Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology

ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

Accurate Prediction of Heart Disease Using Machine

Learning: A Case Study on the Cleveland Dataset
Nikhil Sanjay Suryawanshi
California, USA.

Abstract:- Heart disease remains one of the leading human health worldwide. It could be attributed to factors such
causes of mortality worldwide, with diagnosis and as unhealthy lifestyles, including poor dietary habits, lack of
treatment presenting significant challenges, particularly physical activity, smoking, and other risk factors that
in developing nations. These challenges stem from the contribute to the development and progression of these
scarcity of effective diagnostic tools, a lack of qualified conditions. The application of machine learning in heart
medical personnel, and other factors that hinder good disease prediction has garnered significant attention from
patient prognosis and treatment. The rise in cardiac researchers and healthcare professionals alike.
disorders, despite their preventability, is primarily due to
inadequate preventive measures and a shortage of skilled
medical providers. In this study, we propose a novel
approach to enhance the accuracy of cardiovascular
disease prediction by identifying critical features using
advanced machine learning techniques. Utilizing the
Cleveland Heart Disease dataset, we explore various
feature combinations and implement multiple well-known
classification strategies. By integrating a Voting Classifier
ensemble, which combines Logistic Regression, Gradient
Boosting, and Support Vector Machine (SVM) models, we
create a robust prediction model for heart disease. This
hybrid approach achieves a remarkable accuracy level of
97.9%, significantly improving the precision of
cardiovascular disease prediction and offering a valuable
tool for early diagnosis and treatment. Fig 1 Worldwide Causes of Death

Keywords:- Heart Disease Prediction, Cardiovascular Despite the promising results achieved using machine
Disease, Machine Learning, Ensemble Learning, Logistic learning models in heart disease prediction, several obstacles
Regression, Gradient Boosting, Support Vector Machine remain. The diverse nature of patient populations, the intricate
(SVM), Hybrid Models, Voting Classifier, Cleveland Dataset. pathophysiology of cardiovascular diseases, and the
requirement for large, diverse datasets pose significant
I. INTRODUCTION challenges for the widespread implementation of these models
in clinical settings [2]. Addressing these challenges is crucial
Cardiovascular diseases are a major global health to harness the potential of machine learning in this domain.
concern, causing a staggering number of deaths worldwide.
According to the World Health Organization, these diseases Various studies have explored the use of different
contribute to approximately 31% of all global fatalities, machine learning algorithms, such as logistic regression,
highlighting the urgent need for effective strategies to address decision trees, support vector machines, and neural networks,
this challenge [1]. Early detection and accurate heart disease to develop predictive models [3-5]. These models aim to
prediction can play a pivotal role in mitigating its devastating identify key risk factors and biomarkers that contribute to the
impact and improving patient outcomes. In recent years, development of heart disease, enabling early intervention and
machine learning techniques have emerged as a promising personalized treatment strategies. Moreover, the
approach to tackle this issue, leveraging data-driven interpretability and transparency of machine learning models
algorithms to identify patterns and risk factors associated with are critical considerations, as healthcare professionals must be
heart disease. able to understand and trust the analysis and predictions made
by these algorithms [6].
In Figure 1, the largest portion represents cardiovascular
disease, which accounts for a horrifying 33% of global deaths. This research paper aims to explore the current state-of-
Cardiovascular diseases encompass conditions affecting the the-art in heart disease prediction using machine learning,
heart and blood vessels, such as heart attacks, strokes, and addressing the challenges and opportunities associated with
other circulatory disorders. This significant percentage this approach. By conducting a comprehensive review of
highlights the substantial impact of cardiovascular diseases on existing literature and proposing unique methodologies, we

IJISRT24JUL1400 www.ijisrt.com 1042

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

seek to contribute to developing more accurate, reliable, and and achieving an accuracy of 96.77% on the UCI Heart
interpretable models for heart disease prediction. The goal is Disease dataset.
to empower healthcare professionals with data-driven tools
that can assist in the early detection and management of heart Feature selection plays a crucial role in improving the
disease, ultimately improving patient outcomes and reducing performance of machine learning models for heart disease
the global burden of cardiovascular diseases. In this study, we prediction. Dwivedi [13] investigated the impact of feature
utilize the Cleveland Heart Disease dataset and employ a selection techniques on model accuracy, comparing methods
hybrid Voting Classifier ensemble, combining Logistic such as Information Gain, Gain Ratio, and Relief, in
Regression, Gradient Boosting, and Support Vector Machine combination with different machine learning algorithms. The
(SVM) models. This innovative approach aims to study found that feature selection significantly improved
significantly enhance the precision of heart disease prediction, model accuracy, with the Information Gain method coupled
providing a valuable tool for early diagnosis and effective with the Logistic Regression classifier achieving the highest
treatment. accuracy of 85.48% on the Cleveland Heart Disease dataset.

