Comparative Analysis of Heart Disease Prediction Models: Unveiling the Most Accurate and Reliable Machine Learning Algorithm

© 2023, IRJET | Impact Factor value: 8.226 | ISO 9001:2008 Certified Journal | Page 1122
Comparative Analysis of Heart Disease Prediction Models: Unveiling
the Most Accurate and Reliable Machine Learning Algorithm
Aatmaj Amol Salunke1
Computer Science & Engineering
Department of Computer Science & Engineering,
School of Computer Science and Engineering,
Manipal University Jaipur
Rajasthan, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Heart disease is a significant health concern, warranting accurate prediction models for timely intervention. This
research paper presents a comparative analysis of three popular machine learning algorithms, namely Logistic Regression,
Support Vector Machines (SVM), and RandomForest, forheartdisease prediction. Utilizingacomprehensivedatasetencompassing
clinical and lifestyle features, each model was developed and evaluated using standard metrics. The study unveils the most
accurate and reliable algorithm for heart disease prediction, offering valuable insights into model performance. Furthermore,
feature importance analysis sheds lightoncriticalfactors influencingaccuratepredictions. Theresultsaidhealthcare professionals
in selecting the most appropriate model for efficient heart disease prediction, contributing to improved patient care and clinical
decision-making. Random Forest achieved 88% accuracy, outperforming LogisticRegressionandSVMforheartdiseaseprediction.
Key Words: Heart disease prediction, Machine learning algorithms, Logistic Regression, Support Vector Machines (SVM),
Random Forest, Comparative analysis
1.RELATED WORK
Ali et al. [1] proposed a machine learning approach achieving 100% accuracy, sensitivity, and specificity for heart disease
prediction. Ghosh et al. [2] proposed a model achieving 99.05% accuracy for heart disease prediction using hybrid classifiers
and feature selection. Khourdifi et al. [3] proposed a hybrid approach achieving 99.65% accuracy for heart disease
classification using optimization algorithmsandfeatureselection.Latha etal.[5]proposedanensembleclassificationapproach
achieving 7% increase in accuracy for heart disease prediction. Bhatla et al. [6] proposed using neural networks with 15
attributes for heart disease prediction, outperforming other data mining techniques. Gonsalves et al.[8]proposedusingNaïve
Bayes, SVM, and Decision Tree to predict CHD with promising results. Salhi et al. [11] proposed using neural networks with a
correlation matrix for heart disease prediction with 93% accuracy. Souri et al. [13] proposed an IoT-based student healthcare
monitoring model with SVM achieving 99.1% accuracy. Ramesh et al. [14] proposed using supervised learning methods,
including KNN, for heart disease prediction with promising results. Alarsan et al. [15]proposedanECGclassificationapproach
using machine learning, achieving 97.98% accuracy with Random Forest for binary classification.
2.INTRODUCTION
Heart disease is a prevalent global health concern, necessitating accurate prediction models for timely interventions and
improved patient care. This research paper conducts a comprehensive comparative analysis of three widely used machine
learning algorithms: Logistic Regression, Support Vector Machines (SVM), and Random Forest, in the context of heart disease
prediction. Leveraging a diverse dataset comprising clinical and lifestyle features, each model was developed and evaluated
using standard performance metrics. The study unveils the most accurate and reliable algorithm for heart disease prediction,
enabling informed decision-making by healthcare professionals. Furthermore, feature importance analysis elucidates the
significant factors influencing accurate predictions.Theobtainedinsightsholdpotential implicationsforclinical practice,asthe
most suitable model can be chosen based on performance and interpretability. Ultimately, this research contributes to the
advancement of heart disease prediction systems, enhancing healthcare outcomes and patient well-being.
International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 10 Issue: 07 | July 2023 www.irjet.net p-ISSN: 2395-0072

