An Efficient Artificial Intelligence-Based Early Prediction of Heart Attack Using Deep Learning CNN and SVM Models

265 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about An Efficient Artificial Intelligence-Based Early Prediction of Heart Attack Using Deep Learning CNN and SVM Models Syed Talal Musharraf Bahria University Lahore Campus, Lahore Pakistan/I2C Inc Email: syedtala[email protected] Mian Muhammad Masab (Corresponding Author) AI Engineer, Solutyics, Faisal Town, Lahore, 54000, Pakistan Email: [email protected] Nasir Ayub Deputy Head of Engineering Calrom Limited, M16EG, United Kingdom Email: nasir.ayy[email protected] Shamoon Murtaza Business Developer, CubicSol.inc, Canal Burg, Lahore, 54000, Pakistan Email: [email protected] Habib Ullah Department of Computer Science, Faculty of Computer Science & IT Superior University Lahore, 54000, Pakistan Email: [email protected] Ammar Ahmad Department of Information Technology, Faculty of Computer Science & IT, Superior University Lahore, 54000, Pakistan Email: [email protected] Muhammad Zunnurain Hussain Bahria University Lahore Campus Email: [email protected]du.pk Hamayun Khan Department of Computer Science, Faculty of Computer Science & IT, Superior University Lahore, 54000, Pakistan Email: [email protected] Artificial Intelligence (AI) has become an integral component of modern Cardiovascular disease diagnosis, treatment planning, and risk stratification. Its transformative potential is particularly crucial in developing regions. Cardiovascular issues cause most deaths worldwide. The World Health Organization states that about 17.9 million people die annually. Medical specialists spend considerable time diagnosing to identify the reason behind symptoms. Moreover, doctors cannot always keep up with their patients. Machine learning can assist doctors in predicting heart diseases by utilizing global health sector data, thus strengthening the base of this technology. The proposed system emphasizes temporal 266 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about data modeling through the integration of Parallel CNN (P-CNN) and Support Vector Machine (SVM) classifiers to enable the early detection of Heart Failure (HF). This hybrid architecture facilitates the development of an effective decision-support system for the accurate diagnosis of Congestive Heart Failure (CHF). The model was trained and evaluated using the UCI Machine Learning Heart Disease dataset, and its performance was benchmarked against state-of-the-art algorithms such as Artificial Neural Networks (ANNs) and Recurrent Networks (RNs). Empirical evaluations reveal that CardioHelp achieves outstanding predictive performance, with an accuracy of 97%, an F1-score of 93.4%, and a precision improvement of 0.8% compared to existing baseline models. Furthermore, when compared with advanced hybrid architectures—including CNN-RNN, CNN-LSTM, and CNN-BLSTM—the proposed framework exhibits accuracy enhancements of 2.35%, 1.34%, and 1.95%, Predicting heart disease accurately and quickly is important but difficult. Machine learning becomes highly effective in identifying and categorizing patients. Advanced machine learning and deep learning models, like SVMs and CNNs, can forecast patients' medical histories and general well-being. Heart disease predictions lower global death rates. While reducing the cost of medical care. Minimizing expensive hospital stays, long-term care, and urgent medical help can save lives and reduce family and societal stress. Using machine learning in healthcare can help find heart disease early, make sure patients get treatment on time, and help them live better lives. Moreover, using these technologies can help doctors make better decisions, increase the accuracy of diagnoses, and offer patients custom treatment plans Heart disease is recognized as one of the complex and deadly illnesses affecting humans globally. In the United States, 11.2% of adults are afflicted by heart disease. The prevalence is even higher among individuals aged 75 and older, reaching 37.3%. Research indicates that nearly 80% of heart disease cases might have been avoided through early detection and lifestyle changes [1]. Sadly, heart disease remains the leading cause of death, yet it is also one of the most preventable conditions. Taking timely precautions can significantly lower the risk of heart disease. Unfortunately, a lack of adequate knowledge within the general population results in unnecessary fatalities. The economic impact is substantial, with heart disease costing approximately $417.9 billion between 2020 and 2021 [2]. This figure encompasses healthcare services, medications, and productivity losses from premature deaths. It is crucial to recognize that there is a considerable variation in risk levels among individuals with heart disease. The mortality rates from heart disease differ based on sex, race, and ethnicity. Below are the percentages of all deaths attributed to heart disease in 2021, categorized by race and sex [3, 4]. Several factors contribute to the ongoing challenge of heart disease, including lifestyle-related risks such as poor diet, lack of physical activity, obesity, high blood pressure, smoking, and diabetes. Importantly, these factors are controllable. Additionally, variations in social and economic status, restricted access to preventive healthcare, and an aging population exacerbate the issue. Despite advancements in medical treatment and increased public awareness, the rise in heart disease rates indicates that these efforts are still inadequate [5, 6]. Deep-learning models that are based on a single deep-learning architecture are solo deep-learning models. Models that are produced by connecting two or more deep-learning architectures are hybrid deep-learning (HDL) models. 