Enhanced machine learning and hybrid ensemble approaches for Coronary Heart Disease prediction

Maurice Wanyonyi; Zakayo Ndiku Morris; Faith Mueni Musyoka; Dominic Makaa Kitavi

doi:10.1371/journal.pone.0328338

Peer Review History

Original SubmissionJune 29, 2025
4 Sep 2025 Decision Letter - Vijayalakshmi Kakulapati, Editor PONE-D-25-35231Enhanced machine learning and hybrid ensemble approaches for coronary heart disease predictionPLOS ONE Dear Dr. Wanyonyi, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Oct 19 2025 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. Please include the following items when submitting your revised manuscript: A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols. We look forward to receiving your revised manuscript. Kind regards, Vijayalakshmi Kakulapati, Ph.D Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf. 2. Please note that PLOS One has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, we expect all author-generated code to be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse. 3. We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 8 in your text; if accepted, production will need this reference to link the reader to the Table 4. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented. Reviewer #1: Yes Reviewer #2: Yes Reviewer #3: Partly ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: No Reviewer #3: Yes ****** 3. Have the authors made all data underlying the findings in their manuscript fully available? The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified. Reviewer #1: Yes Reviewer #2: Yes Reviewer #3: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English? PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here. Reviewer #1: Yes Reviewer #2: Yes Reviewer #3: No ****** 5. Review Comments to the Author Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters) Reviewer #1: The author used an enhanced machine learning and hybrid ensemble approaches for coronary heart disease prediction. This is a very clear, easy to read and understand work. However, my comments are as fellows. 1. The Abstract was apt and improvement made from this work is empirical obvious. 2. The introduction and related work need to be made more robust because there are very good literatures that this papers did not capture. This studies will make this paper capture wider audience for instance the author shold look at these article for instance. a) Obaido G, Ogbuokiri B, Swart TG, Ayawei N, Kasongo SM, Aruleba K, Mienye ID, Aruleba I, Chukwu W, Osaye F, Egbelowo O. F, Simphiwe S, E. Esenogho. “An Interpretable Machine Learning Approach for Hepatitis B Diagnosis” Applied Sciences. 2022; 12(21):11127. https://doi.org/10.3390/app122111127. b) Sarah Alexandria Nabofa-Ebiaredoh, E. Esenogho, Theo G. Swart, et al “A Machine Learning Method with Filter-Based Feature Selection for Improved Detection of Chronic Kidney Disease” Bioengineering 2022, vol. 9, no. 8, 350; https://doi.org/10.3390/bioengineering9080350, Switzerland. c) Sarah Alexandria Nabofa-Ebiaredoh, E. Esenogho., Theo G. Swart “Improved machine learning methods for classification of imbalanced data” https://hdl.handle.net/10210/481973 d) Sarah Alexandria Nabofa-Ebiaredoh, E. Esenogho, T. G. Swart “Integrating Enhanced Sparse Autoencoder Based Artificial Neural Network Technique and SoftMax Regression for Medical Diagnosis Published MDPI Electronics Journal, 2020, 9(11), 1963; https://doi.org/10.3390/electronics9111963. e) Sarah Alexandria Nabofa-Ebiaredoh, E. Esenogho., Theo G. Swart “Artificial Neural Network Technique for Improving Prediction of Credit Card Default: A Stacked Sparse Autoencoder Approach” International Journal of Electrical and Computer Engineering, Vol. 11, No. 5, October 2021, pp. 4392~4402, v11i5.pp4392-4402. and many more Good work great well done Reviewer #2: 1-The dataset was sourced from a U.S.-based survey (BRFSS) and may not have fully captured region-specific risk factors, genetic predispositions, or healthcare access patterns prevalent in LMICs like Kenya, potentially limiting the immediate generalizability of the models. 2-The reliance on self-reported data for CHD diagnosis and risk factors introduced the potential for recall bias and misclassification, as clinical verification was not available. 3-The computational complexity of some advanced ensemble techniques, such as stacking and Bayesian Model Averaging, was high, which could pose implementation challenges in real-world, resource-constrained healthcare environments with limited computing infrastructure. 4-Despite the incorporation of explainability techniques like SHAP, certain elements of the ensemble models remained inherently complex ("black-box"), which could potentially limit clinician trust and adoption without further integration of explainable AI (XAI) frameworks. 