Epilepsy detection based on multi-head self-attention mechanism

Yandong Ru; Gaoyang An; Zheng Wei; Hongming Chen

doi:10.1371/journal.pone.0305166

Peer Review History

Original SubmissionFebruary 21, 2024
17 Mar 2024 Decision Letter - Mustafa Sameer, Editor PONE-D-24-07075Epilepsy Detection Based on Multi-Head Self-Attention MechanismPLOS ONE Dear Dr. An, 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 Apr 29 2024 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, Mustafa Sameer, 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, all author-generated code must 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. Thank you for stating the following financial disclosure: "The author(s) received no specific funding for this work." At this time, please address the following queries: a) Please clarify the sources of funding (financial or material support) for your study. List the grants or organizations that supported your study, including funding received from your institution. b) State what role the funders took in the study. If the funders had no role in your study, please state: “The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.” c) If any authors received a salary from any of your funders, please state which authors and which funders. d) If you did not receive any funding for this study, please state: “The authors received no specific funding for this work. Please include your amended statements within your cover letter; we will change the online submission form on your behalf. 4. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section. 5. Thank you for stating the following in the Acknowledgments Section of your manuscript: "This work was supported by the Foundation for Talent of Zhejiang Ocean University (JX6311061523)." We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form. Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows: "The author(s) received no specific funding for this work." Please include your amended statements within your cover letter; we will change the online submission form on your behalf. [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: Yes ******** 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: Yes ****** 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: (1) The authors propose a framework that combines CNN and multi-head attention to detect epilepsy. However, combining CNN and multi-head attention is not an innovation, and much work has been done nowadays, so it is not innovative enough. (2) The figures are not clear enough. (3) References are not novel enough, authors should cite more literatures from the last three years. Reviewer #2: The aim and method are concise with details but the authors could include the following comments on 1. How does the number of self-attention heads and their dimension affect the model's ability to learn intricate relationships between different feature vectors? 2. The paper introduces a "light multi-head self-attention" mechanism to reduce computational cost. How significant is this reduction compared to the original Transformer approach? 3. How does the combination of 1x1 and 3x3 depth-wise separable convolutions within the residual network impact the feature representation compared to using only one type of convolution? 4. The choice of hyper parameters for the convolutional layers and the number of self-attention heads is crucial. How sensitive is the model's performance to these choices? 5. How does the choice of loss function (e.g., cross-entropy) impact the model's training process and the final classification performance? 6. Epilepsy events are typically rare compared to normal EEG recordings. How does the model handle this class imbalance issue, and are there any specific techniques used to address it? Reviewer #3: Please find the following comments on your paper: 1. The motivation for research is not clear from the abstract. 2. Paper contributions should be considered in the introduction section. 3. Paper organization is missing in the introduction section. 4. The related work section is missing in the paper. 5. Please improve the figure quality. 6. Please consider a system overview or block diagram of research for a better understanding of the paper. 7. Data distribution graphs should be considered for the analysis. 8. Limitations of the proposed approach should be discussed. 9. In Figs 7 and 8, the y label should be written. ****** 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: Yes: Dr. M. Laavanya Reviewer #3: Yes: Alwin Poulose ******** [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. Attachments Attachment Submitted filename: Comments_PONE-D-24-07075.docx https://doi.org/10.1371/journal.pone.0305166.r001
Revision 1
26 Apr 2024 Author Response Revision Description We are very grateful to the reviewers for their valuable comments, which are of great significance to our research. We have studied the reviewers’ comments and revised the paper carefully according to the comments. Changes to the original text are shown in red in the text, and new additions are shown in blue. The specific revision are as follows: Comments from reviewer 1 1. The authors propose a framework that combines CNN and multi-head attention to detect epilepsy. However, combining CNN and multi-head attention is not an innovation, and much work has been done nowadays, so it is not innovative enough. Revision description： Thank you very much for your careful review and valuable feedback on our paper. Here, we would like to explain and clarify your concerns about the novelty of our work. Firstly, this paper focuses on the research of cross-patient epilepsy detection. While CNN and self-attention mechanisms have been applied in single-patient epilepsy detection, there remains a gap in the field of cross-patient epilepsy detection. Due to the complex differences between different individuals in cross-patient detection, it is more challenging but also holds greater practical value. Therefore, this paper combines CNN with self-attention mechanisms for cross-patient epilepsy detection, contributing to further advance this field. Additionally, this paper introduces an optimized Transformer module. The uniqueness of this module lies in its structure, which integrates CNN layers, light multi-head self-attention layers, and residual feed-forward networks, forming a new architecture distinct from traditional attention mechanism modules. This design significantly improves computational efficiency, greatly reduces training time, and enhances the overall efficiency of the model while maintaining detection performance. 2. The figures are not clear enough. Revision description： Due to figure format issues in the previous paper, the figures were not clear enough. This problem has now been resolved, ensuring that all figures in the paper are clear enough. 3. References are not novel enough, authors should cite more literatures from the last three years. Revision description： To address this issue, we have re-examined the latest research findings in the relevant field during the revision process and selected more representative and cutting-edge literature from the past three years for citation. Please refer to the third to sixth paragraphs in the "Introduction" section and the "Comparison of literature" section in the paper for detailed changes. Comments from reviewer 2 1. How does the number of self-attention heads and their dimension affect the model's ability to learn intricate relationships between different feature vectors? Revision description： Due to experimental requirements, we have integrated the section on the impact of the number of self-attention heads on the model from our previous paper into the "Hyper-parametric analysis" section of the current paper and revised its description. Please refer to the last paragraph of the "Hyper-parametric analysis" section and Table 2 in the paper for detailed changes. 