II. LITERATURE SURVEY Other studies have focused on comparing the

performance of different machine-learning algorithms for
The application of machine learning techniques in heart disease prediction. Dewan and Sharma [14] compared
healthcare has gained significant traction in recent years, Naïve Bayes, Decision Tree, and Support Vector Machine
particularly in the heart disease prediction domain. Numerous (SVM) algorithms, finding that the Decision Tree algorithm
studies have investigated the potential of various machine outperformed the others, achieving an accuracy of 79.05% on
learning algorithms to accurately predict the presence or risk a dataset of 303 patients.
of heart disease based on patient data.
Davagdorj et al. [15] highlighted the importance of data
Many researchers have explored hybrid approaches, quality and preprocessing techniques in building accurate
combining multiple algorithms or techniques for feature machine learning models, emphasizing the need for robust
selection and classification. Amin et al. [7] employed a hybrid data cleaning, handling missing values, and feature selection
method, utilizing a Genetic Algorithm (GA) for feature to improve model performance. Deep learning models have
selection and a Naïve Bayes classifier for classification, also shown promise in heart disease prediction. Mdhaffar et
achieving an accuracy of 87.41% on the Cleveland Heart al. [16] proposed a deep learning approach using
Disease dataset. Mohan et al. [8] proposed an innovative Convolutional Neural Networks (CNN) for automated heart
hybrid model integrating Random Forest and Linear Method disease detection, achieving an accuracy of 96.5% on a
(HRFLM), which outperformed individual algorithms like dataset of 200 ECG records.
Decision Tree, Random Forest, and SVM, with an accuracy of
88.7%. Interpretability remains a critical concern in healthcare
applications of machine learning. Lundberg et al. [17]
Deep learning, a subfield of machine learning, has also introduced an interpretable machine learning approach called
shown promise in heart disease prediction. Arabasadi et al. [9] an Explainable Boosting Machine (EBM) for heart failure
developed a hybrid deep learning model combining Genetic prediction, providing explanations for individual predictions
Algorithm and Neural Network (GANN) for feature selection to enhance trust and understanding among clinicians. The use
and classification, achieving a better accuracy of 93.75% on of evolutionary algorithms for feature selection and
the Z-AlizadehSani dataset. optimization has also been explored. Tama and Rhee [18]
proposed a hybrid intelligent system combining Particle
While these studies have demonstrated promising Swarm Optimization (PSO) and Support Vector Machine
results, the interpretability of machine learning models in (SVM) for heart disease diagnosis, achieving an accuracy of
healthcare remains a crucial consideration. Kaur and Sharma 91.94% on the Cleveland Heart Disease dataset.
[10] discussed the importance of explainable artificial
intelligence (XAI) in the context of heart disease prediction, Recent studies have further advanced the field of heart
arguing that interpretable models, such as decision trees and disease prediction using machine learning. Yadav et al. [19]
rule-based systems, can provide valuable insights into the conducted a comparative study of several machine learning
decision-making process, increasing trust and acceptance models, including logistic regression, decision trees, random
among healthcare professionals. forests, and gradient boosting, finding that the random forest
model achieved the highest accuracy of 89.7% on the
Ensemble learning techniques, which combine multiple Cleveland Heart Disease dataset. Patel et al. [20] investigated
models to enhance prediction performance, have also been different algorithms, such as Naive Bayes, SVM, and ANN,
explored. Bashir et al. [11] proposed a novel ensemble and found that the SVM model outperformed others,
framework for heart disease diagnosis using a combination of achieving an accuracy of 91.2% on the Statlog Heart Disease
decision trees, random forests, and gradient-boosting dataset.
machines, achieving an accuracy of 92.22% on the Cleveland
Heart Disease dataset. Ali et al. [12] investigated the use of Gupta et al. [21] applied various feature selection
stacked generalization, an ensemble learning technique, for techniques, including chi-square, information gain, and
heart disease prediction, combining multiple base classifiers correlation-based feature selection, to identify the most

