All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Thank you for your contribution to PeerJ Computer Science.
[# PeerJ Staff Note - this decision was reviewed and approved by Paula Soares, a PeerJ Section Editor covering this Section #]
Your manuscript has been improved but still in one experts opinion the following should be done:
Regarding the validity of the findings, a proper discussion section is still missing.
The current discussion section has not been effectively improved. It still provides only a summary of the work. A proper discussion section, as stated in the previous revision round, should provide, above all, an interpretation of the results in view and in comparison with the relevant literature in the field.
Some of the above observations are instead reported in the conclusions section. Therefore, I would suggest limiting the conclusions section to a very brief summary of the work. Some parts of the current conclusions section should be moved to the discussion section.
In addition, the discussion section should be widened by better highlighting the impact of the findings in comparison with current literature. Limitations and future outlook should be more detailed.
I believe these are relevant indications to keep the quality of the work high.
We are accepting your article on the condition that you must incorporate the above suggestions in your final version.
Regarding the overall manuscript, the authors have improved the manuscript according to the provided comments and suggestions.
Regarding the methodological aspects, the authors have improved the manuscript according to the provided comments and suggestions.
Regarding the validity of the findings, a proper Discussion section is still missing.
The current Discussion section has not been effectively improved. It still provides only a summary of the work. A proper Discussion section, as stated in the previous revision round, should provide, above all, an interpretation of the results in view and in comparison with the relevant literature in the field.
Some of the above observations are instead reported in the Conclusions section. Therefore, I would suggest limiting the Conclusions section to a very brief summary of the work. Some parts of the current Conclusions section should be moved to the Discussion section.
In addition, the Discussion section should be widened by better highlighting the impact of the findings in comparison with current literature. Limitations and future outlook should be more detailed.
I believe these are relevant indications to keep the quality of the work high.
No additional comments.
Тhe authors corrected the comment regarding the statistical evaluation of the results, there are no other comments
no comment
no comment
no comment
Thank you for revising the paper, mostly the comments are addressed and the reviewers are quite hopeful that in another round the authors can make more improvements in the paper, Their comments are appended with this email. One of the reviewers has suggested citing some related articles, I would like to make it clear that according to journals policy, you should only add the references if they are relevant and truly needed to cite. Not citing the requested references will not have any effect on my decision on your paper.
[# PeerJ Staff Note: It is PeerJ policy that additional references suggested during the peer-review process should only be included if the authors are in agreement that they are relevant and useful #]
The authors properly addressed the comments highlighted in the previous report.
English has been improved and additional literature references have been provided.
Line 144 "...numerous methods for effective interpretation of FHR were proposed here [14]....": reference [14] seems to be old. I would suggest supporting this statement by citing some more recent works. Indeed, some of them have already been reported by the authors in the point-by-point reply but have not been included in the manuscript:
DOI: 10.3390/s21186136; DOI: 10.3390/app9245421; DOI: 10.3390/jcm7080223; DOI: 10.3390/bioengineering9010008; others...
Figures and tables are clear and well presented and results are properly discussed.
The methods are described with adequate details and rigour and the relevant comments provided in the previous round have been properly addressed.
As recommended in the previous review round, a Discussion section has been added.
However, it provides only a summary of the work. Instead, a proper Discussion section should provide a summary of the findings and, above all, an interpretation of the results in view and in comparison with the relevant literature in the field. Indeed, the authors provided a very good literature review, and then in the Discussion section, it is expected that they demonstrate the unique contribution of their research with respect to the state of the art. Some parts of the literature review could be recalled in the Discussion section.
Therefore, I would encourage revising and strengthening the Discussion Section by providing an extensive interpretation of the findings and motivating why the presented findings are relevant, why they should be considered important in both the clinical and technical field of interest, what is the expected impact. Moreover, in this section and in the Conclusion section, limitations and future research directions of the study could be provided.
While it is expected that a Discussion section is wide and extended, on the contrary, the Conclusions should provide a brief summary of the work and could be more concise.
I commend the authors for providing detailed description of all the comments provided before.
The methodology has been explained in more detail.