267 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Role of Machine Learning in Healthcare Machine learning (ML) and deep learning (DL) came to the rescue medical diagnostics has shown tremendous potential, particularly in predicting critical conditions like heart attacks. By processing and analyzing the large amount of data, Machine learning (ML) and multilayered neural network models are useful in predicting heart disease [7, 8]. The ability to examine the early risk of cardiovascular disease (CVD) can help improve the patient’s health and reduce the associated risks. Over the last decade, cardiovascular research has increasingly integrated advanced computational methods, including artificial intelligence (AI), machine learning (ML), and deep learning (DL) [9, 10]. In recent years, most cardiovascular prediction studies have implemented machine learning techniques (ML) like support vector machines (SVM), decision trees (DT), and naive Bayes (NB) [11, 12]. Still, it is actually challenging to harness the vast amount of clinical data and use it for model training. Implanting in clinical settings remains challenging, usually resulting in less accurate results [13, 14]. Deep Learning (DL), a specialized branch of machine learning (ML) that enables the processing of large amounts of data at a very high speed without any accuracy issues. In recent years, researchers and scientists have been continuously exploring the application of deep learning in the medical field, as they have achieved good results in other fields. For example, researchers have successfully used deep learning (DL) models, including, convolutional neural network (CNN), a Long Short-Term Memory (LSTM), and a CNN-LSTM to predict heart diseases using multiple data sets [15]. In addition, many datasets have been developed by researchers to predict heart disease. Remarkable datasets such as the UCI Heart Disease Dataset, the Framingham Heart Study, the Heart Failure Clinical Records, and cardiovascular datasets play a pivotal role in heart disease Studies claim that the proposed feature selection method needs to be able to optimally balance the most relevant features in the dataset to improve the predictions. The authors suggested that a new feature selection method is needed to choose the best mix of important features in a dataset so that prediction performance can be improved. Many heart disease prediction models are not suitable for real-world clinical use because their inner workings are difficult to interpret. Open-source models can help make these tools more 268 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about accessible, but some are still not available to the public. Without open-source software, it is harder for medical professionals to build predictive heart disease prediction systems [16]. Developing new techniques also requires collaboration among experts from different fields. It is important to figure out how these systems can work on real-time data without always needing a physician to supervise them. At present, many large medical institutions do not have strong predictive or early diagnosis systems in place [17]. Heart feature extraction The investigation involves the extraction of diverse attributes from the medical data obtained via healthcare devices. The improvement of prediction accuracy is one step towards extending classical DL approaches with specific and interpretable functions [16]. In addition, DL can also be effectively used along with other technologies to enhance accuracy, feature selection, and data classification. Additionally, stochastic methods might be considered to enhance the performance of heart disease prediction. Nevertheless, such hybrid models could decrease the predictive performance. Furthermore, there is a dearth of standard criteria for selecting performance evaluation procedures and metrics that can be employed in assessing the effectiveness of new technologies. Eq (4) Eq (5) Accordingly, the establishment of new test evaluation indicators and rigorous verification is beneficial to improve sterilization prediction models [18]. These datasets contain a vast number of attributes that help in the accurate prediction of heart diseases. Both changeable and non-changeable factors contribute to the prediction of heart disease. Nonchangeable factors include gender, ethnic background, and family history. On the other hand, changeable risk factors such as cholesterol level, blood pressure, unhealthy lifestyle habits, and smoking can be altered and controlled through specific measures and medical treatment [19, 20]. These characteristics influenced the creation of large datasets, and researchers have made considerable efforts to refine and enhance these datasets. Table 1 underscores the techniques for enhanced preventive care, early diagnosis, and lifestyle changes to mitigate both the health and financial impacts of heart attacks across the nation. Table 1. Summary of existing techniques on heart disease prediction Method Key Contributions Precision% Recall % F1 Score% Ref Traditional Survey of 92 90.3 91.9 [21] 269 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about ML techniques existing approaches for heart disease prediction Supervised learning models Heart disease prediction using supervised learning algorithms 82 91.2 91.1 [22, 23] ML classification methods Summarized classification techniques for heart disease prediction 88.2 94.5 91.3 [24, 25] ML & soft computing Overview of ML and soft computing approaches for heart disease prediction 92.6 94.8 91.9 [26, 27] ML with multiple data modalities Summarized heart disease prediction with ML using multiple data sources 90.5 90.3 92.6 [28, 29] AI in cardiovascular CT Reviewed AIbased approaches in cardiovascular CT and future implications 92.5 89.2 87.5 [30, 31] Soft computing for heart disease Studied use of soft computing in heart disease prediction and diagnosis 88.1 83.4 91.9 [32] ML and DL algorithms Analyzed different heart diseases using both ML and DL methods 86.4 90.3 85.6 [33] 270 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about AI for inherited heart diseases Reviewed applications of AI models in inherited HD 93.2 92.8 96.1 [34] ML on ECG signals Focused on ML techniques for heart disease diagnosis from ECG data 94.2 84.3 86.3 [35, 36] Figure 1. Generalize Heart disease prediction Approaches [37] As shown in Table 1, predicting heart disease has a long history; however, most existing studies employ traditional machine learning (ML) models. However, some use both machine learning (ML) and deep learning models (DL). Each method has its own limitations, 271 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about advantages, and disadvantages. To better understand the research landscape, prior studies on heart disease prediction Methods were reviewed. Table 2 provides a comprehensive review of these studies [38, 40]. Table 2. Relevant Machine Learning Models on heart disease prediction Discription Key Contributions Model Accuracy Ref Soft computing for heart disease Studied use of soft computing in heart disease prediction and diagnosis