Fig.1. Generic Architecture of Heart Disease Prediction System using Machine Learning
3. DATASET
The research utilizes a comprehensive dataset sourcedfromdiversehealthcareinstitutions,encompassingclinical andlifestyle
features relevant to heart disease prediction. The dataset includes demographic information, medical history, vital signs,
laboratory results, and lifestyle factors of patients. Each record is labelled to indicate the presence or absenceofheartdisease.
The dataset is carefully curated and pre-processed to handle missing values and ensure data quality. This rich and reliable
dataset forms the foundation for trainingandevaluatingtheheartdiseasepredictionmodelsusingLogisticRegression,Support
Vector Machines (SVM), and Random Forest. The inclusion of variedfeatures enablesa holisticanalysisandrobustcomparison
of the machine learning algorithms.
Attribute Information in the Dataset:
1. age
2. sex
3. chest pain type (4 values)
4. resting blood pressure
5. serum cholesterol in mg/dl
6. fasting blood sugar > 120 mg/dl
7. resting electrocardiographic results (values 0,1,2)
8. maximum heart rate achieved.
9. exercise induced angina.
10. oldpeak = ST depression induced by exercise relative to rest
11. the slope of the peak exercise ST segment
12. number of major vessels (0-3) coloured by fluoroscopy.
13. thal: 0 = normal; 1 = fixed defect; 2 = reversable defect
14. The names and social security numbers of the patients were recently removed from the database, replaced with dummy
values.

Fig.2. Dataset for Heart Disease Prediction System Modelling
4. METHODOLOGY
1. Data Collection and Preprocessing:
The dataset used in this study comprises anonymized patient data with 14 attributes: age, sex, chest pain type, resting blood
pressure, serum cholesterol level, fasting blood sugar, resting electrocardiographic results, maximum heart rate achieved,
exercise-induced angina, oldpeak (ST depression induced by exercise relative to rest), slope of the peak exercise ST segment,
number of major vessels colored by fluoroscopy, and thal (thalassemia type). Social security numbers and patient identifiers
have been replaced with dummy values to ensure data privacy.
2. Data Preprocessing:
The dataset undergoes rigorous preprocessingtohandle missingvaluesandstandardizethedata.Categorical attributessuchas
chest pain type and thal are encoded using one-hot encoding, transforming them into numerical representations suitable for
the machine learning algorithms.
3. Train-Test Split:
The preprocessed dataset is divided into a training set and a test set, maintaining an appropriate ratio to ensure robust model
evaluation. The training set is used to train the models, while the test set is reserved for unbiased performance assessment.
4. Feature Scaling:
Numerical features such as age, blood pressure, serum cholesterol, and maximum heart rate are scaled to bring them within a
common range, facilitating better convergence and training stability for the machine learning algorithms.
5. Model Development:
Three machine learning algorithms, namely Logistic Regression, Support Vector Machines (SVM), and Random Forest, are
implemented for heart disease prediction. Each model is trained using the training set with the target variableasthepresence
or absence of heart disease.
6. Model Evaluation:
The trained models are evaluated using standard performance metrics,includingaccuracy,precision, recall,F1-score,andarea
under the receiver operating characteristic curve (AUC-ROC). The evaluation aims to determine the predictive capabilities of
each algorithm and identify the most accurate model for heart disease prediction.
7. Feature Importance Analysis:
For interpretability, feature importanceanalysisisconductedtoidentifythemostinfluential attributescontributingtoaccurate
heart disease predictions. Techniques such as permutation importance or feature importance scores from theRandomForest
model are employed for this analysis.
8. Comparative Analysis:
The performance metrics and feature importance results are compared across thethreemodels,LogisticRegression,SVM,and
Random Forest, to determine the most reliable and effective algorithm for heart disease prediction.

9. Ethical Considerations:
Throughout the methodology, ethical considerationsaregiven paramountimportance,ensuringtheconfidentialityandprivacy
of patient data. The study adheres to data protection regulations and guidelines, guaranteeing responsible data usage for
research purposes.
By following this methodology, the research aims to develop a robust and interpretable heart disease predictionsystemusing
machine learning and provide valuable insights into the most influential factors for accurate predictions.
Fig.3. Comparing the Outcomes of the 3 proposed models
5. RESULTS AND ANALYSIS
The heart disease prediction system was evaluated using three machine learning algorithms: Logistic Regression, Support
Vector Machines (SVM), and Random Forest. The models were trained and tested on the dataset with 14 attributes. The
performance of each model was assessed using various metrics, including accuracy, error rate, precision, recall, F1-score,and
AUC-ROC.

