See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/352840718

Review Paper on Prediction of Heart Disease using Machine Learning Algorithms

Article · June 2021

DOI: 10.22214/ijraset.2021.35626

CITATIONS READS
2 2,928

4 authors, including:

Aadar Pandita Sarita Yadav

2 PUBLICATIONS 3 CITATIONS
Bharatividyapeeth's college of Engineering, New Delhi
10 PUBLICATIONS 9 CITATIONS
SEE PROFILE
SEE PROFILE

All content following this page was uploaded by Aadar Pandita on 30 June 2021.

The user has requested enhancement of the downloaded file.

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VI Jun 2021- Available at www.ijraset.com

Review Paper on Prediction of Heart Disease using

Machine Learning Algorithms
Aadar Pandita1, Siddharth Vashisht2, Aryan Tyagi3, Prof. Sarita Yadav4
1, 2, 3, 4
Department of Information Technology, Bharati Vidyapeeth’s College of Engineering, New Delhi

Abstract: Heart disease has been one of the ruling causes for death for quite some time now. About 31% of all deaths every year
in the world take place as a result of cardiovascular diseases [1].
A majority of the patients remain uninformed of their symptoms until quite late while others find it difficult to minimise the
effects of risk factors that cause heart diseases.
Machine Learning Algorithms have been quite efficacious in producing results with a high level of correctness thereby
preventing the onset of heart diseases in many patients and reducing the impact in the ones that are already affected by such
diseases. It has helped medical researchers and doctors all over the world in recognising patterns in the patients resulting in
early detections of heart diseases.
Keywords: Cardiovascular Diseases (CVDs); Support Vector Machine (SVM); K- Nearest Neighbour (KNN); Naive Bayes (NB);
Random Forest (RF); Logistic Regression (LR); Machine Learning (ML); Prediction Model

I. INTRODUCTION
Machine Learning being a subset of Artificial Intelligence has been one of the most vital tools in recent history. It has helped further
the progress of the healthcare industry manifolds. With such a tool, medical professionals and researchers have been able to
diagnose and detect diseases with much accurate precision. It has contributed to saving many lives.
Majority of the heart related diseases can be averted if people focus on their physical activity, have a balanced and nutritious diet
and avoid consumption of products such as tobacco and alcohol that have a damaging effect on the heart and their overall health.
Therefore, it becomes paramount that these diseases are detected as early, so that its effects can be managed with medical advice
and medicines. In this paper, we are going to review various advancements and recent works that have been done using Machine
learning in the prediction of heart diseases. Heart diseases are a result of a multitude of aspects that can influence the cardiovascular
health of an individual such as age, blood sugar, blood pressure, cholesterol etc.

II. LITERATURE SURVEY

Apurv Garg et al. [2] implemented KNN and Random Forest machine learning algorithms in order to predict heart diseases. After
obtaining and analysing the data, its balancing was checked and correlation was found between various attributes and their effect on
the target value. The dataset obtained was the UCI dataset available at Kaggle. It was divided in 80-20 ratio for training and testing
respectively. It was found that Chest Pain and Maximum heart rate achieved had a positive correlation with the target attribute. This
model provided an accuracy of 86.885% using KNN and 81.967% accuracy using Random Forest.
Rishabh Magar et al. [3] proposed a web application based predictive model trained on the UCI dataset with a 75-25 training and
testing division of the dataset. Logistic Regression based predictive models were found to be the most accurate with 82.89%
accuracy, followed by SVM at 81.57% and Naive Bayes and Decision Tree at 80.43% each. The web application can be used by the
end user as a preliminary test for checking their heart condition and seeking medical advice if needed.
Apurb Rajdhan et al. [4] proposed a system where four classification algorithms such as Random Forest, Decision Tree, Logistic
Regression and Naive Bayes are used to predict the patient’s condition. Data is split into 80% training data and 20% testing data. A
confusion matrix depicting true and false positives as well as true and false negatives was created. Maximum accuracy obtained was
90.16% using Random Forest classification.
Devansh Shah et al. [5] proposed a system of models using supervised learning methods through the WEKA tool. Four individual
classification techniques including NB, KNN, RF, DT were used to predict the chances of having a heart disease. The dataset was
initially cleaned, transformed by smoothening, normalisation, and aggregation, integrated and reduced. The maximum accuracy
obtained was through KNN method.