NB, DT, DF, and KNN classifiers KNN:90.7, DT: 80.2, RF: 84.2, NB: 88.15 [41] ML and DL algorithms Analyzed different heart diseases using both ML and DL methods SVM, NB LR, DNN, DT, RF, and K-NN. SVM:97.41, NB: 91.38, LR: 96.29, DNN: 98.15, DT: 96.42, RF: 90.46, KNN: 96.42 [42, 43] AI for inherited heart diseases Reviewed applications of AI models in inherited HD KACGANbased model 98.15, DT: [44, 45] ML on ECG signals Focused on ML techniques for heart disease diagnosis from ECG data RF 96.42, RF: [46, 47] ML with multiple data modalities Summarized heart disease prediction with ML using multiple data sources KNN 90.46, KNN: [48, 49] AI in cardiovascular CT Reviewed AI-based approaches in cardiovascular CT and future implications CNN classifiers 89.2 [50] Analyze relevant research on predicting heart disease using machine learning (ML) and deep learning (DL) models to execute this research. Studies that applied machine learning (ML) or deep learning (DL) were excluded from the analysis. From the selected papers, we extracted details such as the year of publication, prediction techniques (ML, DL, or hybrid 272 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about approaches), datasets used (public or self-created), and reported contributions. Studies were categorized into three groups: (i) classical machine learning (ML), (ii) deep learning (DL), and (iii) integrated approaches. Eventually, we recapitulated our findings in Tables, which provide a detailed overview of the surveyed and analyzed methods. This table ensures that the review is wide, organized, and equitable. We applied numerous search methods to collect all appropriate literature in this review. We concentrated on keywords and subject headings related to heart disease prediction models and deep learning, using the method from earlier studies and searching only English sources. To be thorough and avoid overlap, we included both forward and backward searches. We also relied on a keyword-based strategy to keep the search both broad and targeted. Literature Review This subsection summarizes the research on existing work on using deep learning (DL) technology for heart disease. Table 6 comprehensively introduces the existing work of authors on deep learning for heart disease. In the field of heart disease prediction, researchers have deployed a huge array of methods with great achievements [51, 52]. This started the process of convolutional neural networks (CNNs) in categorizing individuals as fit or unfit within a Cleveland dataset [53, 54]. Eq (6) Their model attained the impressive test accuracy of 96% and training accuracy of 97%. Especially, they integrated the clinical parameters to identify the patients' risk factors, enabling early risk prediction. In addition, they expanded on the benefits of a balanced dataset to overcome the limitations posed by traditional machine learning (ML) [55, 56]. Utilized the power of a convolutional neural network (CNN) to deal with early-stage heart disease. Their study clearly defined the dominance of a convolutional neural network (CNN) over classical methods using the Cleveland dataset and achieved the admirable accuracy of 94.78% [57, 58]. Their model passes the pre-processing and feature extraction, and also in prediction, highlighting its commendable capabilities [59, 60]. √ Eq (7) Eq (8) 273 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Table 3 provides the detailed information about an overview of deep learning (DL) approaches for heart disease prediction. Table 3. Comparative Overview of deep learning (DL) Techniques for Heart Disease Prediction Model Dataset Model Precission Accuracy Ref Ensemble DL + Feature Fusion Clinical + sensor data 98.5% Combining wearable sensor and clinical data improved prediction [60, 61] Deep Neural Network (DNN) Heart disease datasets 93.33% Deeper network outperformed ANN and simpler models [62, 63] Multi-task Deep & Wide NN Heart failure data Superior forecasting (no % given) Multi-tasking captured shared features Effectively [64, 65] CNN Cleveland dataset 94.78% CNN achieved strong predictive performance vs. traditional ML Models [66, 67] CNN Cleveland dataset 96% Classified ―fit‖ vs ―unfit‖ patients with high accuracy [68, 69] LSTMDBN Four ECG datasets 88.42% Used time-frequency ECG signals; broader dataset coverage [70, 71] CNN Cleveland dataset 94.78% CNN achieved strong predictive performance vs. traditional ML Models [72, 73] CNN Cleveland dataset 96% Classified ―fit‖ vs ―unfit‖ patients with high accuracy [74] LSTMDBN Four ECG datasets 88.42% Used time-frequency ECG signals; broader dataset coverage [75, 76] 280 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about (ANNs). Employed open-access datasets, which were partitioned into training and testing sets. Their analysis revealed that the Cleveland dataset is one of the most commonly employed datasets in heart disease research. such as those by Sujatha and Mahalakshmi. Random Forest (RF) worked best with 95.60% accuracy [130]. A deep neural network (DNN) model using Talos fine-tuning and trained it on real patient records. Talos performed better than other optimizers and achieved 90.78% accuracy, proving it can help make heart disease predictions more dependable. The studies used various datasets and were carefully done to improve prediction models for future research [131]. Chicco and his team built a complete system that joined distinct methods, like Fuzzy Logic (FL) with models like Decision Trees (DT), Support Vector Machines (SVM), and Artificial Neural Networks (ANN), being tested, and Adaboost. The Accomplishment of this system came from using feature reduction and feature selection methods, like LASSO and MRMR [40]. After that, Zeleznik and his team created a model known as the Hybrid Random Forest with Linear Model (HRFLM) approach. This method used a feature selection technique along with a Random Forest (RF) model to enhance the model's ability to predict outcomes. The model worked well at finding the most important features for predicting heart disease and had an accuracy of 88.7% [132]. Neural Networks with Decision Trees (DT) to develop a special hybrid system. This system performed better than older methods and showed higher accuracy in classifying heart disease. All these studies represent major progress in heart disease prediction and help in building systems that are both accurate and efficient. According to Straw and Wu, common supervised learning