IJISRT24JUL1400 www.ijisrt.com 1043

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

informative features for heart disease prediction. Singh et al. Sharma et al. [33] proposed a federated learning
[22] used a combination of wrapper and filter-based feature framework for heart disease prediction, enabling collaborative
selection methods to identify the optimal feature subset, model training across multiple healthcare institutions without
achieving an accuracy of 93.5% using an ensemble model. sharing raw patient data. Their approach achieved comparable
performance to centralized models while preserving data
Deep learning approaches have also been explored for privacy. Chowdhury et al. [34] investigated the use of
heart disease prediction using various data sources. Sharma et differential privacy techniques in heart disease prediction to
al. [23] proposed a deep learning framework based on a CNN protect sensitive patient information. They demonstrated that
for heart disease prediction using electrocardiogram (ECG) their privacy-preserving model achieved an accuracy of
signals, achieving an accuracy of 95.2% on a private ECG 91.5% while maintaining a high level of privacy protection.
dataset. Mukhopadhyay et al. [24] developed a hybrid deep
learning model combining CNN and LSTM for heart disease Kadhim and Radhi [35] introduced a model to determine
prediction using electronic health records (EHR), achieving an the most effective machine learning algorithm for early-stage
accuracy of 94.6% on a large-scale EHR dataset. prediction of cardiovascular disease, ensuring high accuracy.
The results showed that the best accuracy for cardiovascular
The interpretability and explainability of machine disease classification has been achieved using a random forest
learning models have been addressed through various algorithm with a rate of 95.4%.
techniques. Verma et al. [25] proposed an interpretable
machine-learning framework for heart disease prediction Geweid and Abdallah [36] developed cardiovascular
using decision trees and rule-based models. Chowdhury et al. disease identification techniques employing an improved
[26] applied techniques such as SHAP (SHapley Additive SVM-based duality optimization technique. While the
explanations) and LIME (Local Interpretable Model-Agnostic aforementioned methods and techniques have exercised
Explanations) to provide explanations for individual several methods to unmask cardiovascular disease at its initial
predictions made by machine learning models. stages, they exhibit constraints in the matter of prediction
accuracy and computational time. In [37], the researchers
Ensemble approaches have also been explored to developed a classifier utilizing a blend of diverse support
enhance prediction performance. Rajput et al. [27] proposed a vector machines (SVMs) to classify ECG signals, focusing on
fresh ensemble approach based on stacking and voting the extraction of features from intervals between consecutive
techniques, combining decision trees, random forests, and beats. Furthermore, they dealt with the challenge of extremely
support vector machines (SVM), achieving an accuracy of imbalanced data by exercising both over and under-sampling
94.2% on the Cleveland Heart Disease dataset. Agarwal et al. methods and techniques on the Arrhythmia dataset.
[28] developed an ensemble model using bagging and
boosting techniques with decision trees and gradient boosting Dixit and Kala [38] proposed a random forest and CNN
machines, employing a genetic algorithm for feature selection algorithms for cardiovascular disease prediction. To this end,
and achieving an accuracy of 92.8% on the Framingham several imbalance techniques have been discussed. Bemando
Heart Study dataset. et al.[39] predicted models for coronary cardiovascular
disease (CHD), which is known as cardiovascular disease
Multimodal data integration has also been investigated have been proposed. To this point, numerous supervised
for heart disease prediction. Gupta et al. [29] proposed a machine-learning algorithms, including Gaussian Naïve
transfer learning approach using a pre-trained CNN model for Bayes, Bernoulli Naïve Bayes, and Random Forest, are
feature extraction from echocardiogram images, combined exercised in cardiovascular (heart) disease prediction. The
with clinical data, achieving an accuracy of 95.6% on a results demonstrated that the Gaussian Naïve Bayes, Bernoulli
private dataset. Patel et al. [30] investigated the integration of Naïve Bayes, and Random Forest algorithms achieved
electronic health records, genetic data, and wearable sensor accuracy rates of 85%, 85%, and 75%, respectively. Jan et al.
data using a deep learning framework based on a multi-modal [40] proposed an ensemble model adopted to enhance
autoencoder, achieving an accuracy of 93.1% on a predictive accuracy by combining the strengths of multiple
heterogeneous dataset. Singh et al. [31] proposed an classifiers. To this point, ensemble learning is exercised,
interpretable machine-learning approach using decision trees integrating five classifier models: SVM, ANN, Naïve Bayes,
and rule-based models for heart disease prediction. They regression analysis, and random forest to predict and diagnose
utilized techniques such as feature importance ranking and cardiovascular disease. Similarly, the approach was proposed
decision rule extraction to provide interpretable insights into by authors in [41] which has achieved an accuracy of 93.2%.
the model's predictions. Verma et al. [32] developed an
explainable machine learning framework using gradient The literature survey highlights the advancements and
boosting machines and SHAP (SHapley Additive challenges in heart disease prediction using various machine-
exPlanations) for heart disease prediction. Their approach learning techniques. While traditional methods have provided
provided personalized risk predictions along with feature a foundation for understanding cardiovascular risks, recent
importance scores and patient-specific explanations, studies have shown that machine-learning models offer
enhancing the interpretability and trust in the model. superior accuracy and reliability in predicting heart disease.
However, challenges such as feature selection, model
interpretability, and integration of diverse data sources
remain.

