J. Inst. Eng. India Ser.

B (April 2022) 103(2):681–697

https://doi.org/10.1007/s40031-021-00644-z

REVIEW PAPER

Fuzzy Logic and Hybrid based Approaches for the Risk of Heart
Disease Detection: State-of-the-Art Review
Jagmohan Kaur1 • Baljit S. Khehra2

Received: 10 December 2020 / Accepted: 30 May 2021 / Published online: 2 August 2021
Ó The Institution of Engineers (India) 2021

Abstract Artificial Intelligence, Machine Learning, Fuzzy Introduction

Logic, Neural Network, Genetic Algorithm and their
hybrid systems play vital role in the medical sciences to The heart disease involves many risks like gender, age,
diagnose various diseases efficiently in the patients. The obesity, high blood pressure, high cholesterol, diabetes,
problems related to the heart are widely comon in today’s family history, drinking alcohol and smoke. Apart from
world. The risk of heart failure develops due to the nar- these, there are many other threats like underemployment
rowness and blockage in the coronary arteries of the heart industrialization, stress due to over work, physical inac-
as excess cholesterol deposits in the arteries and blood tivity, depression, stress, diet/routine changes and dis-
vessels that results in fatigue, chest pain, dyspnoea, tressed sleep have increased in this technical and modern
sleeping difficulties and depression. This research aims to world. In addition to this, change in life style, egoism,
explore diverse work done on FL and Hybrid-based tech- greed and narcissistic approach of people have triggered
niques to identify the risk of heart disease among the the occurrence of chronic diseases in patients.
patients. The present study reveals publications along with Heart diseases are categorized into distinct types like
the strength, operating system, accuracy rate and other congenital heart disease, coronary heart disease (CHD),
specifications used in the identification of heart disease arrhythmia, dilated cardiomyopathy, heart failure,
based on FL and Hybrid-based approaches since 2010. This myocardial infarction (MI), hypertropic cardiomyopathy
survey contributes motivation for research scholars to and mitral regurgitation. The patient having heart problem
generate more innovative ideas and continue their research may feel chest pain, faintness or coolness in arms and legs.
work in the respective field. Moreover, the future model for However, controlling blood pressure, eating a well-bal-
direct service of the patients from old age homes to the anced nutritious diet, regulating exercise regime, limited
Intensive Care Unit through ambulance services is also usage of alcohol, tobacco, smoking and gazettes can treat
presented in this paper. illness and malfunction of the heart.
The mortality rate of the heart patients is the highest in
Keywords Artificial intelligence (AI)  two countries namely: Africa and India, which is 34% and
Machine learning (ML)  Fuzzy logic (FL)  23%, respectively, and it is the lowest in France, Japan and
Neural network (NN)  Genetic algorithm  Hong Kong. Both developed and underdeveloped countries
Intensive care unit (ICU) have been facing this chronic disease since four decades
due to the excess consumption of alcohol, smoking and
tobacco. Nowadays, the most of the deaths occur due to
coronary heart disease. However, in some European
& Jagmohan Kaur
[email protected]
countries, death rate due to heart failure has declined as
they are followed by better health/education services and
1
IK Gujral Punjab Technical University, Jalandhar, Punjab, other preventive measures. According to the report of
India WHO [1], near 17.8 million older and adults face this
2
BBSB Engineering College, Fatehgarh Sahib, Punjab, India deadly disease. In fact, the death rate in adults is greater

123
682 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

than that of the aged people. Centers for Disease Control

and Prevention (CDC) [2] have announced that ratio of
death rate in Americans is 1:4 that means one out of every
four is dead due to heart stroke and 1 out of each 5 people
doesn’t know that they are affected by silent attack. One
more description from British Heart Foundation (BHF) [3]
revealed that the panic of pandemic COVID-19 (coron-
avirus disease) in the year 2020 has become the reason of
people dying with heart stroke due to psychological stress
and social distancing. All this has even affected the econ-
omy of the world as well as of other countries.
With the advancement of power of computer applica-
tions can make drastic changes in the medical field by
predicting disorder of tumors, lungs, thyroids and heart at
very early stage in the patients. Computer-aided heart
disease risk diagnose system has many features:
(a) Modern technique
(b) Time saving Fig. 1 Publications of fuzzy logic and hybrid system
(c) Independent from medical experts
(d) Minimize expenditure
(e) Minimal errors
(f) Less human effort
(g) Diminish mortality rate
This era of artificial intelligence comes up with another
invention made by the scholars of the Oxford University
[4], well known as ‘‘fingerprint’’ a biomarker named as
FRP (Fat Radiomic Profile) which is helpful in diagnosing
future stroke via scarring and inflammation of blood
vessels. The British Heart Foundation (BHF) and the
National Institute of Health Research (NIHR) are providing
funds for the same.

Research Methodology

The selection and collection of papers in journals and Fig. 2 Collection of research papers per year
international conferences have been chosen from state-of-
the-art and well-recognized publishers that ensures the and 3, respectively. The Table 1 reveals total number of
quality of this review paper. Total 53 papers are summa- publications of SCI Index.
rized in this paper, out of which 4 revealed the status of
heart disease patients in the world by WHO, BHF, NIHR
and CDC and rest 49 papers bring forward current Table of Abbreviations of Symptoms and Heart
approaches and algorithms of AI being used for the iden- Disease Database
tification of heart risk since 2010. In the Fig. 1, pie
chart shows 73% of area of publications is covered by The following Tables 2 and 3 give the information of
hybrid system and 27% belongs to FL. abbreviations used for symptoms and heart disease data-
Further, references of 21 IEEE, 13 Springer, 9 Elsevier base throughout the paper.
and 4 ACM papers are provided to bring forward the dis-
tinct techniques and their combinations being used in heart
disease detection along with their data bases, approaches,
accuracy and other specifications. Collection of research
papers per year and per publication is presented in Figs. 2

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 683

[8, 18, 20, 22, 23, 28, 41, 48, 51], K-means Clustering
Algorithm [22], Logistic Regression (LR) [23, 42], Mul-
tivariate Adaptive Regression Splines (MARS) [42], Rough
Set (RS) [3, 42], Support Vector Machine (SVM) [33, 34],
Random Forest (RF) [21, 23], K-Nearest Neighbourhood
(K-NN) [18, 30, 37, 46], Logistic Regression SVM (LR
SVM) algorithm in [23], has been summarized.

Review of Research Papers from 2010–2013 of FL

and Hybrid-Based Approaches

Adeli et al. [52] demonstrated the fuzzy expert system

(2010) using 11 input attributes from database of CCF,
LBF and V.A medical Center: AG, CP, CHL, MHR, RBS,
RECG, THR, BP, OP, GR, EXANG and 1 output attribute
with 5 features: sick (s1-s4) and healthy. Fuzzification was
structured in trapezoidal, triangular, R and L membership
Fig. 3 Collection of research papers per publisher
functions. Implementation of Mamdani inference system
was made with rule-based system of 44 rules. ‘COG’
Table 1 Total no. of publications of SCI index technique was used for defuzzification and yielded 94%
Publisher Research paper No. of researches
accuracy. Design of the system was implemented in
MATLAB. Explanation of hybrid fuzzy-evidential infer-
IEEE [5–25] 21 ence engine (2010) was made by Vahib Khatibi et al. [39]
Springer [26–38] 13 using fuzzy sets and Demster–Shafer theory of evidence.
Elsevier [39–47] 9 The hybrid system achieved 91.58% accuracy from the
ACM [3, 48, 49, 51] 4 data set of patients from V.A medical Center, CCF, HIC
and UHS.
An article (2011) by Anooj in [40] revealed the system
Review of FL and Hybrid-Based Approaches based on self-received information from the database.
for the Risk of Heart Disease Detection Extension in a clinical decision support system was made
by using triangular membership functions along with
A mathematical discipline in which the structure of human Mamdani fuzzy inference system in programming lan-
behavior in the form of uncertainties, true or false forms guage MATLAB 7.10. Defuzzification technique ‘COA’
the basis of FL. FL has following features: was utilized to achieve 62.35% of accuracy rate. CCF, V.A
(a) Fascinating area of research medical center and UHS were the main sources of data-
(b) Computational approach base. E. P. Ephzibah in the paper [26] presented a hybrid
(c) A matter of degrees of truth system (2011) based on GA and FL in MATLAB 7.3.0 for
(d) A limiting case of estimated reasoning the smooth work in healthcare field. CCF data set was used
(e) Performs decision-making with approximate values for the implementation of the work. Due to GA, 14 selected
attributes were reduced to 6 attributes: CP, RBP, EXANG,
The flow chart of algorithm of Fuzzy logic-based approach
MHR, OP and CA. Trapezium and triangular membership
for the risk of heart disease is shown in the Fig. 4. The
functions were used in the strategy of fuzzification.
combination of two or more than two technologies to find
Another development in the framework of fuzzy expert
the solution of the peculiar problem, overcome the
system (2012) was initiated by Bhuvaneswari Amma [5] to
deficiency of one technique and strengthen the system are
demonstrate an adaptable disease diagnosis system using
known as hybrid systems. In this study, hybridization of
hybridization of GA along with NN. It consisted of 13
techniques has been described as, FL [8, 9, 11, 12,
input variables AG, CP, GR, CHL, RBP, FBS, THS,
14, 15, 19, 20, 25–32, 35, 36, 38–41, 45, 48, 50, 52, 53],
RECG, MHR, EXANG, OP, CA, SLOPE and 1 output
GA [5, 7, 10, 14, 18, 26, 27, 36, 44], Artificial Neural
variable for the data base of UIC, Irvine. Accuracy rate was
Network (ANN) [5, 7, 10, 13, 35, 42–44, 51], Particle
obtained 94.17% which was evaluated through Root Mean
Swarm Optimization (PSO) [13, 33, 41], Ant Colony
Square Error. The system (2012) described by Ephzibah
Optimization (ACO) [37], Naive Bayes (NB)
et al. [27] made use of GA to select salient and 6 prime
[16, 18, 23, 29, 34], Decision Tree (DT)
features responsible for the rapid and error free diagnosis