techniques—including Random Forest (RF), Decision Tree (DT), and ensemble approaches—are essential in predictive modelling and are very useful in analyzing medical data. These algorithms play an important role in diagnosing different types of heart diseases [133]. Introduced a new method that combined multi-layer neural networks employing a hierarchical, component-based learning framework. This enhanced heart disease prediction because it understood the complex relationships between different risk factors more effectively. It also gave better results than traditional methods. Furthermore, Das and his colleagues used important risk-related features that were derived using time-series analysis of patient data. Then they applied a rough set technique to study the relationships between those features. Their research showed that this method can predict heart disease effectively and plays an important role in healthcare analytics [134]. The analysis found that the most commonly used dataset is the Cleveland dataset, which comes from the UCI Repository. Even though this dataset has 76 features, most of the models used only 14 of them, such as age, gender, chest pain type, blood pressure, ECG results, maximum heart rate, blood sugar, and the number of major blood vessels. In terms of popularity, the Statlog (Heart) dataset came in second, and the Hungarian dataset was third [135]. The developers of heart disease prediction have used various languages. Python is one of the preferred programming languages, with 75% of the researchers describing it as a favorite programming language, as described in figure 8. A number of them are probably MATLAB, 9%, R, 8%, Java, 7%, and Knowledge representation language SWRL (Semantic Web Rule Language), 1%. each is also significant in the case of device +DL and IoT +DL. All other strategies are less frequently utilized: 6%, 6%, 4%. The paper provides a succinct overview of the Deep Learning (DL), Enhanced 281 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Transfer Deep Learning (ETDL), and Integrated DL approaches. To improve prediction results, bigger and more varied datasets are needed. Algorithms need to be validated by integrating multiple risk elements within very large cohorts [136, 137]. API or cloud-based datasets can be hosted for research purposes, since cloud computing efficiently handles large volumes of patient records IoT devices enable the real-time recording of key clinical indicators. Collaboration with physicians is equally important to collect meaningful data for model improvements. Enhancing predictive accuracy by subjecting models to multi-facility datasets is most likely to be more effective [42]. Nevertheless, validation remains a problem, though lab test results are some of the most valuable for evaluating prediction accuracy [138,139]. More general medical records may enhance the accuracy of some prediction models, for instance, for cardiac CT scans. Finally, researchers suggest using real-world datasets instead of just theoretical ones for simulations. Models more efficient when working with datasets that have a lot of missing information, their feature selection methods should be changed, this will enhance the results of the model as a whole. Pairwise classification with extra features has also been proposed for increased accuracy. In order to improve the efficiency of models that take datasets with substantial missing values, their feature selection strategies should be altered. It has been proposed in studies [140, 141]. Implementing ensemble classifiers with extra features provides better models for predicting the staging and severity of a given condition, thus enhancing the results of the model as a whole. Pairwise classification with extra features has also been proposed for increased accuracy. Managing a big number of features in combination with medical data records is a task that is known to be difficult, and thus, a dedicated technique for feature reduction is needed. It is also crucial to develop a better technique that features feature elimination, missing data imputation, noise handling, and more, to boost the accuracy of a model’s predictions [142, 143]. Method and Materials The primary objective of this study is to develop a computerized framework for predicting the probability of heart disease, thereby providing valuable decision-support tools for healthcare professionals and improving patient outcomes. To achieve this goal, a range of machine learning algorithms was applied to a structured dataset, and the corresponding results are comprehensively analyzed in this report. The proposed methodology is further enhanced through rigorous data preprocessing, encompassing data cleaning, elimination of non-contributory variables, and incorporation of additional clinical parameters such as Mean Arterial Pressure (MAP) and Body Mass Index (BMI). Following preprocessing, the dataset is partitioned based on gender to enable more nuanced analysis, and k-modes clustering is employed to uncover latent patterns within subpopulations. The refined dataset is then utilized to train predictive models capable of producing robust and reliable diagnostic outcomes. This improved methodological framework is expected to substantially enhance predictive precision and overall model efficacy, as demonstrated in the subsequent sections of this study. 282 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Proposed Enhanced Heart disease prediction Model This study seeks to develop a robust and comprehensive framework for precise heart disease prediction through the integration of advanced machine learning paradigms, feature selection mechanisms, and dimensionality reduction techniques. By leveraging sophisticated algorithms, the proposed model effectively uncovers and interprets complex, nonlinear relationships embedded within clinical data. The incorporation of ensemble deep learning architectures and innovative feature fusion strategies enables the system to deliver accurate and timely diagnostic predictions, providing healthcare professionals with an intelligent decision-support tool to improve diagnostic accuracy and patient outcomes. The hybrid predictive framework is structured into three principal phases: data acquisition, preprocessing, and classification—each serving a crucial function in optimizing model performance. During the preprocessing phase, meticulous procedures are implemented to ensure data integrity and enhance computational