2. The paper introduces a "light multi-head self-attention" mechanism to reduce computational cost. How significant is this reduction compared to the original Transformer approach? Revision description： In the comparative experiments, a new performance evaluation dimension, training time, has been added to highlight the computational efficiency of the proposed "light multi-head self-attention" mechanism. The experimental results are shown in Table 6, which comprehensively demonstrates the advantages of this mechanism in computational efficiency. Please refer to the last two sentences of the second-to-last paragraph in the "Cross-Patient Detection Performance" section and Table 6 in the paper for detailed changes. 3. How does the combination of 1x1 and 3x3 depth-wise separable convolutions within the residual network impact the feature representation compared to using only one type of convolution? Revision description： Added experiments named the "Convolutional structure analysis of RFFN" to the "Experimental parametric" section in this paper. The experimental results are shown in Table 5. Based on Table 5, this paper provides a detailed analysis of the impact of combining 1x1 and 3x3 depth-wise separable convolutions compared to using a single convolution type in residual networks on model performance Please refer to the "Convolutional structure analysis of RFFN" section in the paper for detailed changes. 4. The choice of hyper parameters for the convolutional layers and the number of self-attention heads is crucial. How sensitive is the model's performance to these choices? Revision description： Added experiments named the "Hyper-parametric analysis" to the "Experimental parametric" section in this paper. The experimental results are shown in Table 2, which can be seen that the model performance varies with different values of batch size, learning rate, and number of self-attention heads. Please refer to the "Hyper-parametric analysis" section in the paper for detailed changes. 5. How does the choice of loss function (e.g., cross-entropy) impact the model's training process and the final classification performance? Revision description： Added experiments named the "Loss function analysis" to the "Experimental parametric" section in this paper. The experimental results are shown in Fig 9, which can be seen that the effect of different loss functions on the performance of the model. Please refer to the "Loss function analysis" section in the paper for detailed changes. 6. Epilepsy events are typically rare compared to normal EEG recordings. How does the model handle this class imbalance issue, and are there any specific techniques used to address it? Revision description： To address the issue of data imbalance, we used oversampling techniques in the "Data preprocessing methods" section of our previous paper. In the current paper, we have further added a detailed description of the oversampling technique. Please refer to the last paragraph and the second-to-last paragraph of the "Data preprocessing methods" section in the paper for detailed changes. Comments from reviewer 3 1. The motivation for research is not clear from the abstract. Revision description： The motivation for research in the abstract has been revised to make it clearer in this paper. Please refer to the first two sentences of the "Abstract" section in the paper for detailed changes. 2. Paper contributions should be considered in the introduction section. Revision description： Added the contributions of the paper to the "Introduction" section in this paper. Please refer to the second-to-last paragraph of the "Introduction" section in the paper for detailed changes. 3. Paper organization is missing in the introduction section. Revision description： Added the organization of the paper to the "Introduction" section in this paper. Please refer to the last paragraph of the "Introduction" section in the paper for detailed changes. 4. The related work section is missing in the paper. Revision description： Added the related work to the "Introduction" section in this paper. Please refer to the fourth to sixth paragraphs in the "Introduction" section in the paper for detailed changes. 5. Please improve the figure quality. Revision description： Due to figure format issues in the previous paper, the figures were not clear enough. This problem has now been resolved, ensuring that all figures in the paper are clear enough. 6. Please consider a system overview or block diagram of research for a better understanding of the paper. Revision description： Added the system overview of the research to the "Methods" section in this paper. And the Fig 1 in this paper shows the research block diagram. Please refer to the first paragraph in the "Methods" section in the paper for detailed changes. 7. Data distribution graphs should be considered for the analysis. Revision description： Added time-domain and frequency-domain distribution graphs of epileptic EEG signals and description to the "Data preprocessing methods" section in this paper. The Figs 6 and 7 of this paper show the distribution of epileptic EEG signals in the time and frequency domains respectively. Please refer to the first paragraph and the third paragraph of the "Data preprocessing methods" section in the paper for detailed changes. 8. Limitations of the proposed approach should be discussed. Revision description： Added a description of the limitations of the proposed approach to the "Conclusion" section in this paper. Please refer to the last paragraph of the "Conclusion" section in the paper for detailed changes. 9. In Figs 7 and 8, the y label should be written. Revision description： Due to experimental requirements, Tables 2 and 6 are currently used in this paper to present the information instead of Figs 7 and 8. Attachments Attachment Submitted filename: Response to Reviewers.docx https://doi.org/10.1371/journal.pone.0305166.r002
27 May 2024 Decision Letter - Yuvaraj Rajamanickam, Editor Epilepsy Detection Based on Multi-Head Self-Attention Mechanism PONE-D-24-07075R1 Dear Dr. An, 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. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org. 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, Yuvaraj Rajamanickam, Ph.D Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation. Reviewer #1: All comments have been addressed Reviewer #3: All comments have been addressed ******** 2. 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 #3: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #3: Yes ****** 4. 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 #3: Yes ****** 5. 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 #3: Yes ****** 6. 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: (No Response) Reviewer #3: Dear Authors, Thank you for addressing all my comments and I don't have any further concerns on the paper. Regards ****** 7. 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 #3: No ******** https://doi.org/10.1371/journal.pone.0305166.r003
Formally Accepted
31 May 2024 Acceptance Letter - Yuvaraj Rajamanickam, Editor PONE-D-24-07075R1 PLOS ONE Dear Dr. An, 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 If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks 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. 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. Yuvaraj Rajamanickam Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0305166.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 .