5-The study did not formally evaluate scalability metrics such as training time, inference latency, memory usage, or model size, which are critical for assessing deployment feasibility in low-resource settings. Reviewer #3: The paper presents a hybrid ensemble approach for the prediction of coronary heart disease (CHD) using models such as Adaptive Noise–Resistant Decision Trees (ANRDT), Hybrid Imbalanced Random Forest (HIRF), and Pruned Gradient Boosting Machines (PGBM). The authors use SMOTE to address class imbalance and evaluate model performance on several metrics, including AUC, accuracy, and sensitivity. The ensemble models consistently outperform baseline classifiers, and achieve strong generalizability and robust calibration, especially the stacking ensemble model. The use of SMOTE addresses the imbalance between the classes, but the paper does not assess whether the synthetic samples have realistic clinical relationships. The hybrid models, including ANRDT and HIRF, offer performance improvements, but ablation studies that isolate the contribution of each component to overall performance are lacking. While the evaluation focuses on accuracy, AUC, and sensitivity, additional metrics such as fairness or model calibration across demographic groups would provide more depth, especially for healthcare applications. The study compares ensemble methods, but more recent techniques such as AutoML or deep ensemble methods could be considered for further performance benchmarking. Although time complexity is discussed, there is no comparison with state-of-the-art scalable models or real-world applications in LMICs. Interpretability is limited to feature importance; using techniques such as SHAP for individual-level explanations would improve clinical applicability. Consider the following studies: https://doi.org/10.1038/s41598-024-73570-x https://doi.org/10.3390/app13085188 https://doi.org/10.1038/s41598-025-07350-6 ****** 6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy. Reviewer #1: No Reviewer #2: No Reviewer #3: No ******** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step. https://doi.org/10.1371/journal.pone.0328338.r001
Revision 1
12 Sep 2025 Author Response Response to Reviewers Manuscript Title: Enhanced Machine Learning and Hybrid Ensemble Approaches for Coronary Heart Disease Prediction Manuscript ID: PONE-D-25-35231 We sincerely thank the Academic Editor and Reviewers for their careful evaluation of our manuscript and for the constructive comments provided. We have carefully revised the paper in response to the feedback. Below, we provide a detailed, point-by-point response. Reviewer comments are shown in bold, followed by our responses. Reviewer #1 Comment 1: The Abstract was apt and improvement made from this work is empirically obvious. Response: We thank the reviewer for this positive comment and appreciation of the clarity of our abstract and the strength of our empirical contributions. No changes were necessary in this section, but we are encouraged by the reviewer’s recognition. Comment 2: The introduction and related work need to be made more robust because there are very good literatures that this paper did not capture. These studies will make this paper capture wider audience for instance the author should look at these articles for instance. Response: We appreciate this valuable suggestion. We have carefully revised the Introduction and Related Work sections to incorporate the recommended literature and strengthen the contextual foundation of our study. Specifically: • Added discussion of Obaido et al. (2022) demonstrating the value of interpretable ML with SHAP in hepatitis B diagnosis, highlighting interpretability in medical AI. • Integrated Nabofa-Ebiaredoh et al. (2022) emphasizing feature selection with cost-sensitive AdaBoost for chronic kidney disease detection, reinforcing the role of feature selection in robust diagnostic modeling. • Cited Nabofa-Ebiaredoh et al. (2022) to address class imbalance mitigation with sparse autoencoders. • Included Nabofa-Ebiaredoh et al. (2020) showcasing enhanced sparse autoencoders with Softmax regression achieving high predictive accuracy in CKD, cervical cancer, and heart disease, supporting our hybrid framework. • Cited Nabofa-Ebiaredoh et al. (2021) on stacked sparse autoencoders for credit card default prediction, demonstrating versatility of ensemble and feature-learning approaches beyond healthcare. These additions have been strategically placed throughout the Introduction to enrich the background and broaden the relevance of our study. Reviewer #2 Comment 1: The dataset was sourced from a U.S.-based survey (BRFSS) and may not have fully captured region-specific risk factors, genetic predispositions, or healthcare access patterns prevalent in LMICs like Kenya, potentially limiting the immediate generalizability of the models. Response: We agree and have highlighted this more explicitly in the Discussion (Limitations) section. While BRFSS is U.S.