efficiency. These include the imputation of missing values, accomplished through the ML-HDPM approach, which accurately estimates incomplete data entries. In addition, an extensive feature selection process is employed to identify the most discriminative attributes relevant to heart disease prediction. This process leverages a hybrid optimization strategy that synergistically combines the Genetic Algorithm (GA) and Recursive Feature Elimination Method (RFEM), thereby enabling the extraction of salient features that significantly contribute to predictive precision and model robustness. Figure 4. Proposed Framework for Heart Disease Prediction Approaches 283 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about The experimental framework of this study was systematically designed to evaluate model performance under varying conditions through three distinct testing scenarios: (1) utilizing the complete dataset without applying any data reduction techniques, (2) employing a reduced dataset derived from a representative subset of the original data, and (3) implementing a modified bee algorithm on the reduced dataset. Evaluation of the Enhanced Heart Disease Prediction Model The modified bee algorithm, an enhanced adaptation of the conventional bee optimization method, was specifically engineered to improve performance by iteratively exploring and fine-tuning model parameters. All experiments were conducted on a computational setup equipped with an Intel i5 processor and 3 GB of RAM, using MATLAB version 9.2 as the primary development environment. In each scenario, the dataset was systematically partitioned for training and testing in accordance with the experimental design. To ensure methodological rigor and mitigate overfitting, k-fold cross-validation was employed, wherein the dataset was divided into k mutually exclusive subsets. Each subset was sequentially used as a validation set while the remaining subsets were utilized for training. This process was repeated k times, and the mean performance metrics were computed to provide a robust and unbiased evaluation of the model’s predictive capability across diverse conditions. ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑ 284 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Figure 5. Accuracy analysis of predicting heart disease Figure 6. Precision analysis of predicting heart disease. 285 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Figure 7. False positive rate analysis of predicting heart disease Figure 8. True positive rate analysis of predicting heart disease The ML-HDPM approach exhibits superior performance in terms of F-score, achieving 91.5% during training and 89.6% during testing, surpassing all comparative algorithms as shown in Figure 10. This outstanding performance is attributed to the comprehensive design of the model, which effectively integrates feature selection, data balancing, and deep learning optimization techniques. By capturing significant patterns within heart disease datasets, ML-HDPM attains an optimal balance between precision and recall, resulting in consistently higher F-scores. These results emphasize the model’s strong predictive 286 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about capability and its potential to enhance diagnostic accuracy and improve overall patient care outcomes. Figure 9. F score analysis of predicting heart disease. The simulation results underscore the robustness and predictive efficacy of the proposed Machine Learning Hybrid Deep Predictive Model (ML-HDPM) in forecasting heart disease outcomes. Achieving training and testing accuracies of 95.5% and 89.1%, respectively, the ML-HDPM demonstrates superior performance relative to competing algorithms. This enhanced capability is further validated by consistent accuracy measures of 94.8% (training) and 88.3% (testing), as illustrated in Figure 11, confirming the model’s generalizability across diverse datasets. Furthermore, the ML-HDPM achieves a marked reduction in false positive rates (FPR)—8.2% during training and 14.7% during testing—indicating its proficiency in minimizing misclassification errors and enhancing overall predictive precision. In parallel, the model attains notably higher true positive rates (TPR) of 96.2% (training) and 90.8% (testing), reflecting its capacity to correctly identify heart disease cases with high reliability. Collectively, these results affirm ML-HDPM’s substantial potential as a robust predictive tool for accurate and reliable heart disease diagnosis. 287 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Figure 10. ROC–area under curve of (a) MLP, proposed Machine Learning Hybrid Deep Predictive Model (b) RF, (c) DT, and (d) XGB TABLE 5: Comparative Analysis Proposed Model Using Multiple Classifiers based on UCI ML Dataset Clas sifier Model Comput ation Time JC Dice Score Sensitiv ity Acc urac y Speci ficity Preci sion FSco re NBB 3-Layer CNN K = 30 0.7 44 2 0.7442 0.6486 0.64 15 0.64 37 0.74 42 0.64 83 ECNN U-Net K = 30 0.6 43 3 0.6433 0.6486 0.52 45 0.64 85 0.64 33 0.64 85 SV M VGG19 K = 30 0.1 43 0.8444 0.6485 0.74 42 0.74 42 0.74 42 0.64 86 288 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about 2 RNN Inceptio nV3 K = 30 0.2 12 0.5245 0.6485 0.64 33 0.14 32 0.64 33 0.14 32 Prop osed Mod el Efficien tNetB4 K = 30 0.6 22 2 0.7627 0.6486 0.74 84 0.21 2 0.64 33 0.21 2 Figure 10. Sensitivity vs specificity 289 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Figure 11. Sensitivity vs specificity for HHD and ARVC Conclusion This article evaluates a Deep Learning (DL), Enhanced Transfer Deep Learning (ETDL), and hybrid Deep Learning approaches for forecasting heart disease prediction. The meticulous analysis shows that CNN is the leading DL-based strategy, Hybrid deep learning (DL)-based Enhanced Transfer Deep Learning Enhanced Transfer Deep Learning is the most influential Enhanced Transfer Deep Learning -based technique and the most utilized integrated paradigm in the context of DL. In recent years, there has been a growing interest in research that combines DL and other cooperative techniques. The model was trained and evaluated using the UCI Machine Learning Heart Disease dataset, and its performance was benchmarked against state-of-the-art algorithms such as Artificial Neural Networks (ANNs) and Recurrent Networks (RNs). Empirical evaluations reveal that CardioHelp achieves outstanding predictive performance, with an accuracy of 97%, an