©IJRASET: All Rights are Reserved 2937

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VI Jun 2021- Available at www.ijraset.com

Harshit Jindal et al. [6] implemented a system that uses three different classification algorithms, KNN, RF, LR and results in an
accuracy of 87.5%. In this EHDPS i.e, effective heart disease prediction system, Logistic Regression and KNN outperform RF with
KNN providing an accuracy of 88.52% which is highest amongst the three techniques used.
Aadar Pandita et al. [7] proposed a predictive model that implements 5 machine learning algorithms and uses the technique with the
highest accuracy to build a web application that takes in patient’s medical details and predicts if they have a heart disease or not.
The web application is built using HTML/CSS and Flask based framework. The maximum accuracy obtained was obtained using
KNN, i.e, 89.06% while Logistic Regression contributed with least accuracy of 84.38%.
N. Saranya et al. [8] proposed a time and money efficient model of predicting heart disease using a web application. The model
works on two different methods : Random Forest and KNN. The dataset has been taken from one of Coimbatore’s hospitals which
produces an accuracy of 100% using Random Forest and 91.36% using KNN after cleaning and pre-processing of the dataset. An
ensemble model with and without Logistic Regression is also used to predict the chances with an accuracy of 98.77% and 95.06%
respectively.
Aravind Akella et al. [9] applied six predictive models on the UCI dataset and achieved a maximum accuracy of 93.03% with
Neural Networks with a recall of 93.8 indicating low chances of false negatives and therefore extremely precise result, while the
other five had an accuracy of almost 80% and more.
Ravindhar NV et al. [10] implemented five algorithms: Logistic Regression, Naive Bayes, Fuzzy KNN, K-Means Clustering and
back propagation Neural-Network. A 10-fold cross validation method is used in the experimental analysis of heart conditions. The
maximum accuracy was gathered using back propagation Neural Network with 98.2% accuracy and 87.64% recall and 89.65%
precision.

A comparative study of various papers on heart disease prediction using ML models

Year Author Paper Name Algorithms Used Accuracy Obtained

2020 Rishabh Heart disease prediction using Logistic Regression 82.89%

Magar et al. machine learning SVM 81.57%
[3] Naive Bayes 80.43%
Decision Tree 80.43%

2020 Apurb Heart disease prediction using Logistic Regression 85.25%

Rajdhan et al. machine learning
[4] Decision Tree 81.97%

Random Forest 90.16%

Naive Bayes 85.25%

2020 Devansh Shah Heart disease prediction using Naive Bayes 88.157%
et al. [5] machine learning techniques
KNN 90.789%

Random Forest 86.84%

Decision Tree 80.263%

2021 Harshit Jindal Heart Disease prediction using KNN 88.52%

et al. [6] machine learning algorithms
Logistic Regression 88.5%

KNN & LR based model 87.5%

©IJRASET: All Rights are Reserved 2938

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VI Jun 2021- Available at www.ijraset.com

2021 Aadar Pandita Prediction of Heart Disease Logistic Regression 84.38%

et al. [7] using Machine Learning
Algorithm KNN 89.06%

SVM 87.50%

Naive Bayes 85.94%

Random Forest 87.50%

2020 N. Saranya et Heart Disease prediction using Random Forest 100%

al. [8] Machine Learning
KNN 91.36%

Logistic Regression 87.65%

Ensemble model with 95.06%

Logistic Regression

Ensemble model without

Logistic Regression 98.77%

2021 Aravind Machine learning algorithms Generalized linear 87.64%

Akella et al. for predicting coronary artery model
[9] disease : efforts toward an
open source solution Decision Tree 79.78%

Random Forest 87.64%

SVM 86.52%

Neural Network 93.03%

KNN 84.27%

2019 Ravindhar NV Intelligent Diagnosis of Logistic Regression 81.86%

et al. [10] Cardiac Disease Prediction Naive Bayes 61.46%
using Machine Learning Fuzzy KNN 87.33%
K-Means Clustering 43.24%
BP-Neural Network 98.20%

III. CONCLUSION
Majority of researchers have used the Cleveland Heart Disease Dataset available from the UCI repository containing 76 attributes
and 303 instances, of which only 14 attributes are used due to missing values [11]. There are huge benefits to having feature
selection methods so as to minimise the number of attributes that one has to use in order to build an accurate model by checking the
correlation between various attributes and their impact on the accuracy of the models. It can be seen from various research papers in
the field that KNN and Neural Network works quite accurately in most cases for the prediction of heart diseases.

