Fetal Hypoxia Detection Using Machine Learning: A Narrative Review

Round 1

Reviewer 1 Report

The authors decribed and compared current literature on fetal hypoxia using different machine learning methods. They also did some discussion and point out future direction. I belive this is a nice review and merits publication.

The English is fine

Author Response

Thank you very much for taking the time to review this manuscript. Please find attached the detailed response.

Author Response File: Author Response.pdf

Reviewer 2 Report

I have evaluated this manuscript taking into consideration SANRA methodology for the quality assessment of narrative review articles.

1. JUSTIFICATION OF THE IMPORTANCE OF THE ARTICLE:

-        The authors outline the topic/clinical problem discussed into their paper as being the detection of fetal hypoxia during pregnancy and labor stage.

-        The authors explicitly justify that they selected this clinical problem due to its importance in the detection of fetal hypoxia for reduction of neonatal mortality and of serious effects on the baby’s survival and well-being.

2. STATEMENT OF CONCRETE AIMS

-        The specific objective of the paper seems to be clearly presented in the Introduction: “conduct a comprehensive review of past studies focused on fetal hypoxia detection using ML techniques during pregnancy and labor - stage” (lines 45-47). The authors are interested to “identify gaps, limitations, and opportunities in current methodologies“(lines 47-48).

-        The objectives of the paper are also presented in the Methodology chapter: a) “outline the development of knowledge on fetal hypoxia” (line 53) and b) “guide the reader with an exhaustive look at the research literature in chronological order” (lines 52-53) are not the same with the one stated in the Introduction: “conduct a comprehensive review of past studies focused on fetal hypoxia detection using ML techniques during pregnancy and labor stage” (lines 45-47).

The authors have to decide the number and content of the objective(s) of the paper and to include it/them only into the Introduction.

3. DESCRIPTION OF THE LITERATURE SEARCH

-        The authors only refer briefly to the literature search without mentioning the search terms. It is necessary to provide more details.

4. REFERENCING

-        Here are presented the papers referred by the authors in different parts of their manuscript (that is in Figure 1, in the text of the paper (lines 64-406) and in Tables 1, 2 and 3):

The following inconsistencies are presented below:

o   The specification of the references in Figure 1 is not consistent with the text of the paper (lines 64-406) or with Tables 1, 2 and 3.

o   The total number of the papers analyzed by the authors is 36 and not 35 as resulting from Tables 1, 2 and 3.

o   Figure 2 and 3 are consistent with the values from A and B points (see table before) and not with Tables 1, 2, 3.

The authors have to solve all these inconsistencies.

5. SCIENTIFIC REASONING

-        Appropriate evidence is generally present.

6. APPROPRIATE PRESENTATION OF DATA

-        Data are generally presented appropriately, but the already found inconsistencies have to be solved in order to have the final format of Figures 2 and Figure 3.

SOME OTHER RECCOMENDATIONS

-        In my opinion, the content of the “Manuscript organization” has to be included at the end of the Introduction.

-        FHR, CTG, FIGO, ML, DL, UC, UCI, MAE, DT, GBM, RF, SVM, GLMNET, KNN, ADA-B, LR, AUC, ANN, LSTM, NN, MLPNN, PNN, RP, LM, CAD, VGG16, NB, TPR, AUROC, GLVQ, .BT, SMOTE, AC, EEG, GB, LGBM abbreviations are not defined. It would be good to include into the manuscript a table with all the abbreviations used into the text.

-        Figures 1-3 and Tables 1-3 are not referred into the text.

I hope my feedback is useful to the authors in improving their paper and wish them all the best in pursuing this important area of research.

Author Response

Thank you for taking the time to review this manuscript. Please find attached the detailed response, with comments' responses highlighted in yellow.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and suggestions

This paper provides a clear overview of the importance of detecting fetal hypoxia and highlights the significance of utilizing technology, particularly machine learning, for more accurate and timely detection. The paper is well-organized and looks impressive.