F1-score of 93.4%, and a precision improvement of 0.8% compared to existing baseline models. Furthermore, when compared with advanced hybrid architectures—including CNN-RNN, CNN-LSTM, and CNN-BLSTM—the proposed framework exhibits accuracy enhancements of 2.35%, 1.34%, and 1.95%, Predicting heart disease accurately and quickly is important but difficult. Machine learning becomes highly effective in identifying and categorizing patients. Advanced machine learning and deep learning models, like SVMs and CNNs, can forecast patients' medical histories and general well-being. It has also been found that the Python language is the most preferred one among all languages for implementing these technologies because it provides extensive libraries as well as community support. Also, the fact that most of the influential studies come from journals proves how academic this field is; mostly from giant publishers like IEEE, Springer, and Elsevier. Though the domain has moved a long way, enormous challenges still exist; importantly, big and diverse datasets are nowhere to be found. This inadequacy puts a very high limitation on how much DLbased approaches can assist in improving accuracy, robustness, and overall reliability when it comes to heart disease predictions. FUNDING STATEMENT: The authors received no specific funding for this study. CONFLICTS OF INTEREST: The authors declare that they have no conflicts of interest to report regarding the present study. REFERENCES National Heart, Lung, and Blood Institute, Medical Expenditure Panel Survey (MEPS), unpublished tabulation (2019–2020) in AHA Statistical Summary 2025, Tables 28-1 and 28-2. Centers for Disease Control and Prevention, ―Heart Disease Facts,‖ U.S. Department of Health & Human Services, Oct. 24, 2024. [Online]. Available: https://www.cdc.gov/heart-disease/data-research/facts-stats/index.html The Global Statistics, ―United States Heart Disease Statistics 2025 | Mortality Stats,‖ 2025. [Online]. Available: https://www.theglobalstatistics.com/united-states-heart- 296 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about MACHINE LEARNING. Vol.-20, No.-3, March (2025) pp 54 - 72 Yousaf, M., Khalid, F., Saleem, M. U., Din, M. U., Shahid, A. K., & Khan, H. (2025). A Deep Learning-Based Enhanced Sentiment Classification and Consistency Analysis of Queries and Results in Search Using Oracle Hybrid Feature Extraction. Spectrum of Engineering Sciences, 3(3), 99-121. Khan, H., Usman, R., Ahmed, B., Hashimi, U., Najam, Z., & Ahmad, S. (2019). Thermalaware real-time task schedulabilty test for energy and power system optimization using homogeneous cache hierarchy of multi-core systems. Journal of Mechanics of Continua and Mathematical Sciences, 14(4), 442-452. Ali, M., Cheema, S. M., Ayub, N., Naz, A., & Aslam, Z. (2022, December). Impact of adopting robots as teachers: a review study. In 2022 International Conference on Emerging Technologies in Electronics, Computing and Communication (ICETECC) (pp. 1-9). IEEE. Naveed, A., Khan, H., Imtiaz, Z., Hassan, W., & Fareed, U. (2024). Application and Ethical Aspects of Machine Learning Techniques in Networking: A Review. Spectrum of engineering sciences, 2(3), 455-501. Ayub, N., Bakhet, S., Arshad, M. J., Saleem, M. U., Anam, R., & Fuzail, M. Z. (2025). AN ENHANCED MACHINE LEARNING AND BLOCKCHAIN-BASED FRAMEWORK FOR SECURE AND DECENTRALIZED ARTIFICIAL INTELLIGENCE APPLICATIONS IN 6G NETWORKS USING ARTIFICIAL NEURAL NETWORKS (ANNS). Spectrum of Engineering Sciences, 3(4), 348364. Ghafoor, U., Ayub, N., Yaseen, A., Anas, M., Farooq, I., Khan, S., & Naghman, N. F. (2025). AI Assisted Heart Disease Prediction and Classification and Segmentation based on PIMA and UCI Machine Learning Datasets. Annual Methodological Archive Research Review, 3(7), 248-276. Sarwar, H. Khan, I. Uddin, R. Waleed, S. Tariq, "An Efficient E-Commerce Web Platform Based on Deep Integration of MEAN Stack Technologies", Bulletin of Business and Economics (BBE)., vol. 12, no. 4, pp. 447-453, Jun. 2023 Ali, M., Cheema, S. M., Ayub, N., Naz, A., & Aslam, Z. (2022, December). Blockchainbased Privacy Preservation Framework for IoT-Based Information Systems. In 2022 3rd International Conference on Innovations in Computer Science & Software Engineering (ICONICS) (pp. 1-7). IEEE, 2022 Asghar, M. A., Aslam, A., Bakhet, S., Saleem, M. U., Ahmad, M., Gohar, A., & Khan, H. (2025). An Efficient Integration of Artificial Intelligence-based Mobile Robots in Critical Frames for the Internet of Medical Things (IoMTs) Using (ADP2S) and Convolutional Neural Networks (CNNs). Annual Methodological Archive Research Review, 3(4), 160-183. Ali, M., Cheema, S. M., Aslam, Z., Naz, A., & Ayub, N. (2023, March). CBAI: Cloud-Based Agile Infrastructure for Enhancing Distributed Agile Development. In 2023 4th International Conference on Computing, Mathematics and Engineering Technologies (iCoMET) (pp. 1-6). IEEE. Ayub, N., Yaseen, A., Amin, M. N., Rizwan, S. M., Farooq, I., & Hussain, M. Z. (2025). 297 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Reliable Federated Learning (Rdl) Assisted Intrusion Detection And Classifications Approach Using (Ssl/Tls) For Network Security. Annual Methodological Archive Research Review, 3(7), 376-400. Jabeen, T., Mehmood, Y., Khan, H., Nasim, M. F., & Naqvi, S. A. A. (2025). Identity Theft and Data Breaches How Stolen Data Circulates on the Dark Web: A Systematic Approach. Spectrum of engineering sciences, 3(1), 143-161. Khan, A. K., Bakhet, S., Javed, A., Rizwan, S. M., & Khan, H. (2025). Framework for Predicting Customer Sentiment Aware Queries and Results in Search Using Oracle and Machine Learning. Spectrum of Engineering Sciences, 3(2), 588-617. Abdullah, M. M., Khan, H., Farhan, M., & Khadim, F. (2024). An Advance Machine Learning (ML) Approaches for Anomaly Detection based on Network Traffic. Spectrum of engineering sciences, 2(3), 502-527. Hashmi, U., & ZeeshanNajam, S. A. (2023). Thermal-Aware Real-Time Task Schedulabilty test for Energy and Power System Optimization using Homogeneous Cache Hierarchy of Multi-core Systems. Journal of Mechanics of Continua and Mathematical Sciences, 14(4), 442-452. Sultan, H., Rahman, S. U., Munir, F., Ali, A., Younas, S., & Khan, H. (2025). Institutional dynamics, innovation, and environmental outcomes: a panel NARDL analysis of BRICS nations. Environment, Development and Sustainability, 1-43. Hussain, M., Ahmed, H. A., Babar, M. Z., Ali, A., Shahzad, H. M., Rehman, S. U., ... & Alshahrani, A. M. (2025). An