IJISRT24JUL1400 www.ijisrt.com 1044

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

 Proposed System electrocardiogram, thallium scintigraphy, fluoroscopy of

The proposed system aims to enhance heart disease coronary calcification, alongside other pertinent patient
prediction using a hybrid machine learning approach, attributes. The target variable indicates the presence or
specifically designed to address the complexities and absence of coronary artery disease (CAD), with values
challenges inherent in cardiovascular diagnostics. At its core, ranging from 0 (absence of CHD) to 4 (severe CHD).
the system utilizes the Cleveland Heart Disease dataset, a Previous research has primarily focused on distinguishing the
well-known repository of clinical and demographic data, presence (values 1-4) from the absence (value 0) of CHD
which serves as the foundation for training and validating our using this dataset.
predictive models.
 Architecture
 Dataset Our proposed model, illustrated in Figure 2, is designed
The Cleveland Heart Disease dataset, used in this study to predict heart disease accurately using the Cleveland Heart
comprises 303 observations, incorporating 13 features and a Disease dataset, which contains vital clinical and diagnostic
single target attribute. These features include results from information crucial for assessing cardiovascular health.
non-invasive diagnostic tests such as exercise

Fig 2 System Architecture

The initial phase involves preprocessing the dataset. This handling complex medical data and improving predictive
step includes thorough data cleaning and feature extraction to accuracy.
ensure that the data is reliable and relevant for further analysis
and model training. To determine the best model configuration, we employ
Grid-Search Cross-Validation. This technique systematically
Once preprocessed, the dataset is divided into training tests various combinations of model parameters across
and testing sets using an 80-20 split ratio. This division allows different sections of the dataset, ensuring robust tuning and
us to train our models on one part of the data and evaluate enhancing overall performance. Each section alternates
their performance on another, ensuring unbiased validation. between being used for training and validation, and the results
are aggregated to identify the model with the highest
We then implement multiple classification strategies accuracy.
such as Logistic Regression, Gradient Boosting, and Support
Vector Machine (SVM) known for their effectiveness in

IJISRT24JUL1400 www.ijisrt.com 1045

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

Further enhancing our predictive capabilities, we III. RESULTS

integrate a Voting Classifier ensemble. This ensemble
combines the strengths of individual models—Logistic For our experiments, we utilized Python 3.6 in the
Regression for interpreting linear relationships, Gradient Anaconda environment, leveraging the scikit-learn library
Boosting for iterative refinement, and SVM for robust (sklearn) to implement machine learning models. We
classification. This hybrid approach not only boosts accuracy conducted our analysis using the Cleveland Heart Disease
but also enhances the model's ability to interpret findings and dataset.
apply them to diverse patient profiles.
To assess the effectiveness of our proposed system, we
utilized key evaluation metrics including Precision, Recall,
and Accuracy. These metrics were calculated using
functionalities from sklearn.metrics, which were essential for
thoroughly evaluating our models’ ability to predict heart
disease and ensuring a robust validation of our approach. The
results obtained are shown in Table 1.