Findings are well provided with underlying data in a robust, and controlled manner.
At some places, there are some typing mistakes that should be improved before publication.
no comments
no comments
no comments
The authors gave satisfactory answers to all the reviewer's comments, except for the question about the statistical procedure used in the expert evaluation of the quality of the results. Perhaps this question is not so important for the authors.
Your manuscript has been reviewed by the experts and they are suggesting a number of changes for your paper. Please carefully incorporate the changes and resubmit
Also, Reviewer 2 has requested that you cite specific references. You may add them if you believe they are especially relevant. However, I do not expect you to include these citations, and if you do not include them, this will not influence my decision.
[# PeerJ Staff Note: Please ensure that all review comments are addressed in a rebuttal letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate. It is a common mistake to address reviewer questions in the rebuttal letter but not in the revised manuscript. If a reviewer raised a question then your readers will probably have the same question so you should ensure that the manuscript can stand alone without the rebuttal letter. Directions on how to prepare a rebuttal letter can be found at: https://peerj.com/benefits/academic-rebuttal-letters/ #]
Please check the review report attached.
Please check the review report attached.
Please check the review report attached.
Please check the review report attached.
The language is almost clear, despite some sentences that need to be clarified to help the reader follow the paper flow. Minor spelling mistakes should be revised.
Adequate background is provided and appropriate references have been cited. However, some content should be supported by additional relevant and recent references on the topic of the work.
The content organization and presentation can be improved and clarified, as detailed in the following suggestions to improve the quality of the manuscript:
- Line 89: please define acronym CTG before using it.
- Line 90: please define acronym FHR before using it.
- Line 94: please define acronym UC before using it.
- Lines 90-92: "The majority of studies...", please provide references to support this statement.
- Lines101-103: "...Initially, electrocardiography (ECG), ultrasound Doppler method, phonocardiography (PCG) and magnetocardiography (MCG) are amongst the well-known techniques for measuring CTG...", please clarify the meaning of this sentence. ECG, PCG, MCG are NOT techniques for measuring CTG. The measured signals are different, please clarify this concept.
- Line 108: please define acronym RCOG before using it.
- Lines 107-109: "...As aforementioned, obstetricians referred strictly to the RCOG guideline for...", please revise the content of this section. Indeed, RCOG has not been mentioned earlier in the Introduction section.
- Lines 113-114: "...There persists a lack in finding a common consensus in defining FHR variability (FHRV)...", please provide references to support this statement. I would suggest looking at the most recent literature on this topic, e.g. some of the following works could be of interest to the authors to enrich their manuscript: DOI: 10.3389/fped.2021.661400; DOI: 10.3390/s21186136; DOI: 10.3389/fnins.2020.561186; DOI: 10.3389/fphys.2020.578898
- Lines 121-122: "...As previously stated, numerous methods for effective interpretation of FHR were proposed here...", provide additional references to support this statement.
- Line 162: "...Various methods for extracting FHR features have been used to enhance efficiency...", please provide references to support this statement. I would still suggest looking at the previously suggested works or even other papers to strengthen the sentence: DOI: 10.3389/fped.2021.661400; DOI: 10.3390/s21186136; DOI: 10.3389/fnins.2020.561186; DOI: 10.3389/fphys.2020.578898
- Lines 162-163: "...The Rule-Based Algorithm for Intrapartum is effective and widely applicable...", please support this sentence. Are there works confirming this hypothesis. Is it an assumption of the proposed work?
- Line 166: "...cardiotocograph (CTG) classification...", please define acronym CTG the first time it appears in the Introduction section.
- Paragrah 2 "Literature Review": it is appreciated to have a literature review in the manuscript. However, my suggestion is to summarize the literature review to the most representative example and include it in the Introduction section. In this way, the reader will not be distracted by the main aim of the work that is presented at the end of the Introduction section. In addition, the authors could provide some framework before presenting the literature review, e.g. by explaining how they chose the papers to include in the review, if they focused on specific applications/techniques, etc...
The research is in line with the scope of the journal.