Enhanced Convolutional Neural Network (CNN) based P-EDR Mechanism for Diagnosis of Diabetic Retinopathy (DR) using Machine Learning. Engineering, Technology and Applied Science Research, 15(1), 19062-19067. Ramzan, M. S., Nasim, F., Ahmed, H. N., Farooq, U., Nawaz, M. S., Bukhari, S. K. H., & Khan, H. (2025). An Innovative Machine Learning based end-to-end Data Security Framework in Emerging Cloud Computing Databases and Integrated Paradigms: Analysis on Taxonomy, challenges, and Opportunities. Spectrum of engineering sciences, 3(2), 90-125. Mujtaba, A., Zulfiqar, M., Azhar, M. U., Ali, S., Ali, A., & Khan, H. (2025). ML-based Fileless Malware Threats Analysis for the Detection of Cyber security Attack based on Memory Forensics: A Survey. The Asian Bulletin of Big Data Management, 5(1), 1-14. Hussain, S., Sarwar, N., Ali, A., Khan, H., Din, I., Alqahtani, A. M., ... & Ali, A. (2025). An Enhanced Random Forest (ERF)-based Machine Learning Framework for Resampling, Prediction, and Classification of Mobile Applications using Textual Features. Engineering, Technology & Applied Science Research, 15(1), 1977619781. Ahmad, I., Nasim, F., Khawaja, M. F., Naqvi, S. A. A., & Khan, H. (2025). Enhancing IoT Security and Services based on Generative Artificial Intelligence Techniques: A Systematic Analysis based on Emerging Threats, Challenges and future Directions. Spectrum of engineering sciences, 3(2), 1-25. Khan, H., Imtiaz, M. A., Siddique, H., Rana, M. T. A., Ali, A., Baig, M. Z., ... & Alsaawy, 298 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Y. (2025). An Enhanced Task Migration Technique Based on Convolutional Neural Network in Machine Learning Framework. Ahmed, A., Javed, M. A., Qureshi, J. N., Khan, H., & Yousaf, H. F. (2024). An insightful Machine Learning based Privacy-Preserving Technique for Federated Learning. The Asian Bulletin of Big Data Management, 4(4), 332-343. Farooq, I., Ahmed, S. A., Ali, A., Warraich, M. A., Aqeel, M., & Khan, H. (2024). Enhanced Classification of Networks Encrypted Traffic: A Conceptual Analysis of Security Assessments, Implementation, Trends and Future Directions. The Asian Bulletin of Big Data Management, 4(4), 500-522. Fawy, K. F., Rodriguez-Ortiz, G., Ali, A., Jadeja, Y., Khan, H., Pathak, P. K., ... & Rahman, J. U. (2025). Catalytic exploration metallic and nonmetallic nano-catalysts, properties, role in photoelectrochemistry for sustainable applications. Reviews in Inorganic Chemistry, (0). Liaqat, M. S., Sharif, N., Ali, A., Khan, H., Ahmed, H. N., & Khan, H. (2024). An Optimal Analysis of Cloud-based Secure Web Applications: A Systematic Exploration based on Emerging Threats, Pitfalls and Countermeasures. Spectrum of engineering sciences, 2(5), 427-457. Adil, M. U., Ali, S., Haider, A., Javed, M. A., & Khan, H. (2024). An Enhanced Analysis of Social Engineering in Cyber Security Research Challenges, Countermeasures: A Survey. The Asian Bulletin of Big Data Management, 4(4), 321-331. Maqsood, M., Dar, M. M., Javed, M. A., & Khan, H. (2024). A Survey on the Internet of Medical Things (IOMT) Privacy and Security: Challenges Solutions and Future from a New Perspective. The Asian Bulletin of Big Data Management, 4(4), 355368. Khawar, M. W., Salman, W., Shaheen, S., Shakil, A., Iftikhar, F., & Faisal, K. M. I. (2024). Investigating the most effective AI/ML-based strategies for predictive network maintenance to minimize downtime and enhance service reliability. Spectrum of Engineering Sciences, 2(4), 115-132. Ahmad, J., Salman, W., Amin, M., Ali, Z., & Shokat, S. (2024). A Survey on Enhanced Approaches for Cyber Security Challenges Based on Deep Fake Technology in Computing Networks. Spectrum of Engineering Sciences, 2(4), 133-149. Ayub, N., Ejaz, A., Hassan, B., Hussain, M. Z., Nadeem, M., Sabir, L., & Fatima, S. (2025). An Efficient Machine Learning And Deep Learning Based Deep Packet Security Framework For Detection Of Computing Network Faults In The Iots. Spectrum of Engineering Sciences, 3(5), 659-674. Ayub, N., Imtiaz, M. A., Ali, E., Alqahtani, A. M., Ali, A., Ashurov, M., ... & Law, F. L. (2025). A Decision Framework for Intra Task Fixed Priority INTEL PXA270 Distributed Architecture for Soft RT-Applications Based on Deep Learning. Engineering, Technology & Applied Science Research, 15(3), 2355323558. Ayub, N., Waheed, A., Ahmad, S., Akbar, M. H. A., Fuzail, M. Z., & Hashmi, A. H. (2025). Strengthening Network Security: An Efficient DL Enabled Data Protection and Privacy Framework for Threat Mitigation and Vulnerabilities Detection in IoT 299 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Network. Annual Methodological Archive Research Review, 3(6), 1-25. Farooq, M., Younas, R. M. F., Qureshi, J. N., Haider, A., & Nasim, F. (2025). Cyber security risks in DBMS: Strategies to mitigate data security threats: A systematic review. Spectrum of engineering sciences, 3(1), 268-290. Ayub, N., Habib, Z., Bakhet, S., Riaz, S., Rizwan, S. M., Abid, M., ... & Khan, H. (2025). An Optimal Ai & Deep Learning Mechanism For Mitigating Hacking Threat Identification Using Secure Network Infrastructure Based On Linux And SoftwareDefined Network (Sdn). Spectrum of Engineering Sciences, 3(5), 675-687. Aslam, I., Tariq, W., Nasim, F., Khan, H., Khawaja, M. F., Ahmad, A., & Nawaz, M. S. (2025). A Robust Hybrid Machine Learning based Implications and Preventions of Social Media Blackmailing and Cyber bullying: A Systematic Approach. Ayub, N., Anwer, M. A., Iqbal, A., Rizwan, S. M., Shahbaz, A., Abid, M. H., & Rafi, S. (2025). Enhanced ML Framework based on Artificial Neural Network for countermeasures of Data Protection and Network Vulnerabilities Detection in Industrial Internet of Things. Annual Methodological Archive Research Review, 3(5), 410-431. Akhtar, M. H., Ali, A., Ali, S., Nasim, F., Aziz, M. H., Khan, H., & Naqvi, S. A. A. (2025). A Novel Machine Learning Approach for Database Exploitation to Enhance Database Security: A Survey. Spectrum of Engineering Sciences, 3(2), 26-57. Jabeen, T., Mehmood, Y., Khan, H., Nasim, M.F. and Naqvi, S.A.A., 2025. Identity Theft and Data Breaches How Stolen Data Circulates on the Dark Web: A Systematic Approach. Spectrum of engineering sciences, 3(1), pp.143-161. A. S. Oliver, K. Ganesan, S. A. Yuvaraj, T. Jayasankar, M. Y. Sikkandar, and N. B. Prakash, ―Accurate prediction of heart disease based on biosystem using regressive learningbased neural network classifier,‖ Journal