©IJRASET: All Rights are Reserved 2939

 International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.429
Volume 9 Issue VI Jun 2021- Available at www.ijraset.com

IV. FUTURE WORK

Future researchers should work towards improving the existing accuracies. They can create their own dataset using the existing
datasets available in order to increase the sample size and allow predictive models to train on a larger dataset thereby increasing the
chances of obtaining improved accuracy. Special focus should be put towards removing false positives and false negatives from the
existing models. The predictive models should be accessible to the people in the form of a web or a mobile application so that
people can try to be aware of their heart condition and consult a medical professional if their results predict a related disease.
Although users of such applications should keep in mind that these applications are not a substitute for a doctor, rather just a model
trained on a certain dataset which could have errors.
REFERENCES
[1] https://www.who.int/en/news-room/fact-sheets/detail/cardiovascular-diseases-(cvds)
[2] Garg, Apurv & Sharma, Bhartendu & Khan, Rizwan. (2021). Heart disease prediction using machine learning techniques. IOP Conference Series: Materials
Science and Engineering. 1022. 012046. 10.1088/1757-899X/1022/1/012046.
[3] "HEART DISEASE PREDICTION USING MACHINE LEARNING", International Journal of Emerging Technologies and Innovative Research
(www.jetir.org), ISSN:2349-5162, Vol.7, Issue 6, page no.2081-2085, June-2020, Available :http://www.jetir.org/papers/JETIR2006301.pdf
[4] Apurb Rajdhan , Avi Agarwal , Milan Sai , Dundigalla Ravi, Dr. Poonam Ghuli, 2020, Heart Disease Prediction using Machine Learning, INTERNATIONAL
JOURNAL OF ENGINEERING RESEARCH & TECHNOLOGY (IJERT) Volume 09, Issue 04 (April 2020)
[5] Shah, D., Patel, S. & Bharti, S.K. Heart Disease Prediction using Machine Learning Techniques. SN COMPUT. SCI. 1, 345 (2020).
https://doi.org/10.1007/s42979-020-00365-
[6] Harshit Jindal et al 2021 IOP Conf. Ser.: Mater. Sci. Eng. 1022 012072
[7] Aadar Pandita, Siddharth Vashisht, Aryan Tyagi, Prof. Sarita Yadav."Prediction of Heart Disease using Machine Learning Algorithms", Volume 9, Issue V,
International Journal for Research in Applied Science and Engineering Technology (IJRASET) Page No: 2422-2429, ISSN : 2321-9653, www.ijraset.com
[8] N. Saranya, P. Kaviyarasu, A. Keerthana, C. Oveya. Heart Disease Prediction using Machine Learning International Journal of Recent Technology and
Engineering (IJRTE) ISSN: 2277-3878, Volume-9 Issue-1, May 2020, Page No: 700-70
[9] Akella, Aravind and Akella, Sudheer. Machine learning algorithms for predicting coronary artery disease: efforts toward an open source solution. Future
Science OA Volume 7, Number 6, Pages FSO698, 2021, https://doi.org/10.2144/fsoa-2020-0206
[10] Ravindhar NV, Anand, Hariharan Shanmugasundaram, Ragavendran, Godfrey Winster. Intelligent Diagnosis of Cardiac Disease Prediction using Machine
Learning. Volume-8 Issue-11, September 2019, ISSN: 2278-3075 (Online). Page No: 1417-1421. DOI: 10.35940/ijitee.J9765.0981119
[11] https://archive.ics.uci.edu/ml/datasets/Heart+Disease

©IJRASET: All Rights are Reserved 2940

View publication stats