The research question and the gaps in the current knowledge could be better clarified. In particular I would clarify the following points:
- Lines 147-148: "...Hence, this study is significant as it proposes...", the significance of the study should be discussed at the end of the manuscript (Discussion and / or Conclusion section). It does not make sense to state that the work is significant before showing and commenting on the obtained results.
- Lines 149-161: This section appears to be not connected with the previous text. At first, the authors comment on the CTG feature extraction methods, then they talk about labour complications, and then they talk again about the feature extraction methods.
The methods are described with adequate details and rigour.
Figures formatting should be revised.
In particular:
- Figure 4 resolution is poor
- Figure 6 is too stretched
- Figure 23 and Figure 24 resolution is very poor
The results are well described and commented on but the Discussion and comparison with the relevant literature in the filed could be improved.
No Discussion section is provided. I would suggest adding a Discussion Section where the authors should discuss the main findings in view of the current knowledge in the field.
Conclusions are well stated.
n.a.
The English language is used professionally but it is not clear and it is ambiguous.
In the Abstract, at line 69, correct the tense of the verb ... it should be .... was classified ....it should not be ....... will be classified.
Line 69 and 70 … need to specify the name of the classifier
Line 76… algorithms? … proposed algorithm’s results or any other algorithm
Line 78: heading not in Bold format
Line 83 and 83 …. Which algorithms were developed by MATLAB?
In the Introduction section, the paragraphs in terms of information are not linked to each other. Long paragraphs with switching and mixing of concepts are confusing for the reader.
Line 89 … needs to mention full term with CTG
Line 90 … which studies … citation is required
Line 90 and 91 …
What is FHR? Write full term with an acronym for this in Introduction when using FHR for 1st time.
Which feature extraction?
Which computer vision approach?
Line 94… what is UC?
Line 105 to 125, 131 to 143 … font style is different
Line163 … citation is required for … The Rule-Based Algorithm for Intrapartum is effective and widely applicable.
Line 166 … how authors can say? … which will help obstetricians.
Comments for the section about Literature Review:
The literature review is not appropriately compiled and linked with the proposed approach. Explanation of a number of papers is included without creating any proper link or connection with the proposed work.
Line 186 to 202 … long explanation of reviewed paper in two paragraphs without relating with the proposed approach.
Line 198 … classification of what? Which classes?
Line 200 and 201 … no explanation of paper [29]. In this paper what issue was addressed and what was the final outcome? How this paper is related to your study? Why only refer for classifier performance?
Line 202 … What is R?
Line 204 and 205 … authors are not the reviewers of [30]… why using ‘trends should be
Long paragraph … line 235 to 267.
Line 224 is related to which paper? [35] or [36]? If related to [35] why in the next paragraph?
Line 265 … Moreover, in [47], the researchers will reveal … When? This paper is part of a literature review not of future work.
Comments about Figures:
Figure 12 and 13: caption is not appropriate or incomplete in terms of meaning. e.g. Need to mention that an algorithm or flowchart or step by step procedure of ….
Figure 14: Why back arrow is pointing in upward direction? Why there is an independent rectangle with red text in a flowchart?
Figure 16: The shown information is difficult to interpret because of missing legends.
Figure 17: Identification of different modes of deceleration
What is the condition to select early Deceleration or late Deceleration?
Figure 18,19 and 21: Caption is not descriptive or appropriate. Figure is difficult to interpret. Authors may need to use proper legends.
Figure 21: What is (A) and (B). Why independent rectangles are shown with red font in a flowchart.
Figure 22:
General caption is missing. Figure is blur and not readable. Redraw to present clear information.
Figure 23: Figure is blur and difficult to read.
Figures 24 to 34: x-axis and y-axis labels are missing. What is on x-axis and y-axis?
Figure 25 and 26: Figure caption is not appropriate.
The methodology should be explained with reasons, more details and explanations. The dataset should be clearly described.
In the section Materials and Methods:
DESCRIPTION OF DATASETS needs a proper breakdown of paragraphs and clear explanation to increase the understandability of used datasets.