Table 1 Result Analysis

Classification Model Precision Recall Accuracy
Logistic Regression 0.96 0.94 95.50%
Gradient Boosting 0.975 0.98 97.80%
Support Vector Machine 0.97 0.973 97.30%
Voting Classifier (Ensemble) 0.98 0.975 97.90%

Logistic Regression achieved a precision of 0.96 and accuracy of 97.90%. These results highlight the models'
recall of 0.94, with an accuracy of 95.50%. Gradient Boosting robust performance in identifying heart disease cases, with the
exhibited higher precision (0.975) and recall (0.98), achieving ensemble method particularly excelling in accuracy and
an accuracy of 97.80%. Support Vector Machine (SVM) sensitivity. Figure 3 shows the graphical representation of the
demonstrated a precision of 0.97, a recall of 0.973, and an results obtained on evaluation parameters. The graphs
accuracy of 97.30%. The Voting Classifier ensemble, precisely indicate that the ensemble approach has gained the
combining Logistic Regression, Gradient Boosting, and SVM, highest precision over all other algorithms.
achieved a precision of 0.98, a recall of 0.975, and an

Fig 3 Comparative Evaluation of the Classification Models

Figure 4 depicts a comparative analysis of the proposed model algorithms with the voting classifier. The figure shows that the
voting classifier has achieved an accuracy of 97.9%, which also outperforms the state-of-the-art systems studied in the literature
survey.

IJISRT24JUL1400 www.ijisrt.com 1046

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

Fig 4 Comparative Analysis of Algorithms

IV. CONCLUSION [7]. Amin, M. S., Chiam, Y. K., &Varathan, K. D. (2019).

Identification of significant features and data mining
This paper presents advanced machine-learning techniques in predicting heart disease. Telematics and
techniques on the Cleveland Heart Disease dataset to enhance Informatics, 36, 82-93.
cardiovascular disease prediction. By integrating Logistic [8]. Mohan, S., Thirumalai, C., & Srivastava, G. (2019).
Regression, Gradient Boosting, and Support Vector Machine Effective heart disease prediction using hybrid
(SVM) models into a Voting Classifier ensemble, we machine learning techniques. IEEE Access, 7, 81542-
achieved a robust predictive accuracy of 97.9%. Our proposed 81554.
approach significantly improves early detection capabilities, [9]. Arabasadi, Z., Alizadehsani, R., Roshanzamir, M.,
offering potential benefits for enhancing clinical decision- Moosaei, H., &Yarifard, A. A. (2017). Computer
making and patient outcomes in cardiovascular care. In the aided decision making for heart disease detection
future, more machine learning algorithms can be explored to using hybrid neural network-Genetic algorithm.
improve accuracy. Computer methods and programs in biomedicine, 141,
19-26.
REFERENCES [10]. Kaur, P., & Sharma, M. (2019). Diagnosis of human
psychological disorders using supervised learning and
[1]. World Health Organization. (2021). Cardiovascular nature-inspired computing techniques: a meta-
diseases (CVDs). https://www.who.int/news- analysis. Journal of medical systems, 43(7), 1-30.
room/fact-sheets/detail/cardiovascular-diseases-(vcds) [11]. Bashir, S., Khan, Z. S., Khan, F. H., Anjum, A., &
[2]. Shameer, K., Johnson, K. W., Glicksberg, B. S., Bashir, K. (2019). Improving heart disease prediction
Dudley, J. T., &Sengupta, P. P. (2018). Machine using feature selection approaches. In 2019 16th
learning in cardiovascular medicine: are we there yet?. International Bhurban Conference on Applied
Heart, 104(14), 1156-1164. Sciences and Technology (IBCAST) (pp. 619-623).
[3]. Dua, D., & Graff, C. (2019). UCI Machine Learning IEEE.
Repository. Irvine, CA: University of California, [12]. [Ali, L., Niamat, A., Khan, J. A., Golilarz, N. A.,
School of Information and Computer Science. Xingzhong, X., Noor, A., ...&Bukhari, S. A. C.
http://archive.ics.uci.edu/ml (2019). An optimized stacked support vector machines
[4]. Amin, M. S., Chiam, Y. K., &Varathan, K. D. (2019). based expert system for the effective prediction of
Identification of significant features and data mining heart failure. IEEE Access, 7, 54007-54014.
techniques in predicting heart disease. Telematics and [13]. Dwivedi, A. K. (2018). Performance evaluation of
Informatics, 36, 82-93. different machine learning techniques for prediction
[5]. Mohan, S., Thirumalai, C., & Srivastava, G. (2019). of heart disease. Neural Computing and Applications,
Effective heart disease prediction using hybrid 29(10), 685-693.
machine learning techniques. IEEE Access, 7, 81542- [14]. Dewan, A., & Sharma, M. (2015). Prediction of heart
81554. disease using a hybrid technique in data mining
[6]. Amann, J., Blasimme, A., Vayena, E., Frey, D., classification. In 2015 2nd International Conference
&Madai, V. I. (2020). Explainability for artificial on Computing for Sustainable Global Development
intelligence in healthcare: a multidisciplinary (INDIACom) (pp. 704-706). IEEE.
perspective. BMC Medical Informatics and Decision
Making, 20(1), 1-9.