123
684 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

Table 2 Table of abbreviations of symptoms

Sr. Symptoms Units Abbreviation Limit
No

1 Age AG 30–86
2 Chest pain (typical angina, atypical angina, non- anginal pain, CP Yes, no
asymptomatic)
3 Cholesterol mgndL CHL 125–200
4 Serum cholesterol mgndL Serum CHOL Less than 200
5 Low density lipoprotein mgndL LDL CHOL 18–232
6 High density lipoprotein mgndL HDL CHL 15–111
7 Resting blood pressurenBlood pressure mmnHg RBPnBP 75–120
8 Diastolic blood pressure mmnHg DBP 80-greater than 120
9 Systolic blood pressurenHigh blood pressure mmnHg SBPnHBP 130-greater than 180
10 Blood pressure mmnHg BP 90–190
11 ElecrocardiogramnResting elecrocardiogram ms ECGnR ECG 120–200
12 Maximum heart ratenMaximum heart rate achieved during bpm MHRnTHALACH Varies
exercise
13 Resting heart rate bpm Thal rest 60–100
14 Gender GR Yes, no
15 Old peak (ST depression) OP Variable
16 Slope of the peak-exercise ST-segment SLOPE Upsloping, flat, downsloping
17 Blood creatinine level mgndL BCL 0.5–2.2
18 Blood uric acid level mgndL BUAL 2.4–6.0 ( female), 3.4–7.0
(male)
19 Blood sodium level .125, .129, BSL 135–145
.141mEq/L
20 Height Feet HT
21 Weight Kg WT
22 Pulse rate ppm PR 50–110
23 Fasting blood sugar mgndL FBS 62–400
24 History of hypertension HTN Yes, no
25 Family history of coronary artery disease FAMHIST Yes, no
26 History of diabetes DM Yes, no
27 Number of major vessels colored by fluoroscopy CA 0–3
28 Exercise-induced Angina EXANG Yes, no
29 Obesity OBES Yes, no
30 Thallium scan THAL Normal, fixed defect, reversible
defect

of disease and then implemented FL. MATLAB was used The another system (2012) designed by Muthukaruppan
as programming language for the data base taken from UCI [41] was based on DT and PSO containing 13 input vari-
ML respository. Akhil jabbar et al. [6] revealed the ables: AG, CHOL, BP, GR, THALACH, CP, RECG, OP,
methodology (2012) of association rules which were gen- EXANG, SLOPE, THAL, CA and 1 output variable. The
erated by using hybrid feature subset selection. The paper phenomenon of ‘if-then’ rule in fuzzy rule-base was
consisted of mainly 11 input variables: AG, GR, THA- applied. The system was designed in MATLAB 7 and
LACH, LDL CHL, HDL CHL, CHOL, Rural/Urban, BP yielded accuracy of 93.27%.
Systolic, BP Diastolic and Serum Triglycerides. Approach The demonstration of another fuzzy expert system
was implemented in the database of Andra Pradesh which (2013) was presented by Sanjeev Kumar et al. [53]. It
yielded 95% accuracy. explained 6 input variables: CP, BS, BP, CHL, MHR, OP
and one is decision variable with 5 features. Mamdani

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 685

Table 3 Table of abbreviations of heart disease database

Sr. No Heart disease data base Country Abbreviation

1 Cleveland Clinic Foundation Ohio, US CCF

2 The Veterans Health Administration US V. A. medical center
3 Hungarian Institute of Cardiology Budepast, Hungary HIC
4 University of California Irvine, California UCI
5 Unified Medical Language System US UMLS
6 Massachusetts Institute of Technology, Boston’s Beth Israel Hospital US MIT-BIH
7 Long Beach Foundation Long Beach, California LBF
8 The University Hospital Zurich, Switerzland UHS
9 Cleveland Clinic Foundation and Hungarian Institute of Cardiology US and Hungry CH

the uncertainty received from the medical experts by using

FL and GA with an accuracy of 69.22%. The fuzzy rule
base generated ‘if-then’ rules by making use of C4.5
algorithm of the DT. Experiment was conducted on 299
patients of Gill medical center, Korea by taking 7 input
variables GR, AG, CHL, HDL CHL, RBP, DM and
SMOKE. The Table 4 presents the list of research papers
from 2010–2013 of FL and Hybrid-based approaches.

Review of Research Papers from 2014–2017 of FL

and Hybrid-Based Approaches

Explanation of hybrid model (2014) was made by Yuehjen

E. Shao et al. [42] consisted of many hybrid approaches
MARS, LR, RS and ANN techniques. Input variables: AG,
CP, GR, RBP, FBS, Serum CHOL, RECG, EXANG,
THALACH, OP, CA, THAL were used in this hybrid
Fig. 4 Flow chart of algorithm of FL-based approach for the risk of model and attained 83.93% accuracy evaluated by Root
heart disease detection Mean Square Error in RESE software. Jan Bohacik et al.
[8] applied cumulative application of FL Controller and DT
inference system was implemented with rule-based system in heart disease diagnosis (2014). Coding was developed in
(22 rules). Rule base was characterized by ‘if-then’ rules weka tool with a data base from Hull York medical school
using logical combinations of inputs with AND operator in and University of Hull, England with sensitivity and
MATLAB. Center of Gravity (COG) approach was applied specificity 34.23% and 91.01% respectively.
for defuzzification, and database was taken from Smt. Again Jan Bohacik et al. [9] developed an Algorithmic
Parvati Devi hospital, EMC hospital, medical center, Model (2015) for heart disease patients by employing the
Amritsar with accuracy of 92%. The demonstration of same on 2032 patients of Hull York medical school and
another fuzzy expert system (2013) was presented by Syed University of Hull, England with an accurate rate 64.41%
Umar Amin et al. [7] using a hybrid system of ANN along and specificity 63.27%. The system considered 9 input
with GA for prediction of heart disease. Total 12 input variables: BCL, BUAL, GR, BSL, AG, WT, PR, HT and
variables: GR, AG, CHL, BP, FAMHIST, SMOKE, alco- NT-proBNPlevel. Software tool Java was implemented in
hol intake, physical activity, DM, diet, OBES, stress and 1 fuzzy inference system. The hybrid technique (2015)
output variable were used. The article used input (12), designed by Ankita Dewan [10] evolved a prototype to find
hidden (10) and output (2) nodes, respectively. Imple- and execute unknown information from the database of
mentation was done in MATLAB R2012a. Accuracy of heart disease. Genetic Algorithm was hybrid with back
92% was measured by applying least mean square error propagation method. The study used 20, 10 and 10 as input,
(MSE). Jae-Kwon Kim et al. [28] represented prediction hidden and output nodes, respectively. MATLAB R2012a
model (2013) of heart disease in MATLAB to overcome was used for the implementation of the task. The proposed