Line 305: To measure signal loss, the number of values less than 50 b.p.m. is counted…. Why less than 50… what’s the reason?
Line 298: Mention if this is part of pre-processing stage … The signal condition- stage used the if statement in MATLAB source code to delete the breakdown to zero
line 329: Which feature extraction and classification algorithms are used?
Line 359: The best value of ()… What is inside these parenthesis ()?
Table 3: What is α? Mention in table caption and header.
What’s the reason for comparing with krupa et al 2008? Why only one and so old?
Paper needs major revision before acceptance.
The overall idea is good and worthwhile. Experimentation is sufficient. Results seem meaningful. Paper contribution or work is not justifiable by a long list of authors.
Introduction and Literature review needs to rewrite to link with main idea and proposed work. Explanation of datasets needs organization and clarity. Steps of Methods should be clearer and more organized. A lot of loopholes are there. A number of concepts need to explain with reason. Include some more explanation of MATLAB pre-processing, features extraction and rule-based classification algorithms.
Figures and Tables should present information with more clarity.
There is a need to write easy but meaningful sentences to maintain the understandability of your work. Complicated sentences are distracting and lose focus. Mixed-use of the past and present tense in writing the same time scenario is inappropriate.
Write full term with the acronym in Introduction or where writing for 1st time in the paper.
Authors need to revise the paper carefully for Spelling Mistakes. Two spelling mistakes have been highlighted here: Figure 18: periodas, Figure 21: classification
Authors need to be consistent with one style for caption, legends, x-axis and y-axis text for figures and Tables.
no comment
Despite the rather large literature review, the knowledge gap is not explicitly highlighted, and thus the specifics of the problem that the authors are solving remains unclear.
The research questions are not explicitly formulated. According to the reviewer, the following statements can be distinguished as a candidate for a research question, but the content and results of the article are vaguely or incompletely related to these statements and, in turn, raises questions from the reviewer. Namely:
Lines 60-63: <The goal of this project is to create a reliable feature extraction algorithm for the FHR as well as a systematic and viable classifier for the CTG utilizing the MATLAB platform, all the while adhering to the recommendations established by the Royal College of Obstetricians and Gynecologists (RCOG)>. However, the algorithm implemented by the authors (Figure 2) is much narrower (contains much fewer parameters and transitions) than the one contained in the recommendations (The use of electronic fetal monitoring Clinical Practice Algorithm, https://ctgutbildning.se/images/Referenser/NICE-guidelines-FHR-monitoring-2001.pdf).
Lines 147-148: <this study is significant as it proposes reliable pre-processing, feature extraction algorithms along with CTG signals classification>
As far as preprocessing is concerned, as far as the reviewer knows, noise reduction in CTG is a serious problem, for which a variety of signal processing methods are used (see for example https://www.researchgate.net/publication/332333574_Automatic_Cardiotocography_Diagnostic_System_Based_on_Hilbert_Transform_and_Adaptive_Threshold_Technique). In the article under review, the authors limit themselves to using a moving average filter with a single parameter w (the units of measurement are not named), and they give a very brief description (Lines 306-309): <the ideal value for the filter window size w is determined. According to the statistics, the window size should be between 30 and 50 points. Experts chose 308 w=30 for the moving average filter in this research paper because it provides the best visual representation>. Why is this magnitude ideal? What statistical procedure the experts used? How does that very "best visual representation" coincide with RCOG recommendation and in what way is it proved? Тhe figures referred to by the authors (Figures 21 and 22) do not clarify the situation.
As concerning to feature extraction algorithms, <A virtual imaginary baseline R is assumed in this study, which is equal to the mean value of the FHR signal of a 30-minute segment> (lines 344-345). However, the formula given by the authors (lines 344-345) describes the calculation of P as a constant, while in Figure 4 P is shown as a moving average. Then, on line 359, the value of alpha is entered, but the formula for its calculation is not provided, thus further calculations for the minimum (L) and maximum (H) limits of the FHR signal are not clear.
Since the original question investigated has not been formulated, and there are doubts about the formulation of research tasks (see above), it is not possible to assess the completeness of the conclusions.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.