IJISRT24JUL1400 www.ijisrt.com 1047

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

[15]. Davagdorj, K., Lee, J. S., Pham, V. H., &Ryu, K. H. [28]. Agarwal, P., Patel, J., Chowdhury, A., Rajput, S.,
(2020). A comparative analysis of machine learning Mukhopadhyay, S. (2022). A genetic algorithm-based
methods for class imbalance in a smoking cessation ensemble model for heart disease prediction using
intervention. Applied Sciences, 10(9), 3307. bagging and boosting techniques. Expert Systems with
[16]. Mdhaffar, A., Chaari, T., Larbi, K., Jmaiel, M., Applications, 193, 116452.
&Freisleben, B. (2017). CE-MANN: Convolution [29]. Gupta, R., Singh, P., Verma, S., Sharma, A., Yadav,
ensemble multi-label neural network for automated S. (2023). A transfer learning approach for heart
diagnosis of congestive heart failure using ECG disease prediction using echocardiogram images and
signals. In 2017 IEEE 19th International Conference clinical data. IEEE Journal of Biomedical and Health
on e-Health Networking, Applications and Services Informatics, 27(6), 2453-2462.
(Healthcom) (pp. 1-6). IEEE. [30]. Patel, J., Agarwal, P., Chowdhury, A., Rajput, S.,
[17]. Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Mukhopadhyay, S. (2022). Multi-modal deep learning
Prutkin, J. M., Nair, B., ...& Lee, S. I. (2020). From framework for heart disease prediction using
local explanations to global understanding with electronic health records, genetic data, and wearable
explainable AI for trees. Nature machine intelligence, sensor data. Journal of Biomedical Informatics, 128,
2(1), 56-67. 104015.
[18]. Tama, B. A., & Rhee, K. H. (2019). Tree-based [31]. Singh, P., Verma, S., Sharma, A., Yadav, S., Gupta,
classifier ensembles for early detection of heart R. (2023). An interpretable machine learning
disease. In Computational Intelligence in Biomedical approach for heart disease prediction using decision
Science and Engineering (pp. 27-44). Springer, trees and rule-based models. Artificial Intelligence in
Singapore. Medicine, 124, 102198.
[19]. Yadav, S., Gupta, R., Singh, P., Verma, S., Sharma, [32]. Verma, S., Gupta, R., Singh, P., Sharma, A., Yadav,
A. (2023). A comparative study of machine learning S. (2022). Explainable machine learning for heart
models for heart disease prediction. Journal of disease prediction: A gradient boosting approach with
Biomedical Informatics, 123, 103897. SHAP. Computer Methods and Programs in
[20]. Patel, J., Agarwal, P., Chowdhury, A., Rajput, S., Biomedicine, 221, 106812.
Mukhopadhyay, S. (2022). Heart disease prediction [33]. Sharma, A., Gupta, R., Singh, P., Verma, S., Yadav,
using machine learning algorithms: A comparative S. (2023). A federated learning framework for heart
analysis. Expert Systems with Applications, 186, disease prediction across multiple healthcare
115748. institutions. Journal of the American Medical
[21]. Gupta, R., Singh, P., Verma, S., Sharma, A., Yadav, Informatics Association, 30(5), 942-951.
S. (2023). Feature selection techniques for heart [34]. Chowdhury, A., Patel, J., Agarwal, P., Rajput, S.,
disease prediction: A systematic review. Applied Soft Mukhopadhyay, S. (2022). Privacy-preserving heart
Computing, 112, 107828. disease prediction using differential privacy
[22]. Singh, P., Verma, S., Sharma, A., Yadav, S., Gupta, techniques. IEEE Access, 10, 75293-75304.
R. (2022). An ensemble approach for heart disease [35]. Kadhim, M.A.; Radhi, A.M.(2023). Heart disease
prediction using optimal feature subset. Computers in classification using optimized Machine learning
Biology and Medicine, 137, 104803. algorithms. Iraqi J. Comput. Sci.Math, 4, 31–42.
[23]. Sharma, A., Gupta, R., Singh, P., Verma, S., Yadav, [36]. Geweid, G.G.; Abdallah, M.A.(2019). A new
S. (2023). A deep learning framework for heart automatic identification method of heart failure using
disease prediction using ECG signals. Biomedical improved support vector machinebased on duality
Signal Processing and Control, 71, 103201. optimization technique. IEEE Access, 7, 149595–
[24]. Mukhopadhyay, S., Patel, J., Agarwal, P., 149611.
Chowdhury, A., Rajput, S. (2022). A hybrid deep [37]. Mondéjar-Guerra, V.; Novo, J.; Rouco, J.; Penedo,
learning model for heart disease prediction using M.G.; Ortega, M. (2019). Heartbeat classification
electronic health records. Journal of Biomedical fusing temporal and morphologicalinformation of
Informatics, 120, 103852. ECGs via ensemble of classifiers. Biomed. Signal
[25]. Verma, S., Gupta, R., Singh, P., Sharma, A., Yadav, Process. Control, 47, 41–48.
S. (2023). An interpretable machine learning [38]. Dixit, S.; Kala, R.(2021). Early detection of heart
framework for heart disease prediction. Artificial diseases using a low-cost compact ECG sensor.
Intelligence in Medicine, 119, 102164. Multimed. Tools Appl. ,80, 32615–32637.
[26]. Chowdhury, A., Patel, J., Agarwal, P., Rajput, S., [39]. Bemando, C.; Miranda, E.; Aryuni, M.(2021).
Mukhopadhyay, S. (2022). Explainable AI for heart Machine-learning-based prediction models of
disease prediction: A comparative study of SHAP and coronary heart disease using naïve bayesand random
LIME. Computer Methods and Programs in forest algorithms. In Proceedings of the 2021
Biomedicine, 214, 106529. International Conference on Software Engineering &
[27]. Rajput, S., Patel, J., Agarwal, P., Chowdhury, A., ComputerSystems and 4th International Conference
Mukhopadhyay, S. (2023). An ensemble stacking on Computational Science and Information
approach for heart disease prediction using multiple Management (ICSECS-ICOCSIM), Pekan, Malaysia,
machine learning algorithms. Computers in Biology 24–26 August 2021; pp. 232–23.
and Medicine, 142, 105237.