-based, its behavioral and clinical variables are relevant in LMICs. The trained models can be validated and fine-tuned using LMIC-specific datasets for improved adaptability. Comment 2: The reliance on self-reported data for CHD diagnosis and risk factors introduced the potential for recall bias and misclassification, as clinical verification was not available. Response: We acknowledge this limitation and now discuss it explicitly in the Discussion (Limitations) section. Self-reported data introduces potential recall bias and misclassification; future research should incorporate clinically validated datasets. Comment 3: The computational complexity of some advanced ensemble techniques, such as stacking and Bayesian Model Averaging, was high, which could pose implementation challenges in real-world, resource-constrained healthcare environments with limited computing infrastructure. Response: We thank the reviewer for highlighting this important point. In response, we have expanded the manuscript to include a dedicated Scalability Analysis of Enhanced Models section (pp. 40–42). In this section, we evaluate computational aspects of the proposed methods, including training time, inference latency, model size, and memory usage. The results are summarized in Table 10. This analysis demonstrates that while some techniques such as stacking and Bayesian Model Averaging indeed have higher computational demands, boosting-based methods (e.g., PGBM, Boosting) provide a more favorable trade-off between predictive accuracy and scalability. By explicitly quantifying resource requirements, we show how the proposed models can be realistically deployed in real-world and low-resource healthcare settings. Comment 4: Despite the incorporation of explainability techniques like SHAP, certain elements of the ensemble models remained inherently complex ("black-box"), which could potentially limit clinician trust and adoption without further integration of explainable AI (XAI) frameworks. Response: A new Results section on Model Interpretability and Explainability now presents SHAP analyses for global and individual feature impacts and a surrogate decision tree that approximates the behavior of the stacking ensemble. Clinically meaningful features (age, hypertension, diabetes) and protective lifestyle factors (physical activity, diet) are highlighted. We acknowledge that black-box elements remain, and future work will incorporate comprehensive XAI frameworks. Comment 5: The study did not formally evaluate scalability metrics such as training time, inference latency, memory usage, or model size, which are critical for assessing deployment feasibility in low-resource settings. Response: We have addressed this concern by formally evaluating scalability metrics, including training time, inference latency, model size, and memory usage, for both baseline and enhanced machine learning models. Furthermore, the Discussion has been updated to emphasize the trade-offs between predictive performance and computational feasibility, ensuring that scalability is explicitly considered alongside accuracy and interpretability. Reviewer #3 The paper presents a hybrid ensemble approach for the prediction of coronary heart disease (CHD) using models such as Adaptive Noise–Resistant Decision Trees (ANRDT), Hybrid Imbalanced Random Forest (HIRF), and Pruned Gradient Boosting Machines (PGBM). The authors use SMOTE to address class imbalance and evaluate model performance on several metrics, including AUC, accuracy, and sensitivity. The ensemble models consistently outperform baseline classifiers, and achieve strong generalizability and robust calibration, especially the stacking ensemble model. Comment 1: The use of SMOTE addresses the imbalance between the classes, but the paper does not assess whether the synthetic samples have realistic clinical relationships Response: We acknowledge the reviewer’s point. While SMOTE is effective for mitigating class imbalance, its synthetic samples may not fully reflect realistic clinical relationships. We have now discussed this limitation explicitly in the Limitations section and noted that future work could explore domain-aware oversampling strategies or generative approaches such as GANs and VAEs to better preserve clinically plausible patterns. Comment 2: The hybrid models, including ANRDT and HIRF, offer performance improvements, but ablation studies that isolate the contribution of each component to overall performance are lacking. Response: Ablation studies are acknowledged as valuable but are beyond the current manuscript’s scope. This is now noted in the Discussion, with plans for future experiments to isolate component contributions in hybrid ensembles. Comment 3: While the evaluation focuses on accuracy, AUC, and sensitivity, additional metrics such as fairness or model calibration across demographic groups would provide more depth, especially for healthcare applications. Response: Calibration curves have been presented for baseline and enhanced models, improving reliability assessment. Fairness metrics are limited by BRFSS dataset constraints; future studies should incorporate subgroup analyses. Comment 4: The study compares ensemble methods, but more recent techniques such as AutoML or deep ensemble methods could be considered for further performance benchmarking. Response: While AutoML and deep ensembles are powerful, they are often computationally intensive. Our focus on tree- and SVM-based methods balances performance and efficiency. Benchmarking against these advanced frameworks is suggested as future work. Comment 5: Although time complexity is discussed, there is no comparison with state-of-the-art scalable models or real-world applications in LMICs. Response: We expanded the Discussion to address LMIC deployment feasibility. Enhanced ensemble methods are suitable for standard hardware; future studies will compare against scalable distributed systems. Comment 6: Interpretability is limited to feature importance; using techniques such as SHAP for individual-level explanations would improve clinical applicability. Response: We have now included SHAP analyses both global feature importance and local directional impact to our study. This strengthens transparency and clinical trust. Related studies are cited to contextualize interpretability benefits. Comment 7: Consider the following studies: https://doi.org/10.1038/s41598-024-73570-x, https://doi.org/10.3390/app13085188, https://doi.org/10.1038/s41598-025-07350-6 Response: All recommended citations have been incorporated, as they align well with our study. Their inclusion broadens the literature review and strengthens the contextual grounding. • Saleem et al. (2025): Imbalance-handling in stroke prediction. • Javeed et al. (2023): Feature selection and ensemble models for cardiac mortality prediction. • Saleem et al. (2024): Hybrid autoencoder-linear classifier models for stroke prediction. Additional Journal/Editor Requirements Comment: Ensure the manuscript meets PLOS ONE style requirements, including file naming and templates. Response: The manuscript has been revised to comply with PLOS ONE style guidelines and file naming conventions, following the templates provided by the journal. Comment: 2. Please note that PLOS One has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, we expect all author-generated code to be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse. Response: We have ensured that all author-generated code is fully available and openly accessible. The repository is public on GitHub (https://github.com/wanyonyi254-0001/CHD-Baseline-Enhanced-ML-Models), archived on Zenodo with DOI (https://doi.org/10.5281/zenodo.17073723) The dataset used in the study is openly available via Zenodo: https://doi.org/10.5281/zenodo.17073818. All steps to run the baseline and enhanced models are fully documented in the README. Comment: We note you have included a table to which you do not refer in the text of your manuscript. Please ensure that you refer to Table 8 in your text; if accepted, production will need this reference to link the reader to the Table Response: Table 8 is now cited appropriately in the manuscript. Comment: If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. Response: We have reviewed the suggested references and incorporated those deemed relevant, as described in our responses to reviewers above. We thank the Academic Editor and Reviewers again for these constructive comments. Their input has improved the clarity, rigor, and real-world applicability of our manuscript. Sincerely, Maurice Wanyonyi (on behalf of all co-authors) 10/09/2025 Attachments Attachment Submitted filename: Response to Reviewers.docx https://doi.org/10.1371/journal.pone.0328338.r002
7 Dec 2025 Decision Letter - Vijayalakshmi Kakulapati, Editor Enhanced Machine Learning and Hybrid Ensemble Approaches for Coronary Heart Disease Prediction PONE-D-25-35231R1 Dear Dr. Wanyonyi, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Vijayalakshmi Kakulapati, Ph.D Academic Editor PLOS One Reviewers' comments: Reviewer's Responses to Questions Reviewer #1: All comments have been addressed Reviewer #3: All comments have been addressed Reviewer #1: The author has taken time to address all the comments I raised. as such I do not have any reservation against the work Reviewer #3: The authors have addressed all previous comments carefully, and the revised manuscript shows clear improvement in clarity, completeness, and technical soundness. The methodology is now better justified, the experimental discussion is more detailed, and the figures and tables have been improved for readability. The added explanations and updated related work strengthen the overall contribution. The revisions resolve the earlier concerns, and I did not find any remaining issues that would hinder publication. The paper now presents a coherent and well-supported study that fits the scope of the journal. I recommend acceptance, pending minor editorial adjustments for grammar and formatting. https://doi.org/10.1371/journal.pone.0328338.r003
Formally Accepted
Acceptance Letter - Vijayalakshmi Kakulapati, Editor PONE-D-25-35231R1 PLOS One Dear Dr. Wanyonyi, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Vijayalakshmi Kakulapati Academic Editor PLOS One https://doi.org/10.1371/journal.pone.0328338.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .