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.
Author has addressed reviewer comments properly. Thus I recommend publication of the manuscript.
the authors have made all the requested revisions to the manuscript.
the authors have made all the requested revisions to the manuscript. We have carefully reviewed the feedback provided and addressed each point accordingly.
the authors have made all the requested revisions to the manuscript. We have carefully reviewed the feedback provided and addressed each point accordingly.
The original Academic Editor is unavailable. Based on the referee reports, I recommend a major revision of the manuscript. The author should improve the manuscript, taking carefully into account the comments of the reviewers in the reports, and resubmit the paper.
1. The manuscript is presented with a high level of writing quality.
2. Please position your research within the broader context of scientific knowledge and identify how it fits into the already existing landscape of academic literature. This will help to better understand the unique contribution of your work in the specific field of diabetic eye diseases.
3. The editorial recommendations of the journal are ensured.
4. The article demonstrates precision and a rigorous methodology in the formal results section. This establishes the credibility and reliability of the research findings in the field of detection and diagnosis of diabetic eye diseases using a two-phase transfer learning approach.
It is crucial to clearly pinpoint the existing voids in academic literature and illustrate how the outcomes of this investigation can address these gaps more explicitly. A more comprehensive examination of existing literature could enhance the rationale for the significance of the research and underscore the distinctiveness of the anticipated contributions.
A more detailed description of the statistical robustness of the data, the control methods employed, and the rigor of the collection process is necessary to enhance the credibility of the drawn conclusions
The authors have not included even a single recent manuscript. They have simply written that they have made changes in the manuscript, but have not. Most comments were answered that the changes are made in a few sections. However, most of the responses were not reflecting the proper response to a comment. The resolution of the Figures is not sufficient which can be improved further. The changes made in the current version of the manuscript are not highlighted. Authors are suggested to remove a few words such as "inform" in the manuscript. Simply they have written they have made the changes but not. I don't want to point out every comment again and again. I suggest the authors consider this as a final opportunity to make all the corresponding changes and resubmit the manuscript again.
NA
NA
I have not found the response to reviewers' comments in the current version of the manuscript. I suggest the authors highlight and answer every comment.
Dear Authors,
Thank you for submitting your paper to PeerJ Computer Science. Kindly consider all of the reviewers' comments to improve the quality of the submitted paper. Here are a few important concerns about the paper that need to be addressed:
The research gaps need to be discussed and highlighted well supported by up-to-date references. The proposed solutions are not well-defined, please state clearly your main contributions.
In the experimental design, you need to explain how the hyperparameter tuning of the models was performed.
Discuss if there is any overfitting issue in the results.
Please follow the instructions to re-submit the revised paper with a detailed response to the reviewers' comments.
Kind regards,
Faisal Saeed
PeerJ Computer Science Associate Editor
**PeerJ Staff Note:** Please ensure that all review and editorial comments are addressed in a response letter and any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** The review process has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at [email protected] for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
1. The raw data (or code) of this article can be accessed without any issues, and it is well-described in English. The authors have ensured that the data files or code used in the study are easily openable and accompanied by clear and detailed explanations in the English language. The structure, format, and variables of the raw data are thoroughly documented. Additionally, any necessary instructions or preprocessing steps are clearly outlined, further enhancing the reproducibility and transparency of the study.
2. The language employed in this article is characterized by its clarity, comprehensibility, and professional tone. The information and ideas are effectively conveyed, making it easy for readers to understand and engage with the content. The author's command of language and proficiency in conveying complex concepts in a concise manner is evident throughout the article, ensuring a high level of professionalism in its presentation.
3. The article provides an adequate introduction and background to demonstrate how the work fits into the field of medical image interpretation. It presents a clear overview of the context and motivations of the study, highlighting its significance and relevance in relation to prior research. Furthermore, appropriate references to DED disease diagnosis are provided, showcasing a comprehensive understanding of the existing work in the field. The article establishes a strong connection between the current research and previous knowledge, enabling readers to grasp how this study contributes to the advancement of the field
4. Figs and tables in this article have been presented without any inappropriate manipulation. They accurately represent the data and information being discussed, ensuring transparency and integrity in the visual presentation of the findings. The figures and tables serve as reliable visual aids that enhance the understanding of the content and provide essential support for the arguments and conclusions presented in the article. The authors have maintained the highest standards of ethical conduct in the representation of the data, ensuring that readers can trust the information conveyed through these visual elements.
5. This submission is self-contained and represents an appropriate unit of publication. It encompasses all the necessary components and findings relevant to the hypothesis being investigated. The article provides a comprehensive overview of the proposed two-phase transfer learning approach for detecting and diagnosing diabetic eye diseases. It includes detailed descriptions of the methodology, experimental setup, and evaluation metrics employed. The results presented in the article demonstrate the effectiveness and accuracy of the proposed approach in detecting and diagnosing diabetic eye diseases. Additionally, the article discusses the implications of the findings and their significance in the broader context of medical image analysis and diabetic eye disease diagnosis. As such, the submission stands as a complete and standalone contribution, providing a valuable and comprehensive exploration of the topic at hand.
