Wang, S., & Zhang, Y. (2020). Exposure to

multiple metals and prevalence for preeclampsia in

Taiyuan, China. Environment International, 145

(August), https://doi.org/10.1016/j.envint.2020.106098

Wanti, L. P., Azroha, I. N., & Faiz, M. N. (2019).

Implementasi User Centered Design Pada Sistem Expert

Diagnosis Gangguan Perkembangan Motorik Kasar Pada

Anak Usia Dini. Media Aplikom, 11(1), 1–10.

Wanti, L. P., & Romadlon, S. (2020). Implementasi Forward

Chaining Method Pada Sistem Expert Untuk Deteksi

Dini Penyakit Ikan. Infotekmesin, 11(02), 74–79.

https://doi.org/10.35970/infotekmesin.v11i2.248

Wulandari, R. dan F. A. F. (2012). Faktor Risiko Kejadian

Preeklampsia Berat Pada Ibu Hamil di RSUD Dr

Moewardi Surakarta. Jurnal Kesehatan, 5(1), 29–35.

Zain, A. A., & Astutik, E. Z. (2015). ANALISIS METODE

Certainty Factor Dalam Sistem Expert Untuk

Mendeteksi Penyakit SAPI PEDAGING Kebutuhan

konsumsi sapi pedaging di Indonesia sangat tinggi

hampir setiap hari masyarakat Indonesia

mengkonsumsi daging sapi . Sapi juga merupakan

hewan yang banya.

Zasada, I., Piorr, A., Novo, P., Villanueva, A. J., &

Valánszki, I. (2017). What do we know about decision

support systems for landscape and environmental

management? A review and expert survey within EU

research projects. Environmental Modelling and

Software, 98, 63–74. https://doi.org/10.1016/j.envsoft.

2017.09.012

Zhang, J., Prater, E. L., & Lipkin, I. (2013). Feedback

reviews and bidding in online auctions: An integrated

hedonic regression and fuzzy logic expert system

approach. Decision Support Systems, 55(4), 894–902.

https://doi.org/10.1016/j.dss.2012.12.025

iCAST-ES 2021 - International Conference on Applied Science and Technology on Engineering Science

154

Zieschang, L., Klein, M., Jung, N., Krämer, J., &

Windbergs, M. (2019). Formulation development of

medicated chewing gum tablets by direct compression

using the SeDeM-Diagram-Expert-System. European

Journal of Pharmaceutics and Biopharmaceutics,

144(April), 68–78. https://doi.org/10.1016/j.ejpb.

2019.09.003

Optimization of Certainty Factor Method to Detect Preeclampsia in Women Pregnant

155