123
686 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

methodology achieved nearly 100% with least error but [14] to pre-process data set, select effective attributes and
being stiffen in local minima, the system was unable to then to make a strategy of fuzzy rule base. Accuracy of this
achieve the desired outcome. The focus of Hui Yang et al. approach is very close to near 80%. CCF, HIC, UHS and V.
[29] was to construct a hybrid model (2015) for the heart A. medical center heart disease datasets from the UCI ML
failure identification by combining FL and NB in MAL- repository were used in this system. C?? was used to
LET operating system. The system made use of 7 main risk implement the system.
factors responsible for heart stroke extracted from UMLS An overview of fuzzy logic controller (2017) for coro-
database and yielded 91.5% accuracy with precision nary heart disease with 9 input variables: AG, CHL, BS,
88.4%. Krishnaiah et al. [30] developed a robust system CP, ECG, HR,EX, GR, SMOKE and 1 output variable with
(2015) using data mining techniques: FL Controller and 5 features was proposed by Mohammad Alqudah [31].
K-NN algorithm with a capability of reducing uncertainty Trapezoidal, triangular, R, and L membership functions
available in the medical data. Weka 3.6.6 was used in the were used for all variables. Mamdani inference system with
system which got 91% accuracy. Mainly 13 input attributes rule-based system (66 rules) using logical combinations of
: AG, CR, CP, RBP, CHOL, RECG, FBS, THALACH, inputs with ‘AND’ operator was implemented on patients’
EXANG, OP, SLOPE,CA, THAL were considered and records in Jordan’. Coding was done in Visual Studio 2010
testing was conducted on 550 patients of heart disease C which accomplished the diagnosis by yielding high
databases of CCF and Statlog. accuracy. Tanmey Kasbe et al. [15] designed fuzzy expert
The design of fuzzy expert engine (2016) was developed system (2017) with 10 input variables: CP, CHOL, SBP,
by Wiga Maulana et al. [11]. The system made use of 13 FBS, MHR, ECG, THAL, OP, AG, GR and 1 output
input variable: AG, CP, CHL, MNR, RBS, RECG, THAL, variable with 5 parameters. Mamdani inference mechanism
BP, OP, GR, EXANG, SLOPE, CA and 1 output variable: was implemented with rule-based system (86 rules) char-
the status of angiography. The system used C4.5, CART acterized by ‘if-then’ rules using logical combinations of
and RIPPER in MATLAB (7.12 R2011a). The membership attributes with AND/OR operator. Fuzzification was
functions were optimized by imperialist competitive algo- accomplished using triangular and trapezoidal membership
rithm (ICA). The designed method was experimented on function. MATLAB R2008 tool was used as programming
data sets from CCF and HIC, Budapest with an accuracy of language and Center of Gravity (COG) approach was
81.82%. Assemgul Duisenbayeva et al. [12] revealed more applied for defuzzification and finally achieved 93.33%
adaptable and compatible fuzzy inference system (2016) accuracy. V.A medical center, CCF and LBF were data set
that assist doctors and physicians in their decision-making sources. Coronary illness framework (2017) was applied by
for CAD disease. CP, BP, LDL CHL, BS, GR, MHR and Purushottam Sharma et al. [32] considered 13 input attri-
AG were 7 input and 1 output variable having 5 features: butes: AG, GR, CP, FBS, RBP, RECG, THALACH,
sick (s1–s4), healthy. The given approach was imple- EXANG, OP, SLOPE, CA, THAL and CHOL from the
mented in MATLAB. Coronary heart disease prediction data set of Hungary, Switzerland and U.S. The paper made
system (2016) was developed by Mokeddem et al. [48] use of very common combination of FL and GA, and
hybrid with SIPINA Decision Tree algorithm and Fuzzy brought 88.11% accuracy. Creation of accurate rule set was
logic. The system used 13 input features from the well- the main theme of this article. With a purpose of
known CCF, HIC, LBF and UCI heart disease data sets improvement in accuracy and reduction in computational
using triangular, trapezoidal, R and L membership func- time Kanika Pahwa and Ravinder Kumar [16] proposed
tions. ‘if-then’ rules were used in rule-base using logical prediction method of heart disease (2017) by hybridization
combinations of inputs with AND operator. Centroid of techniques NB and RF. Both these techniques yielded
method was used for defuzzification. This new approach 84.15% and 84.16% accuracy, respectively. Kaan Uyar
has yielded an accuracy of 94.05%. Feshki et al. [13] et al. [43] presented diagnosis of heart disease model
proposed the system of diagnosis (2016) that yielded (2017) using 13 input attributes: AG, GR, CP, RBP,
results in less time and cost with an accuracy of 91.94%. CHOL, FBS, RECG, THALACH, EXANG, OP, SLOPE,
AG, CP, RBP upon admission in hospital, GR, CHL, FBS, CA and THAL. GA-based Recurrent Fuzzy Neural Net-
CA, RECG results, EXANG, MHR, OP, SLOPE and tha- work (RFNN) involved 13 input, 7 hidden and 1 output
lassemia were 13 attributes considered in this paper. The neurons. An accuracy of 97.78% was yielded by experi-
Particle Swarm Optimization was implemented to extract 8 mentation on CCF patients’ data set which was further
features. Further, Feed Forward Back-Propagation Neural evaluated by RMSE (Root Mean Square Error). Program-
network (FFBPNN) was applied to optimize the PSO ming Languages Ubuntu and Java were used for the
algorithm. Cleveland Clinic Foundation (CCF) data set was implementation of the system. Utilization of hybrid data
used. Fuzzy Decision Support System (2016) to predict mining techniques (2017) was demonstrated by Meenal
heart disease was applied by Animesh Kumar Paul et al. Saini et al. [17] using 13 attributes AG, GR, CP, RBP,

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 687

Table 4 List of research papers from 2010–2013 of FL and Hybrid-based approaches

Year Author Title Application AI approach Strength
and Software
paper

2010 Ali Adeli, Mehndi Neshant A Fuzzy Expert System for MATLAB FL Simulation of the expert and doctor
[52] Heart Disease Diagnosis with presentation of definite output
in an undefinate term
2010 Vahib Khatibi, Gholam Ali A Fuzzy -Evidential Hybrid – Demster-Shafer Requisite of a precise, certain and firm
[39] Montazerac Inference Engine for theory of decision-making to formulate a
Coronary Heart Disease Risk evidence and model in various problems of
Assessment fuzzy sets engineering
theory (DFT)
2011 P. K. Anooj Clinical Decision Support MATLAB FL Development of a medical decision
[40] System: Risk Level (7.10) support system based on fuzzy rule
Prediction of Heart Disease engine in order to classify training
using Weighted Fuzzy Rules and testing database depending on
risk level
2011 E. P. Ephzibah A Hybrid Genetic-Fuzzy MATLAB FL and GA Implementation of the diagnosis model
[26] Expert System for Effective 7.3.0 dependant upon knowledge-based
Heart Disease Diagnosis expert system for the smooth work in
the healthcare industries
2012 Bhuvaneswari Amma NG Cardiovasular Disease – GA and NN Analysis of nonlinear functions and
[5] Prediction System using attribute selection to construct a
Genetic Algorithm and compatible model
Neural Network
2012 E. P. Ephzibah, V. A Fuzzy Rule Based Expert MATLAB FL and GA Selection of the salient and prime
[27] Sundrapandian System for Effective Heart features responsible for the rapid and
Disease Diagnosis error free diagnosis of disease
2012 M. Akhil jabbar, Priti Prediction of risk score for WEKA Association rules Associative classification rules and
[6] Chandra, B. heart disease using generated by feature subset selection. Discovery of
L. Deekshatulu associative classification and hybrid feature a new rule set through extraction of
hybrid feature subset subset important information from
selection selection knowledgeable data base of UCI to
detect risk score in Andhra Pradesh
2012 S. Muthukaruppan, M. J. Er A Hybrid Particle Swarm MATLAB 7 Fuzzy Expert Potentiality in the interpretation of the
[41] Optimization Based Fuzzy System, DT judgement through fuzzy expert
Expert System for the and PSO engine based on PSO
Diagnosis of Coronary Heart
Disease
2013 Sanjeev Kumar, Detection of Heart Disease MATLAB Fuzzy expert Itself usage by the patients and less
[53] Gursimranjeet Kaur using Fuzzy Logic system scope of redundancy and conflicts in
the generation of fuzzy rules
2013 S. U. Amin, K. Agarwal Genetic Neural Network based MATLAB GA and NN Insurance of early prediction of the
[7] and R. Beg Data Mining in Prediction of R2012a disease and hence insurance of
Heart Disease using Risk reduction in the expenditure during
Factors clinical tests and treatments of the
disease
2013 Jae-Kwon Kim, Jong-Sik Adaptive mining prediction MATLAB FL Controller and Importance of linking medical expert
[28] Lee, Dong-Kyun Park, model for content DT knowledge and information with data
Yong-Soo Lim, Young- recommendation to coronary mining approaches and reducing
Ho Lee, Eun-Young heart disease patients uncertainty in the prediction
Jung framework using information and
knowledge from medical resource
persons

CHOL, FBS, RECG, CA, THAL, EXANG, THALACH, Network, Bayesian regularized NN, Generalized linear
OP and SLOPE. This approach has considered hybridiza- model, Lasso, MARS, Classification and Regression Tree
tion of 9 classifiers classifiers: SVM, Decision tree, Neural and come out with an accuracy of 82.54%. Arabasadi et al.

123
688 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

[44] presented an adaptive and affordable technique (2017) and achieved 91% of accuracy. J. Vijayashree et al. [33]
for the enhancement of accuracy and finally achieved presented ML framework (2019) using 11 input attributes:
93.85%. The system combined together GA and NN, and AG, GR, CP, RBP, FBS, CHOL, CA, OP, RECG, SLOPE
involved 22 input, 5 hidden and 1 output neurons. Exper- and THALACH. Hybridization of PSO-SVM was utilized
imentations were done on 303 patients with data set from and compared with other classifiers and using the data base
HIC, CCF and LBF. The identification of heart disease of staLog heart disease set and yielded 88.22% accuracy
diagnosis (2017) was derived by Abhishek Rairikar et al. results and implementation was done in MATLAB. The
[18] using hybridization of GA, DT and KNN. The use of Hybrid system (2019) was expanded by M. Tarawnah et al.
13 vital clinical attributes was considered to achieve con- [34] for heart disease identification. The system comprised
siderable accuracy rate. The Table 5 presents the list of of 11 input variables from the data base of CCF and UCI
research papers from 2014–2017 of FL and Hybrid-based ML respository. Implementation of classification tech-
approaches. niques NB, SVM, J4.8, NN, GA and RF was shown and
compared with the hybrid approach of all techniques.
Review of Research Papers from 2018–2020 of FL Accuracy obtained was 89.2% and it was achieved by
and Hybridbased Approaches hybrid approach which was better than all other approa-
ches. Senthil kumar Mohan et al. [21] used hybrid ML
Explanation of a mediative fuzzy logic system (2018) techniques (RF and Linear Model (HRFLM)) to demon-
given by Ion Iancu [45] has considered 11 input variables: strate an effective heart disease prediction framework
CP, CHL, BP, MHR, BS, RECG, EX, THAL, OP, AG, GR (2019) with an accuracy of 88.7%. The study made use of
and 1 output variable with 5 features. Mamdani inference 13 attributes: AG, GR, CP, RBP, CHOL, FBS, RECG, OP,
system was implemented with rule-based system (44 rules) SLOPE, CA, THAL, MHR and EXANG. Classification of
using ‘single input-single output’ phenomenon. Software heart disease was done in R Studio Rattle software tool
MATLAB was used for the implementation of experi- using the data of 297 patients obtained from CCF data set.
mental work from the data base of CCF, V. H. medical Heart disease detection hybrid classifier (2019) was rep-
center and LBF. Defuzzification was done using ‘Middle of resented by Yukti Sharma [22] using hybridization of DT
maxima and middle of minima’ approach. Hasan Kahtan and K-Means Clustering. The study considered 14 attri-
et al. [49] proposed a FL system with 4 inputs variables: butes namely: AG, GR, CP, BP, CHL level, BS, ECG,
AG, BS, BP and CHL. Fuzzification was done using THAL, CA and SLOPE obtained from CCF data base.
trapezoidal membership function for all variables. A rule- Classifier K-Means Clustering and Decision Tree obtained
based system (96) was implemented in JAVA (using Net 49% and 52% accuracy, respectively, and the proposed
Beans IDE 8.2 in Java) with an accuracy of 98%. The work of hybrid classifier yielded 62% accuracy which was
Fuzzy Inference System (2018) designed by Vishu Madaan greater than that of individual classifiers. Saba Bashir et al.
et al. [19] has resolved a peculiar problem of diagnosis by [23] gave an emphasis on approaches and algorithms of
making use of 6 input parameters: AG, BP, CP, CHL, HR attribute selection in datasets of heart disease diagnosis
and DM. The rule base set contained 162 rules (if-then system (2019) using 4 classifiers. Logistic Regression SVM
rule) with the implementation in MATLAB tool. The yielded 84.85% accuracy which is the highest among the
proposed model has acquired 82.65% accuracy. Another accuracies obtained using DT (82.22%), LR (82.56%), RF
hybrid system (2018) based on FL controller and DT was (84.17%) and NB (84.24%) algorithms on UCI database in
described by Oumaima Terrada et al. [20] using 14 clinical Rapid Miner Studio software tool. Presentation of a fuzzy-
input parameters: Total CHL, HDL, LDL, HBP, SMOKE, based framework (2019) was proposed by Padmavathi
OBES, DM, GR, BMI, SBP, TRIG, AG, FAMHIS and Kora et al. [46] for valvular heart disease detection by
sedentary lifestyle. Software MATLAB was used the considering 7 clinical input variables: AG, BP, CHL, DM,
experiments of diagnosis for the rule base consisting of BMI, SMOKE and physical activity with 1 output variable
logical combinations of AND operator and gave 63.24% having 3 features. After training the dataset using NN,
accuracy. Mamdani inference framework was implemented with rule
The study developed by Prerna Jain et al. [50], displayed set system (44 rules) ‘if-then’ rules using logical combi-
the system (2019) considering 8 input variables: AG, CHL, nations of inputs with AND operator. The source of dataset
GR, OBY, HTN, DM, FAMHIS, SMOKE and 1 output was CCF and HIC. Parameter ROC Curve (Receiver
variable with 3 parameters. Fuzzification was done using Operator Characterstic) was used to check the accuracy of
triangular and trapezoidal membership functions for each 99.3% in MATLAB R2019b. The another proposal (2019)
and every all input variable. Mamdani approach was used using Imperialist Competitive Algorithm (ICA) to select
in rule base system (44 rules) using if-else statements. The optimal features and K-nearest neighbor approach (KNN)
system was designed in programming language MATLAB was used to classify heart disease by J. Nourmohammad-

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 689

Table 5 List of research papers from 2014–2017 of FL and Hybrid-based approaches

Year Author Title Application AI approach Strength
and software
paper

2014 Yuehjen E. Shaoa, Chia- Hybrid Intelligent Modeling RESE LR, MARS, Enlargement of a hybrid model using
[42] Ding Houa, and Chih- Schemes for Heart Disease software RS and NN least input variables and having distinct
Chou Chiu Classification techniques classification capability
2014 Jan Bohacik, C. Use of Cumulative Information Weka FL Controller An algorithmic as well as intelligent
[8] Kambhampati, Darryl Estimations for Risk and DT home tele-monitoring model based on
N. Davis, J. G. F. Assessment of Heart Failure cumulative system of Fuzzy Logic and
Clevland Patients Decision Tree
2015 J. Bohacik, K. Matiasko, Algorithmic Model for Risk Java Fuzzy Expert Challenged a task by using AI or
[9] M. Benedikovic, I. Assessment of Heart Failure System computer aided technology to develop
Nedeljakova Patients an effective and intelligent biomedical
diagnosis system
2015 Ankita Dewan and Prediction of Heart Disease MATLAB NN with back A promising, feasible, supportive and
[10] Meghna Sharma using a Hybrid Technique in R2012a propagation suitable system concluding smart
Data Mining Classification method and medical decisions
GA
2015 Hui Yang and Jonathan A Hybrid Model for Automatic MALLET Rule -based A robust knowledge extraction system
[29] Garibaldi Identification of Risk factors methods and with less human effort for automatic
for Heart Disease NB detection of heart disease
2015 V. Krishnaiah, G. Heart Disease Prediction System Weka 3.6.6 FL Controller Capability to eliminate uncertainty
[30] Narsimha and N. Using Data Mining technique and K-NN available in the medical data using data
Subhash Chandra for Fuzzy K-NN Approach algorithm mining techniques
2016 Wiga Maulana Baihaqi, Rule Extraction for Fuzzy MATLAB FL contoller Development of a new prototype along
[11] Noor Akhmad Setiawan Expert System to Diagnose 7.12 with the comparison of performance
and Igi Ardiyanto Coronary Artery Disease (R2011a) with other models and classification of
overall error rate of the data
2016 Assemgul Duisenbayeva,Using Fuzzy logic concepts in MATLAB Fuzzy Assistance to the physicians and
[12] Lyazzat Atymtayeva, creating the decision making inference practitioners by taking decisions,
Iskander Beisembetov expert system for cardio- system reducing time period to determine
vascular diseases (CVD) disease and producing error-free results
2016 Sidahmed Mokeddem and Assessment of Clinical Decision – FL and Necessity of the application of attribute
[48] Baghdad Atmani Support Systems for SIPINA DT selection technique for the removal of
Predicting Coronary Heart the irrelevant and redundant factors
Disease available in the database of heart
disease
2016 M. G. Feshhki and O. Improving the Heart Disease – NN Feed Utilization of artificial intelligence and
[13] S. Shijani Diagnosis by Evolutionary Forward mining of data to extract cognizant
Algorithm of PSO and Feed Back output to mark the improvement in the
Forward NN Propagation framework
and PSO
2016 Animesh Kumar Paul, Genetic Algorithms Based C?? Fuzzy Consumption of only relevant and
[14] Pintu Chandra Shill, Fuzzy Decision Support Decision irredundant features to achieve
Md. Rafiqul Islam Rabin System for the Diagnosis of Support admissible exactness from the
and M. A. H. Akhand Heart Disease System generalization of FDSS
(FDSS) and
GA
2017 Ali Mohammad Alqudah Fuzzy Expert System for Visual Fuzzy Logic Demonstration of the method which is
[31] Coronary Heart Disease Studio Contoller easily applicable on the patients’
Diagnosis in Jordan 2010 clinical data and construct a mobile-
based web or application for e-health
support system
2017 Tanmay Kasbe and Ravi Design of Heart Disease MATLAB Fuzzy Logic A competent, reasoning and powerful
[15] Singh Pippal Diagnosis System using Controller approach in order to manage variability
Fuzzy Logic of the available data

123
690 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

Table 5 continued
Year Author Title Application AI approach Strength
and software
paper

2017 Purushottam Sharma and Application of fuzzy logic and – Fuzzy Logic Efficient and logical engine with the
[32] Kanak Saxena genetic algorithm in heart Controller capability of replicating human
disease risk level prediction decisions
2017 Kanika Pahwa, Ravinder Prediction of Heart Disease – Naive Bayes Optimization of the classification problem
[16] Kumar using hybrid Techniques for (NB) and by adopting hybrid technique of
selecting features Random- attribute selection that helps in
Forest (RF) removing irrelevant and redundant
attributes
2017 Kaan Uyar and Ahmet Ilhan Diagnosis of heart disease using Ubuntu and ANN and Determination and extraction of the
[43] genetic algorithm-based Java RFNN unrevealed information about heart
trained recurrent fuzzy neural disease
networks
2017 Meenal Saini, Niyati Prediction of Heart Disease – Hybrid Demonstration of favorable outcomes
[17] Baliyan, Vineeta Bassi Severity with Hybrid Data Classifier with the utilization of hybrid data
Mining with mining techniques
Weighted
Voting
2017 Zeinab, Roohallah Computer aided decision – GA and NN Affordable and adaptive technique to
[44] Alizadehsani, Mohamad making for heart disease enhance accuracy and increase the
Roshanzamir, Hossein detection using hybrid neural execution of NN
Moosaei, Ali Asghar network-Genetic algorithm
Yarifard
2017 Abhishek Rairikar, Vedant Heart disease prediction using – Genetic Consumption of only relevant and
[18] Kulkarni, Vikas Sabale, data mining techniques Algorithm, irredundant features to achieve
Harshavardhan Kale and NB, DT admissible exactness from the
Anuradha Lamgunde and K-NN generalization of FDSS

Khiarak et al. [35]. The system was user-friendly in the UCI ML respository which obtained 90%, 91% and 89%
strategy of feature selection as well as training and testing accuracy, respectively. Preethi Krishnan et al. [25] pro-
of data set and yielded accuracy of 88.25%. Testing was posed a novel clinical based and intuitive model (2020) for
conducted on 303 patients with 13 clinical parameters of heart disease detection using fuzzy expert system whose
heart disease databases from UCI ML respository and T. accuracy, specificity and sensitivity were 96.6%, 96.8%
S. Rajaei hospital. An intelligent medical heart diagnose and 95.6%, respectively, using MIT-BIH heart disease data
system (2019) was designed by L. Ali et al. [24]. Data set base in Physionet. Better outcomes were generated to
was extracted from CCF which considered 13 input attri- detect PVC beats in electrocardiogram signals [57-63].
butes out of 76 attributes: AG, CHOL, CA, CP, FBS, Fuzzification was done using Gaussian and triangular
EXANG, RECG, THAL, THALACH, RBS, OP. After membership functions. Explanation of another classical
extracting noisy and redundant features using Chi-square model (2020) was described by Anna Karen et al. [47]
statistical model, the proposed method used Deep Neural using 6 ML classifiers out of which maximum accuracy
Network (DNN) to avoid underfitting and overfitting of was achieved by (CHI-PCA) with RF where the data were
network and achieved 93.33% accuracy. derived from CH datasets [54-56] . The paper also revealed
A hybrid approach (2020) to predict heart disease was clear comparative study other classifiers using CCF, HIC
proposed by G. Thippa Reddy et al. [36] considering and CH datasets in Apahe Spark 2.2.0. The prediction
together adaptive GA with FL. The first step followed the model (2020) of hybrid approach using ACO and Hybrid
selection of features through rough set theory and the K-NN (HKNN) was designed by Sowmiya et al. [37].
implementation of hybrid AGAFL classifier is made in Utilization of feature selection was made by ACO, and
second step. The membership functions were designed in then another hybrid classifier was implemented in Net-
trapezoidal functions. The system consisted of 13 input beans IDE. Experiments were conducted in Netbeans IDE
variables, namely AG, RBP, GR, CHL, MHR, OP, CA, with an accuracy of 99.2% using data obtained from CCF.
RECG, FBS, CP, SLOPE, THAL and EXANG. Experi- Comparison with other classifiers along with their accura-
mental work was done with data base from CCF, HIC and cies was presented in this study. Mohammad Ali Hassani