of Ambient Intelligence and Humanized Computing, vol. 13, no. S1, pp. 1–9, 2021. S. Dami and M. Yahaghizadeh, ―Predicting cardiovascular events with deep learning approach in the context of the Internet of Things,‖ Neural Computing and Applications, vol. 33, no. 13, pp. 7979–7996, 2021. F. Ali, S. El-Sappagh, I. S. Riazul, D. Kwak, A. Ali, M. Imran, and K. S. Kwak, ―A smart healthcare monitoring system for heart disease prediction based on ensemble deep learning and feature fusion,‖ Information Fusion, vol. 63, pp. 208–222, 2020. S. S. Sarmah, ―An efficient IoT-based patient monitoring and heart disease prediction system using deep learning modified neural network,‖ IEEE Access, vol. 8, pp. 135784– 135797, 2020. B. Wang, Y. Bai, Z. Yao, J. Li, W. Dong, Y. Tu, W. Xue, Y. Tian, Y. Wang, and K. He, ―A multi-task neural network architecture for renal dysfunction prediction in heart failure patients with electronic health records,‖ IEEE Access, vol. 7, pp. 178392– 178400, 2019. M. Khan, ―An IoT framework for heart disease prediction based on MDCNN classifier,‖ IEEE Access, vol. 8, pp. 34717–34727, 2020. J. Shafi, M. S. Obaidat, P. V. Krishna, B. Sadou, M. Punambal, and J. Gitanjali, ―Prediction of heart abnormalities using deep learning model and wearable devices in smart 300 http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about health homes,‖ Multimedia Tools and Applications, vol. 81, pp. 543–557, 2022. S. Sharma and M. Parmar, ―Heart diseases prediction using deep learning neural network model,‖ International Journal of Innovative Technology and Exploring Engineering, vol. 9, no. 3, pp. 2244–2248, 2020. S. Nandy, M. Adhikari, V. Balasubramanian, V. Menon, X. Li, and M. Zakarya, ―An intelligent heart disease prediction system based on swarm-artificial neural network,‖ Neural Computing and Applications, vol. 35, pp. 14723–14737, 2023. P. Dhaka and B. Nagpal, ―WoM-based deep BiLSTM: Smart disease prediction model using WoM-based deep BiLSTM classifier,‖ Multimedia Tools and Applications, vol. 82, pp. 25061–25082, 2023. N. Sharma, L. Malviya, A. Jadhav, and P. Lalwani, ―A hybrid deep neural net learning model for predicting coronary heart disease using randomized search crossvalidation optimization,‖ Decision Analytics Journal, vol. 9, p. 100331, 2023. C. Zhou, A. Hou, P. Dai, A. Li, Z. Zhang, Y. Mu, and L. Liu, ―Risk factor refinement and ensemble deep learning methods on prediction of heart failure using real healthcare records,‖ Information Sciences, vol. 637, p. 118932, 2023. A. A. Almazroi, ―Survival prediction among heart patients using machine learning techniques,‖ Mathematical Biosciences and Engineering, vol. 19, no. 1, pp. 134–145, 2022. G. Gupta, U. Adarsh, R. N. Subba, and A. B. Rao, ―Comparison of various machine learning approaches used in heart ailments prediction,‖ Journal of Physics: Conference Series, vol. 2161, no. 1, p. 012010, 2022. R. Bharti, A. Khamparia, M. Shabaz, G. Dhiman, S. Pande, and P. Singh, ―Prediction of heart disease using a combination of machine learning and deep learning,‖ Computational Intelligence and Neuroscience, vol. 2021, p. 8387680, 2021. P. Li, Y. Hu, and Z. P. Liu, ―Prediction of cardiovascular diseases by integrating multimodal features with machine learning methods,‖ Biomedical Signal Processing and Control, vol. 66, p. 102474, 2021. R. Biswas, A. R. Beeravolu, A. Karim, S. Azam, M. T. Hasan, M. S. Alam, and P. Ghosh, ―A robust deep learning-based prediction system of heart disease using a combination of five datasets,‖ in Proc. 31st Int. Conf. Computer Theory and Applications (ICCTA), 2021, pp. 223–228. D. Chicco and G. Jurman, ―Machine learning can predict survival of patients with heart failure from serum creatinine and ejection fraction alone,‖ BMC Medical Informatics and Decision Making, vol. 20, no. 16, pp. 108–121, 2020. R. B. Zeleznik, P. Eslami, et al., ―Deep convolutional neural network to predict cardiovascular risk from computed tomography,‖ Nature Communications, vol. 12, no. 1, p. 715, 2021. M. A. Hassani, R. Tao, M. Kamyab, and M. H. Mohammadi, ―An approach of predicting heart disease using a hybrid neural network and decision tree,‖ in Proc. 5th Int. Conf. Big Data and Computing, Chengdu, China, 2020, pp. 84–89. http://amresearchreview.com/index.php/Journal/about Volume 3, Issue 10 (2025) Online ISSN Print ISSN . . http://amresearchreview.com/index.php/Journal/about Page 301 I. Straw and H. Wu, ―Investigating bias in healthcare algorithms: A sex-stratified analysis of supervised machine learning models in liver disease prediction,‖ BMJ Health & Care Informatics, vol. 29, no. 1, pp. 142–156, 2022. O. W. Samuel, B. Yang, Y. Geng, M. G. Asogbon, S. Pirbhulal, D. Mzurikwao, O. P. Idowu, T. J. Ogundele, X. Li, S. Chen, et al., ―A new technique for the prediction of heart failure risk driven by hierarchical neighborhood component-based learning and adaptive multi-layer networks,‖ Future Generation Computer Systems, vol. 110, pp. 781–794, 2020. S. Das, S. K. Pradhan, S. Mishra, S. Pradhan, and P. Pattnaik, ―Prediction of heart diseases using soft computing technique,‖ in Intelligent Systems: Proc. ICMIB, Springer, Berlin, 2022, pp. 155–167. S. Singhal, H. Kumar, and V. Passricha, ―Prediction of heart disease using CNN,‖ American International Journal of Research in Science, Technology, Engineering & Mathematics, vol. 23, no. 1, pp. 257–261, 2018. P. Rani, R. Kumar, N. M. Ahmed, and A. Jain, ―A decision support system for heart disease prediction based upon machine learning,‖ Journal of Reliable Intelligent Environments, vol. 7, no. 3, pp. 263–275, 2021. G. S. Sajja, ―A comprehensive review of various machine learning techniques for heart disease prediction,‖ International Journal of Computer Applications, vol. 183, p. 37, 2021. B. Vaishali, V. Madhushi, C. Pradeep, and S. Gaurav, ―Efficient heart disease prediction using hybrid deep learning classification models,‖ IRBM, vol. 44, no. 5, p. 100786, 2023. A. Mehmood, M. Iqbal, Z. Mehmood, A. Irtaza, M. Nawaz, T. Nazir, M. Masood, and M. Masood, ―Prediction of heart disease using deep convolutional neural networks,‖ Arabian Journal for Science and Engineering, vol. 46, no. 4, pp. 3409–3422, 2021. A. Hauptmann, S. Arridge, F. Lucka, V. Muthurangu, and J. A. Steeden, ―Real-time cardiovascular MR with spatio-temporal artifact suppression using deep learning – proof of concept in congenital heart disease,‖ Magnetic Resonance in Medicine, vol. 81, no. 2, pp. 1143–1156, 2019.