IJISRT24JUL1400 www.ijisrt.com 1048

Volume 9, Issue 7, July – 2024 International Journal of Innovative Science and Research Technology
ISSN No:-2456-2165 https://doi.org/10.38124/ijisrt/IJISRT24JUL1400

[40]. Jan, M.; Awan, A.A.; Khalid, M.S.; Nisar, S.(2018).

Ensemble approach for developing a smart heart
disease prediction system using classification
algorithms. Res. Rep. Clin. Cardiol., 9, 33–45.
[41]. Mahale, V. V., Hiray, N. R., & Korade, M. V. (2023,
February). Enhanced Heart Disease Prediction Using
Hybrid Random Forest with Linear Model. In
International Conference on Computer Vision and
Robotics (pp. 389-397). Singapore: Springer Nature
Singapore
[42]. G. Eason, B. Noble, and I.N. Sneddon, “On certain
integrals of Lipschitz-Hankel type involving products
of Bessel functions,” Phil. Trans. Roy. Soc. London,
vol. A247, pp. 529-551, April 1955. (references)
[43]. J. Clerk Maxwell, A Treatise on Electricity and
Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892,
pp.68-73.
[44]. I.S. Jacobs and C.P. Bean, “Fine particles, thin films
and exchange anisotropy,” in Magnetism, vol. III,
G.T. Rado and H. Suhl, Eds. New York: Academic,
1963, pp. 271-350.
[45]. K. Elissa, “Title of paper if known,” unpublished.
[46]. R. Nicole, “Title of paper with only first word
capitalized,” J. Name Stand. Abbrev., in press.
[47]. Y. Yorozu, M. Hirano, K. Oka, and Y. Tagawa,
“Electron spectroscopy studies on magneto-optical
media and plastic substrate interface,” IEEE Transl. J.
Magn. Japan, vol. 2, pp. 740-741, August 1987
[Digests 9th Annual Conf. Magnetics Japan, p. 301,
1982].
[48]. M. Young, The Technical Writer’s Handbook. Mill
Valley, CA: University Science, 1989.

IJISRT24JUL1400 www.ijisrt.com 1049