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 691

Table 6 List of research papers from 2018–2021 of FL and Hybrid-based approaches

Year Author Title Application AI approach Strength
and Software
paper

2018 Ion Iancu Heart Disease Diagnosis based on MATLAB FL Resolvement of contradictory
[45] mediative Fuzzy Logic situations by dealing with
inconsistent data in order to
calculate mediate output
2018 Hasan Kahtan, Kamal Heart Disease Diagnosis using JAVA FL Controller Exhibition of a promising, feasible
[49] Z.Zamli, Wan Nor Fuzzy Logic and suitable system to assist
Ashikin, Wan Ahmed peculiar diagnosis process
Fatthi, Azma Abdullah,
Noor Shahaiyusniezam
2018 Vishu Madaan, Anjali Goyal X-Cardio: Fuzzy Inference System MATLAB FL Controller Solution to resolve a peculiar
[19] to Diagnose Heart disease problem of diagnosis by dealing
with incomplete and vague
medical records. Moreover,
presentation of an error-free and
user-friendly interface of toolbox
2018 Oumaima Terrada, A fuzzy medical diagnostic support MATLAB FL Contoller Suitable solution based on Fuzzy
[20] Bouchaib Cherradi, system for cardiovascular and DT Logic and further on DT due to
Abdelhadi Raihani and diseases diagnosis using risk more count of rules to improve
Omar Bouattane factors results
2019 Prerna Jain and Amandeep A Fuzzy Expert System for MATLAB FL Controller Provision of an easy-to-use
[50] Kaur Coronary Artery Disease graphical presentation of the
Diagnosis dataset along with working
environment resulting in
constructing predictive analytics
2019 J. Vijayashree, H. Parveen A machine learning framework for MATLAB SVM and PSO Requirement of an effective and
[33] Sultana feature selection in heart disease powerful system in order to trace
Classification using Improved the essential features that
Particle Swarm Optimization contribute towards the diagnosis
with S.V.M classifier of disease
2019 Monther Tarawnah, Ossama Hybrid Approach for Heart Disease – NB and SVM Analysis of hidden patterns,
[34] Embarak Prediction using Data Mining information and relationships
Techniques available in the clinical data of
the patients
2019 Senthil kumar Mohan, Effective Heart Disease Prediction R Studio HRFLM A novel approach with high
[21] Chandrasegar Thirumalai, Using Hybrid Machine Learning Rattle discriminative power to develop
Gautam Srivastava Techniques and frame a model resulting in the
improvement of medical
awareness and reduction in the
expenditure of the treatment
2019 Yukti Sharma, Rikku Hybrid Classifier for Identification – Decision Tree Framework of a model that has
[22] Veliyambara and Rajshree of Heart Disease (DT) and ability to generate high accuracy
Shettar K-means by hybridism classifiers rather
clustering than single individual classifier
algorithm
2019 Saba Bashir, Zain Sikander Improving Heart Disease Prediction Rapidminer LR, DT, LR Prerequisite of a system that
[23] Khan, Farhan Hassan using Feature Selection Studio SVM, RF eliminates unnecessary and
Khan, Khurram Bashir Approaches and NB irrelevant factors present in any
and Aitzaz Anjum data base of heart patients
2019 Dr. Padmavathi Kora, Dr. New hybrid method for heart – Imperialist A boosting and compatible approach
[46] K Meenakshi, Dr. disease diagnosis utilizing Competitive with acceptable accuracy by using
K Swaraja, Dr. A Rajani optimization algorithm in feature Algorithm efficient and reliable classifier
and Md. Kafiul Islam selection (ICC) and
K-NN

123
692 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

Table 6 continued
Year Author Title Application AI approach Strength
and Software
paper

2019 Jalil Nourmohammadi-Khiarak, Detection of Cardiac Arrhythmias MATLAB FL Automatic prediction with the
[35] Mohammad-Reza Feizi- using Fuzzy Logic R2019b Controller application of determination
Derakhshi, Khadijeh Behrouzi, and ANN strategies which are PC based
Samaneh Mazaheri, Yashar followed by artificial
Zamani-Harghalani and intelligence
Rohollah Moosavi Tayebi
2019 Liaqat Ali, Atiqur Rahman, An Automated Diagnostic System Python v2 Statistical Removal of noisy and irrelevant
[24] Aurangzeb Khan, Mingyi for Heart Disease Prediction Model and attributes from feature space
Zhou, Ashir Javeed and Javed Based on v Statistical Model DNN along with automation and
Ali Khan and Optimally Configured Deep improvement in the accuracy of
Neural Network disease diagnosis
2020 G. Thippa Reddy, M. Praveen Hybrid genetic algorithm and a MATLAB FL Application of t-test to show the
[36] Kumar Reddy, Kuruva fuzzy logic classifier for heart 7.12 Controller existence of significant
Lakshmanna, Dharmendra disease diagnosis and GA difference between proposed
Singh Rajput, Rajesh Kaluri algorithm and present one in
and Gautam Srivastava MATLAB R2014a
2020 Preethi Krishnan, Viswanathan A Novel Severity Index of Heart PhysioNet FL Real time applicable clinical based
[25] Rajagopalan, Bashir I. Morshed Disease from Beat-wise Controller and intuitive model which is
Analysis of ECG Using Fuzzy computationally efficient to
Logic for Smart-Health generate better outcomes to
detect PVC beats in
electrocardiogram signals
2020 Anna KarenGarate-Escamila, Classification models for heart Apahe park CHI-PCA Application of feature selection
[50] Amir Hajjam El disease prediction using feature 2.2.0 and RF method and dimensionality
Hassani,Emmanuel Andresb selection and PCA reduction technique after
tracing the features of the risk of
heart disease
2020 C. Sowmiya, P. Sumitra A Hybrid Approach for Mortality Netbeans ACO and Design of an innovative hybrid
[37] Prediction for Heart Patients IDE HKNN approach with various
using ACO-HKNN classifiers by selecting
necessary attributes from the
data set
2020 Mohammad Ali Hassani, Ran An Approach of Predicting Heart – DT and NN Noteworthy approach that deals
[51] Tao, Marjan Kamyab, Disease Using a Hybrid Neural with combination of unique
Mohammad Hadi Mohammadi Network and Decision Tree dataset i.e. CCF and Stalog
heart disease dataset, UCI ML
repository
2021 L. J. Muhammad, Ebrahem A. Fuzzy-based expert system for MATLAB FL and Requirement of a affordable,
[38] Algehyne diagnosis of coronary artery Improved capable, reliable and relatively
disease in Nigeria C4.5 unique technology
algorithm

et al. [51] used hybrid ML techniques (NN and DT) to variables, namely AG, RBP, glucose, LDL CHOL, HDL
demonstrate a prouctive heart disease detection model CHOL, triglycerides, BCL, BMI, MHR, CP and CHL.
(2020) with an accuracy of 98.7%. The study made use of Mamdani approach was applied in the fuzzy inference
13 input attributes: AG, GR, CP, RBP, CHOL, FBS, engine of 87 rules, and defuzzification was completed by
RECG, OP, SLOPE, CA, THAL, THALACH and centroid method. The proposed technique was designed
EXANG. Classification of heart disease was done in Weka and implemented in the MATLAB tool. Experimental work
software tool using the data of 227 patients obtained from was done with data base from State Ministry of health,
CCF and StaLog heart disease data set. A new and unique Kano, Nigeria which obtained 94.55% accuracy. Testing
hybrid technique (2021) using Fuzzy Logic along with dataset was of 100 patients implemented in MATLAB
Improved C4.5 algorithm was designed and implemented software. The Table 6 presents the list of research papers
by Muhammad et al. [38]. The system consisted of 11 input from 2018–2021 of FL and Hybrid-based approaches.

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 693

Table 7 Summary of accuracy rate of heart disease detection based on FL and Hybrid-based approaches
References Research methodology Accuracy Evaluation parameters

[52] FL 94% –
[39] Demster–Shafer theory of 91.58% –
evidence and fuzzy sets theory
[40] FL a) 62.35% (CCF) b) 46.9% (HIC) c) 51.3% K = 10 fold cross validation (CCF, HIC), K =
(UHS) 8 fold cross validation (UHS)
[26] FL ? GA – –
[5] GA ? NN 94.17% Confusion matrix
[27] FL ? GA – –
[6] GA ? Association classification 95% Confusion matrix
rule mining
[41] FL ? DT 93.27% Confusion matrix
[53] FL 92% –
[7] GA ? NN 92% Mean square error (MSQ)
[28] FL ? DT 69.22 –
[42] LR ? MARS ? RS ? ANN 83.93 Type-I/Type-II error
[8] FL ? DT 90.94% K = 10 fold cross validation
[9] FL 64.41% K=10 fold cross validation
[10] NNBP ? GA Near 100%
[29] NB ? Rule based methods 91.5% F-measure
[30] FL?K-NN 91% (k = 1) Confusion matrix
[11] FL 81.82% F-measure
[12] FL – –
[48] FL?DT 94.05% K=10 cross validation
[13] NNFFBP ? PSO 91.94% K = 10 cross validation
[14] GA ? FL Near 80% Confusion Matrix
[31] FL – –
[15] FL 93.33% –
[32] FL 88.11% K=10 cross validation
[16] NB ? RF 84.16% K = 10 cross validation, ROC Curve
[43] ANN ? Recurrent NN 97.78% F-measure, Root Mean Square Error (RMSE)
[17] Hybrid classifier with weighted 82.54%
voting
[44] GA ? NN 93.85% ROC Curve
[18] GA ? NB ? DT ? K-NN – –
[45] FL – –
[49] FL 98% –
[19] FL 82.65% –
[20] FL?DT (C4.5 Algorithm) 63.24% –
[50] FL 91% –
[33] SVM ? PSO 88.22% ROC Curve
[34] NB ? SVM ? ANN?DT 89.2% F-measure
(J4.8)?GA
[21] RF ? Linear Model 88.7% F-measure
[22] DT ? K-means clustering 62% K-fold cross validation
algorithm
[23] LR, DT, LR SVM, RF, NB a) DT-82.22% b) LR-82.56% c) RF-84.17% d) K=5 fold cross validation
NB-84.24% e)LR SVM-84.85%
[46] FL ? ANN 99.3% ROC Curve