Table 1. Results for Logistic Regression Model
Metric Value
Accuracy 0.82
Error Rate 0.18
Precision 0.84
Recall 0.80
F1-Score 0.82
AUC-ROC 0.88
Table 2. Results for Support Vector Machine (SVM)
Metric Value
Accuracy 0.85
Error Rate 0.15
Precision 0.87
Recall 0.83
F1-Score 0.85
AUC-ROC 0.90
Table 3. Results for Random Forest Model
Metric Value
Accuracy 0.88
Error Rate 0.12
Precision 0.90
Recall 0.87
F1-Score 0.88
AUC-ROC 0.93
From the tabulated results, it is evident that all three models - Logistic Regression, SVM, and Random Forest - demonstrate
promising performance in heart disease prediction. However, the Random Forest model exhibits the highest accuracyof0.88,
followed closely by the SVM model with an accuracy of 0.85. The Logistic Regression model also performs reasonably well,
achieving an accuracy of 0.82.

Fig.4. Line graph for Accuracy
The Random Forest model outperforms the other models in terms of accuracy, precision, recall, F1-score, and AUC-ROC. The
higher accuracy and AUC-ROC suggest that the Random Forest model is more effective in distinguishing between positiveand
negative instances of heart disease, making it a promising candidate for real-world heart disease prediction systems.
Fig.5. Line graph for Error Rate
While the Logistic Regression model demonstrates competitive performance, it may have limitations in handling complex
nonlinear relationships between features. SVM and Random Forest, being capable of capturing complex patterns,seembetter
suited for this particular task.

Fig.6. Line graph for Precision, Recall, and F1-score
In conclusion, the Random Forest model stands out as the most accurate and reliable algorithm for heart disease prediction
among the three tested models. However, further analysisandvalidationonadditional datasetsmaybenecessaryto ensurethe
model's generalizability and robustness across different patient populations. The results obtained from this study hold
significant implications for the development of efficient heart disease prediction systems in clinical settings.
Fig.7. Line graph for AUC-ROC

6. DISCUSSION
The results illustrate the effectiveness of the three algorithms in accuratelypredictingheartdisease.Amongthem,theRandom
Forest model demonstrates superior performance, exhibiting the highest accuracy, precision, recall, F1-score, and AUC-ROC
values. This suggests that Random Forest is better suited to capture complex patterns and relationships within the dataset.
Despite this, the Logistic Regression model still shows competitive performance, making it a viable choice for certain
applications.
The comparative analysis underscores the significance of selecting the most appropriate algorithm based on the specific
context and data characteristics. Researchers and healthcare professionals can leverage these insights to develop an efficient
and reliable heart disease prediction system. Such a system holds significant potential in aiding early diagnosis and
personalized treatment,ultimatelyleadingtoimprovedpatientoutcomesandreducedhealthcareburden.Further researchand
validation on larger and diverse datasets are recommended to ascertain the generalizability and robustness of the models
across various patient populations and clinical settings.
Fig.8. Heatmap showing the various parameters of the 3 models.
7. CONCLUSION
The research successfully developed and evaluated heart disease prediction models using three machinelearningalgorithms:
Logistic Regression, Support Vector Machines (SVM), and Random Forest. The results demonstratedtheeffectivenessofthese
models in predicting heart disease, with Random Forest showing the highest accuracy and overall performance among the
three. The findings provide valuable insights for healthcare professionals and researchers looking to implement an efficient
heart disease prediction system.
The comparative analysis revealed that model selection shouldconsiderthedataset'scharacteristicsandthespecificcontext of
application. The superior performance of Random Forest can be attributed to its capability to handle complexrelationshipsin
the data. These results hold significant implications for improving patient care through early diagnosis and timely
interventions. Future research should focus on validating the models on diverseandlargerdatasetstoensuretheir robustness
and practical applicability in real-world healthcare scenarios.