Here are some suggestions:

1.       Consider breaking down complex sentences into shorter ones for improved readability. This will help the reader better grasp the information.

2.       Use precise terms to convey your message. For example, instead of "vitality of interpreting," you might say "critical importance of interpreting."

3.       In the sentence, "Implementing machine learning and deep learning techniques ensures more timely and precise detection," consider specifying how these techniques achieve more timely and precise detection.

4.       Provide a brief introduction that outlines the current challenges in detecting fetal hypoxia and why advanced technologies are necessary. This will give readers a context for the importance of your review.

5.       Ensure a smooth transition between sentences and paragraphs. For instance, connect the sentence about classifying cases with the subsequent mention of machine learning, emphasizing how they complement each other.

6.       Offer specific examples or statistics to support your claims about the benefits of machine learning in fetal health prediction. This can add credibility to your argument.

7.       Overall, your paper is well-structured, and with a few adjustments, it can become even more impactful.

8.       I would like to advise the authors to include the following papers in the reference list of their revised version:

(I)                  Lwin, Tunn & Tun, San & Tin, Pyke & Zin, Thi Thi. (2023). Evaluating Imputation Strategies for Handling Missing Data: A Comparative Study. 508-509. 10.1109/GCCE59613.2023.10315259.

(II)                Thi Thi Zin, Tunn Cho Lwin, Pyke Tin (2024) A Novel Stochastic Model for Analyzing Heart Rate Variability in the Heart-Brain Signal Communication System; NCSP'24, 2024 RISP International Workshop on Nonlinear Circuits, Communications and Signal Processing, Honolulu, Hawaii, Feb. 27 - Mar. 1, 2024

Only a minor check is necessary.

Author Response

Thank you for taking the time to review this manuscript. Please find attached the detailed response, with comments' responses highlighted in pink.

Author Response File: Author Response.pdf

Reviewer 4 Report

The authors did a narrative review on fetal hypoxia detection, the topic is interesting and very important. Overall, the article is written well. Here are my comments:

·         In abstract, overall review results can be given.

·         “Comprehensive” statement should be removed because this is a narrative review not a systematic one.

·         What is FHR signal? Give full name.

·         In p1 line 24, please give year period.

·         In p2 l40, the authors need to give references for AI systems.

·         All abbreviations should be given in full form when they first used.

·         Contributions should be listed and clearly given at the end of introduction.

·         Organizational paragraph that summarizes sections should be added to intro not in 2.2.

·         For sections 3.3-3.6, a discussion paragraph should be added at the end of sections.

·         In order to observe ML-DL trend, a graph by yearly basis can be constructed.

·         Model-wise result graph should be given (ex: how many studies trained RF, SVM?).

·         What about research from 2024?

·         Discussion part should be extended.

·         In the paper, overall, more figures are needed.

·         Fig. 2 is not clearly understandable.

·         Maybe studies that used the same dataset can be grouped together and analyzed.

·         In the future work, authors also need to address future work of this paper particularly.

·         Conclusion should be extended.

Author Response

Thank you for taking the time to review this manuscript. Please find attached the detailed response, with comments' responses highlighted in green.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

DESCRIPTION OF THE LITERATURE SEARCH

-        At my comment “The authors only refer briefly to the literature search without mentioning the search terms. It is necessary to provide more details”, the authors specify the abbreviations used in their paper and not the terms used in searching the literature. This would be a real help for understanding how is made the selection of papers used in this narrative review.

SOME OTHER RECCOMENDATIONS

-        Tables 1-3 are not referred into the text.

I hope my feedback is useful to the authors in improving their paper.

Author Response

Thank you for taking the time to review this manuscript. Please find attached the detailed response, with comments' responses highlighted in yellow.

Author Response File: Author Response.pdf

Reviewer 4 Report

The authors answered all my queries.

I have one minor comment:

In section line 75 authors need to give references to these domains.

Author Response

Thank you for taking the time to review this manuscript. Please find attached the detailed response, with comments' responses highlighted in green.

Author Response File: Author Response.pdf