123
694 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

Table 7 continued
References Research methodology Accuracy Evaluation parameters

[35] K-NN ? ICA 88.25% Confusion matrix

[24] K-NN ? CHI model 93.33% ROC curve, AUC, MCC
[36] AGAFL a) 90% (CCF) b)91% (HIC) c)89% (UHS) K = 10 fold cross validation
[25] FL 96.6% Confusion matrix
[47] Chi-square?PCA?RF a) 98.7% (CCF) b) 99.0% (CCF, HIC) c) 99.4% (HIC) F1-score
[37] HKNN ? ACO 99.2% Precision/recall
[51] DT ? NN 98.7% F-meaasure/ROC
[38] FL? Improved C4.5 algorithm 94.55% ROC

Fig. 5 Year-wise accuracy rate using FL

Fig. 7 Year-wise accuracy rate using FL and hybrid system

Comparative Analysis of Accuracies/Techniques

Accuracy of all respective years for FL and Hybrid System

is considered and shown in Table 7 along with evaluation
parameters. In the Figs. 5 and 6, higher accuracy in that
respective year is shown and further, comparative analysis
of both system’s accuracies is presented in Fig. 7 along
with technique used in that particular paper. All the papers
come up with distinct techniques to improve the accuracy
in the diagnosis system of heart. The highest accuracy
(99.4%) is achieved by using Fuzzy Logic hybrid with
Improved C4.5 algorithm in the current year 2021 [38].
Existence of uncertainty is always there in the detection of
any disease. An extensive amount of research has been
carried out by various researchers using FL on heart dis-
ease detection as FL has enough capability to diagnose
Fig. 6 Year-wise accuracy rate using hybrid system heart disease. Therefore, to handle FL is more suitable but
the only challenge in FL system is the optimization of
rules. If rules are more, complexity of the system increases
and if rules are less, system accuracy decreases. Hence,

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 695

Fig. 8 Future framework of fuzzy logic

selection of optimal rules is a major problem in FL. After motivates young researchers and scholars to identify the
that the next and the most used techniques are ANN, GA gaps in existing work and take-up the work to create new
and DT. ANN is indeed a robust and powerful algorithm models and optimize existing models. The efforts should be
still it faces the challenge of the implementation of the made to make cost-effective real-life software and tools
automatic detection of the disease due to the complex and which can be used by health institutions on a day-to-day
complicated structure of the network. Though GA and DT basis helping to improve heart disease diagnostics.
are easy and simple to understand, yet GA is time-con-
suming algorithm and DT leads to overfitting of the data.
Future Model of Fuzzy Logic

Conclusion In future, model for direct service of the patients from the
old age homes or other home care centers to the Intensive
The paper presents concise information about the diverse Care Unit (ICU) through ambulance services can be plan-
work done by various scholars in the field of developing a ned. An artificially intelligent system will take the data of
heart disease detection algorithm and software using soft clinical parameters from old age homes or other care
computing techniques. The paper is the result of in-depth centers. Fuzzy Logic system will be implemented to get the
study and analysis of various papers published year-on- single output that will reveal distinct stages of patients in
year in various renowned journals on this subject. Key terms of healthy, first/second stage of sickness and critical
finding is that maximum publications are based on FL and stage. The system will show green color if the status of the
FL Hybrid Systems to improve the accuracy using genetic person is healthy, and the respective person will be
fuzzy logic, fuzzy neural, Adaptive Neuro-Fuzzy, Genetic informed via SMS that you are ‘Healthy’. Otherwise, if the
Neuro-Fuzzy and PSO with FL as FL has enough capability person is at the first/second stage of sickness, then a SMS
to diagnose the heart disease. Comparison of accuracy of ‘Do frequent monitoring’ will be sent to his/her mobile
various approaches has been presented graphically, and the number. Furthermore, if the patient is at chronic stage of
best approach is also recommended. The paper also any disease, the system will show red color and the patient

123
696 J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697

will be directly brought to the hospital. The Fig. 8 repre- 14. A.K. Paul, P.C. Shill, M.R. IslamRabin, M.A.H. Akhand, Genetic
sents the future framework of the fuzzy logic system in any algorithm based fuzzy decision support system for the diagnosis
of heart disease, in: Proceedings of International Conference on
type of disease. Informatics, Electronics and Vision(ICIEV), IEEE, pp. 145–150
(2016)
15. T. Kasbe, R.S. Pippal, Design of heart disease diagnosis system
Funding No funding for this paper. using fuzzy logic, in: Proceedings of International Conference on
Energy, Communication, Data Analytics and Soft Computing
Declarations (ICECDS), IEEE, pp. 3183–3187 (2017)
16. K. Pahwa, R. Kumar, Prediction of heart disease using hybrid
Conflict of Interest The authors declare that they have no conflict of technique for selecting features, in: Proceedings of Uttar Pradesh
interest. Section International Conference on Electrical, Computer and
Electronics (UPCON), IEEE, pp. 500–504 (2017)
17. M. Saini, N. Baliyan, V. Bassi, Prediction of heart disease
severity with hybrid data mining, in: Proceedings of International
References Conference on Telecommunication and Networks (TEL-NET),
IEEE, pp. 1–6 (2017)
1. WHO (2019) https://www.who.int/health-topics/cardiovascular- 18. A. Rairikar, V. Kulkarni, H.K.V. Sabale, A. Lamgunde, Heart
diseases/tab disease prediction using data mining techniques, in: Proceedings
2. CDC (2019) https://www.cdc.gov/datastatistics/index.html of International Conference on Intelligent Computing and Control
3. BHF (2020) https://www.bhf.org.uk/informationsupport/heart- (I2C2), IEEE, pp. 1–8 (2017)
matters-magazine/news/behind-the-headlines/coronavirus 19. V. Madaan, A. Goyal, X-cardio: Fuzzy inference system to
4. U. of Oxford, Ai technology can predict heart attacks (2020). diagnose heart diseases, in: Proceedings of International Con-
https://www.ox.ac.uk/news/2019-09-04-AI-technology-can-predi ference on Advances in Computing, Communication Control and
ct-heart-attacks Networking (ICACCCN), IEEE, pp. 1049–1053 (2018)
5. N.G. Bhuvaneswari, Cardiovasular disease prediction system 20. O. Terrada, B. Cherradi, A. Raihani, O. Bouattane, A fuzzy
using genetic algorithm and neural network, in: Proceedings of medical diagnostic support system for cardiovascular diseases
International Conference on Computing, Communication and diagnosis using risk factors, in: Proceedings of International
Applications (ICCCA), IEEE, pp. 1–5, (2012) Conference on Electronics, Control, Optimization and Computer
6. M.A. jabbar, P. Chandra, B.L. Deekshatulu, Prediction of risk Science (ICECOCS), IEEE, pp. 1–6 (2018)
score for heart disease using associative classification and hybrid 21. S. Mohan, C. Thirumalai, G. Srivastava, Effective heart disease
feature subset selection, in: Proceedings of International Con- prediction using hybrid machine learning techniques. IEEE
ference on Intelligent Systems Design and Applications (ISDA), Access. 7, 81542–81554 (2019)
IEEE, pp. 628–634 (2012) 22. Y. Sharma, R. Veliyambara, R. shettar, Hybrid classifier for
7. S.U. Amin, K. Agarwal, R. Beg, Genetic neural network based identification of heart disease, in: Proceedings of International
data mining in prediction of heart disease using risk factors, in: Conference on Computational Systems and Information Tech-
Proceedings of International Conference on Information and nology for Sustainable Solution (CSITSS), IEEE, pp. 1–3(2019)
Communication Technologies, IEEE, pp. 1227–1231 (2013) 23. S. Bashir, Z.S. Khan, F.H. Khan, K. Bashir, A. Anjum, Improving
8. J. Bohacik, C. Kambhampati, D.N. Davis, J.G. Cleland, Use of heart disease prediction using feature selection approaches, in:
cumulative information estimations for risk assessment of heart Proceedings of International Bhurban Conference on Applied
failure patients, in: Proceedings of International Conference on Sciences and Technology (IBCAST), IEEE, pp. 619–623 (2019)
Fuzzy Systems (FUZZ-IEEE), IEEE, pp. 1402–1407 (2014) 24. L. Ali, A. Rahman, A. Khan, M. Zhou, A. Javeed, J.A. Khan, An
9. J. Bohacik, K. Matiasko, M. Benedikovic, I. Nedeljakova, automated diagnostic system for heart disease prediction based on
Algorithmic model for risk assessment of heart failure patients, v2 statistical model and optimally configured deep neural net-
in: Proceedings of International Conference on Intelligent Data work. IEEE Access 7, 34938–34945 (2019)
Acquisition and Advanced Computing Systems: Technology and 25. P. Krishnan, V. Rajagopalan, B. I. Morshed, A novel severity
Applications (IDAACS), IEEE, pp. 177–181 (2015) index of heart disease from beat-wise analysis of ecg using fuzzy
10. A. Dewan, M. Sharma, Prediction of heart disease using a hybrid logic for smart-health, in: Proceedings of International Confer-
technique in data mining classification, in: Proceedings of Inter- ence on Consumer Electronics (ICCE), IEEE, pp. 1–5 (2020)
national Conference on Computing for Sustainable Global 26. E.P. Ephzibah, A hybrid genetic-fuzzy expert system for effective
Development (INDIACom), IEEE, pp. 704–706 (2015) heart disease diagnosis, in: Proceedings of International Confer-
11. W.M. Baihaqi, N.A. Setiawan, I. Ardiyanto, Rule extraction for ence on Advances in Computing and Information Technology,
fuzzy expert system to diagnose coronary artery disease, in: Springer, pp. 115–121 (2011)
Proceedings of International Conference on Information Systems 27. E.P. Ephzibah, V. Sundrapandian, A fuzzy rule based expert
and Electrical Engineering (ICITISEE), IEEE, pp. 136–141 system for effective heart disease diagnosis, pp. 196–203 (2012)
(2016) 28. J.K. Kim, J.S. Lee, D.K. Park, Y. Lim, Y.H. Lee, E.Y. Jung,
12. A. Duisenbayeva, L. Atymtayeva, I. Beisembetov, Using fuzzy Adaptive mining prediction model for content recommendation
logic concepts in creating the decision making expert system for to coronary heart disease patients. Cluster Comput. 17(3),
cardio vascular diseases (cvd), in: Proceedings of International 881–891 (2013)
Conference on Application of Information and Communication 29. H. Yang, J.M. Garibaldi, A hybrid model for automatic identifi-
Technologies (AICT), IEEE, pp. 1–5 (2016) cation of risk factors for heart disease. J. Biomed. inform. 58,
13. M.G. Feshki, O.S. Shijani, Improving the heart disease diagnosis 171–182 (2015)
by evolutionary algorithm of pso and feed forward neural net- 30. V. Krishnaiah, G. Narsimha, N. S. Chandra, Heart disease pre-
work, in: Proceedings of International Conference on Artificial diction system using data mining technique by fuzzy k-nn
Intelligence and Robotics (IRANOPEN), IEEE, pp. 48–53, approach, in: Proceedings of Annual Convention of the Computer
(2016) Society of India (CSI), Springer, pp. 371–384 (2015)