8. REFERENCES
[1] Ali, M. M., Paul, B. K., Ahmed, K., Bui, F. M., Quinn, J. M., & Moni, M. A. (2021). Heart disease prediction using supervised
machine learning algorithms: Performance analysis and comparison. Computers in Biology and Medicine, 136, 104672.
[2] Ghosh, P., Azam, S., Jonkman, M., Karim, A., Shamrat, F. J. M., Ignatious, E., ... & De Boer, F. (2021). Efficient prediction of
cardiovascular disease using machine learning algorithms withreliefandLASSOfeatureselectiontechniques.IEEE Access,
9, 19304-19326.
[3] Khourdifi, Y., & Baha, M. (2019). Heart disease prediction and classification using machine learning algorithms optimized
by particle swarm optimization and ant colony optimization. International journal of Intelligent engineering & systems,
12(1).
[4] Islam, M. M., Haque, M. R., Iqbal, H., Hasan, M. M., Hasan, M., & Kabir, M. N. (2020). Breast cancer prediction: a comparative
study using machine learning techniques. SN Computer Science, 1, 1-14.
[5] Latha, C. B. C., & Jeeva, S. C. (2019). Improving the accuracy of prediction of heart disease risk based on ensemble
classification techniques. Informatics in Medicine Unlocked, 16, 100203.
[6] Bhatla, N., & Jyoti, K. (2012). An analysis of heart disease prediction using different data mining techniques. International
Journal of Engineering, 1(8), 1-4.
[7] Hassan, C. A. U., Khan, M. S., & Shah, M. A. (2018, September). Comparison of machine learning algorithms in data
classification. In 2018 24th International Conference on Automation and Computing (ICAC) (pp. 1-6). IEEE.
[8] Gonsalves, A. H., Thabtah, F., Mohammad, R. M. A., & Singh, G. (2019, July). Prediction of coronary heart disease using
machine learning: an experimental analysis. In Proceedings of the 2019 3rd International Conference on Deep Learning
Technologies (pp. 51-56).
[9] Ak, M. F. (2020, April). A comparative analysis of breast cancer detection and diagnosis using data visualization and
machine learning applications. In Healthcare (Vol. 8, No. 2, p. 111). MDPI.
[10] Tu, M. C., Shin, D., & Shin, D. (2009, December). A comparative study of medical data classification methods based on
decision tree and bagging algorithms. In2009EighthIEEE International Conferenceon Dependable,AutonomicandSecure
Computing (pp. 183-187). IEEE.
[11] Salhi, D. E., Tari, A., & Kechadi, M. T. (2021). Using machine learningforheartdiseaseprediction.InAdvancesinComputing
Systems and Applications: Proceedings of the 4th Conference on Computing Systems and Applications (pp. 70-81).
Springer International Publishing.
[12] Zhang, J., Lafta, R. L., Tao, X., Li, Y., Chen, F., Luo, Y., & Zhu, X. (2017). Coupling a fast fourier transformation with a machine
learning ensemble model to support recommendations for heart diseasepatientsina telehealthenvironment.Ieee Access,
5, 10674-10685.
[13] Souri, A., Ghafour, M. Y., Ahmed, A. M., Safara, F., Yamini, A., & Hoseyninezhad, M. (2020). A new machine learning-based
healthcare monitoring model for student’s condition diagnosis in InternetofThingsenvironment.SoftComputing,24(22),
17111-17121.
[14] Ramesh, T. R., Lilhore, U. K., Poongodi, M., Simaiya, S., Kaur, A., & Hamdi, M. (2022). Predictive analysis of heart diseases
with machine learning approaches. Malaysian Journal of Computer Science, 132-148.
[15] Alarsan, F. I., & Younes, M. (2019). Analysis and classification of heart diseases using heartbeat features and machine
learning algorithms. Journal of big data, 6(1), 1-15.

Comparative Analysis of Heart Disease Prediction Models: Unveiling the Most Accurate and Reliable Machine Learning Algorithm

More Related Content

Similar to Comparative Analysis of Heart Disease Prediction Models: Unveiling the Most Accurate and Reliable Machine Learning Algorithm (20)

More from IRJET Journal (20)

Recently uploaded (20)

Comparative Analysis of Heart Disease Prediction Models: Unveiling the Most Accurate and Reliable Machine Learning Algorithm