123
J. Inst. Eng. India Ser. B (April 2022) 103(2):681–697 697

31. A.M. Alqudah, Fuzzy expert system for coronary heart disease 43. K. Uyar, A. Ilhan, Diagnosis of heart disease using genetic
diagnosis in Jordan. Health Technol. 7(2), 215–222 (2017) algorithm based trained recurrent fuzzy neural networks. Proc.
32. P. Sharma, K. Saxena, Application of fuzzy logic and genetic Comput. Sci. 120, 588–593 (2017)
algorithm in heart disease risk level prediction. Int. J. Syst. Assur. 44. Z. Arabasadi, R. Alizadehsani, M. Roshanzamir, H. Moosaei,
Eng. Manag. 8(2), 1109–1125 (2017) A.A. Yarifard, Computer aided decision making for heart disease
33. J. Vijayashree, H.P. Sultana, A machine learning framework for detection using hybrid neural network-genetic algorithm. Com-
feature selection in heart disease classification using improved put. Methods Programs Biomed. 141, 19–26 (2017)
particle swarm optimization with support vector machine classi- 45. I. Iancu, Heart disease diagnosis based on mediative fuzzy logic.
fier. Program. Comput. Softw. 44(6), 388–397 (2018) Artif. Intell. Med. 89, 51–60 (2018)
34. M. Tarawneh, O. Embarak, Hybrid approach for heart disease 46. P. Kora, K. Meenakshi, K. Swaraja, A. Rajani, M. Kafiul, M.K.
prediction using data mining techniques. Adv. Internet Data Web Islam, Detection of cardiac arrhythmias using fuzzy logic.
Technol. 29, 447–454 (2019) Inform. Med. Unlocked 17, 1–7 (2019)
35. J. Nourmohammadi-Khiarak, M.R. Feizi-Derakhshi, K. Behrouzi, 47. A.K. Garate-Escamilla, A.H.E. Hassani, E. Andres, Classification
S. Mazaheri, Y. Zamani-Harghalani, R.M. Tayebi, New hybrid models for heart disease prediction using feature selection and
method for heart disease diagnosis utilizing optimization algo- PCA. Inform. Med. Unlocked 19, 1–11 (2020)
rithm in feature selection. Health and Technology, pp. 1–12 48. S. Mokeddem, B. Atmani, Assessment of clinical decision sup-
(2019) port systems for predicting coronary heart disease. Int. J. Oper.
36. G. Reddy, M.P.K. Reddy, K. Lakshmanna, D.S. Rajput, R. Res. Inf. Syst. (IJORIS) 7(3), 57–73 (2016)
Kaluri, G. Srivastava, Hybrid genetic algorithm and a fuzzy logic 49. H. Kahtan, K. Z. Zamli, W. N. Ashikin, W. A. Fatthi, A.
classifier for heart disease diagnosis. Evol. Intell. 13, 185–196 Abdullah, N. Shahaiyusniezam, Heart disease diagnosis using
(2020) fuzzy logic, in: Proceedings of International Conference on
37. C. Sowmiya, P. Sumitra, A hybrid approach for mortality pre- Software and Computer Applications, IEEE, pp. 297–301 (2018)
diction for heart patients using ACO-HKNN. J. Ambient Intell. 50. P. Jain, A. Kaur, A fuzzy expert system for coronary artery dis-
Hum. Comput., pp. 1–8 (2020) ease diagnosis, in: Proceedings of International Conference on
38. L.J. Muhammad, E.A. Algehyne, Fuzzy based expert system for Advanced Informatics for Computing Research, IEEE, pp. 1–6
diagnosis of coronary artery disease in Nigeria. Health Technol. (2019)
11(2), 1–11 (2021) 51. M.A. Hassani, M.K.R. Tao, M.H. Mohammadi, An approach of
39. V. Khatibi, G.A. Montazerac, A fuzzy-evidential hybrid infer- predicting heart disease using a hybrid neural network and
ence engine for coronary heart disease risk assessment. Expert decision tree, in: Proceedings of International Conference on Big
Syst. Appl. 37(12), 8536–8542 (2010) Data and Computing, IEEE, pp. 84–89 (2020)
40. P.K. Anooj, Clinical decision support system: risk level predi- 52. A. Adeli, M. Neshat, A fuzzy expert system for heart disease
cation of heart disease using weighted fuzzy rules. J. King Saud diagnosis, in: International Multi Conference of Engineers and
Univ. Comput. Inf. Sci. 24(1), 27–40 (2011) Computer Scientists, 1, pp. 134–139 (2010)
41. S. Muthukaruppan, M.J. Er, A hybrid particle swarm optimization 53. S. Kumar, G. Kaur, Detection of heart diseases using fuzzy logic.
based fuzzy expert system for the diagnosis of coronary artery Int. J. Eng. Trends Technol. 4, 2694–2699 (2013)
disease. Expert Syst. Appl. 39(14), 11657–11665 (2012)
42. Y.E. Shaoa, C. Houa, C. Chiu, Hybrid intelligent modelling Publisher’s Note Springer Nature remains neutral with regard to
schemes for heart disease classification. Appl. Soft Comput. 14, jurisdictional claims in published maps and institutional affiliations.
47–52 (2014)

123