6. The formal results presented in this article exhibit an exceptional level of rigor and depth. The authors have meticulously provided clear and precise definitions of all the relevant terms and theorems employed throughout their study. Each definition is meticulously crafted, leaving no room for ambiguity or misinterpretation. Furthermore, the article showcases a remarkable commitment to mathematical rigor by including detailed and comprehensive proofs for all the stated theorems. The proofs are logically structured, demonstrating the authors' keen understanding of the underlying principles and their ability to rigorously validate their claims. The inclusion of these thorough definitions and proofs elevates the scholarly merit of the article, ensuring the understanding and appreciation of the theoretical foundations on which the research is based.
1. This article is classified as a research article
2. Scientific soundness refers to the validity, reliability, and rigor of the research methods and findings. This includes evaluating the study design, data collection techniques, sample size, statistical analysis, and the overall adherence to established scientific principles. The article should demonstrate a clear research question or hypothesis, appropriate methodology, and a logical interpretation of the results. Additionally, the article provide relevant references to existing literature and theoretical frameworks. Methodological soundness refers to the appropriateness and effectiveness of the chosen methods for addressing the research question. In the case of this article, the evaluation would involve examining the two-phase transfer learning approach used for the detection and diagnosis of diabetic eye diseases. This may include assessing the transfer learning framework, the selection and preparation of the datasets, the training and validation procedures, and the evaluation metrics employed.
3. Based on the evaluation, the article titled "Detection and Diagnosis of Diabetic Eye Diseases using Two Phase Transfer Learning Approach" meets the specified criteria. The research question is well defined, relevant, and meaningful. The study identifies a knowledge gap and clearly states how it contributes to filling that gap.
4. The article titled "Detection and Diagnosis of Diabetic Eye Diseases using Two Phase Transfer Learning Approach" provides an exemplary demonstration of methods described with meticulous detail and comprehensive information, ensuring their reproducibility by fellow investigators. The authors have provided an extensive account of the experimental setup, including the transfer learning framework, dataset selection and preparation, training and validation procedures, and evaluation metrics employed. This level of transparency allows for precise replication and verification of the study's outcomes, establishing a strong foundation for further scientific advancements in this domain.
The article under consideration exemplifies the fulfillment Validity of the Findings
, as its conclusions are meticulously formulated, intricately linked to the original research question, and judiciously confined to align with the supporting findings. The authors demonstrate a commendable adherence to scientific rigor by ensuring that the conclusions drawn are derived exclusively from the empirical evidence generated by the study. Notably, assertions pertaining to causal relationships are substantiated through meticulously controlled experimental interventions, thus reinforcing the distinction between correlation and causation. By adhering to these principles, the article upholds the scholarly standards and enhances the credibility of its findings within the scientific community.
This is a well-written manuscript that only needs to undergo a few minor changes.
Authors should provide improved quality of Figures 1, 4
How did the authors perform the hyperparameter tuning of the models? I understand you reached those values after many experiments, but you should be more verbose on WHY you chose that final values. And How to optimize hyper parameters?
How did the authors avoid the overfitting problem?
Discuss the stability of the system in terms of complexity.
The content is clear, however, the grammatical mistakes should be avoided. The Literature provided is not the latest one and needs to be improved. The authors must include and discuss recent articles. The organization of the manuscript can be improved further. The authors can refer to the journals suggested in the review report. The methodology and results sections must be improved by highlighting the contribution and including the standard performance metrics for segmentation. Most of the Figure's resolution should be improved as mentioned in the review report.
The work is quite exciting and falls within the scope of the journal. The research gaps and solutions are not well-defined. More investigation is required in terms of literature and methodology and results section. The Literature, Methodology, and Results sections can be improved.
The contribution and novelty should be highlighted in the Methodology section with proper explanations and citing equations. Must address the data imbalance issue in the current approach.
The paper is structured well.
There are some syntax errors (grammars) in the language. for example: In the lines 13 and 14, you can remove s from verbs "improves" and "lessens".
For literature references Table 1, I wanted to bring to your attention a minor update that I need to address regarding the paper survey. The papers included are up to the year 2021. Considering the advancements and developments in the field, it would be valuable to incorporate papers published in 2022 and 2023 to ensure your survey reflects the most recent research landscape accurately.
In light of this, I propose that you collaborate to update the survey by adding relevant papers from the years 2022 and 2023. This will provide a more comprehensive and up-to-date overview of the field's progress and contributions.
I wanted to suggest the inclusion of cross-validation in your evaluation process. While your current approach of splitting the dataset into training, validation, and test sets is well-founded, incorporating cross-validation can further strengthen the reliability of your findings.
Cross-validation involves partitioning the dataset into multiple subsets and iteratively training and testing the model on different combinations of these subsets. This helps mitigate the potential bias that can arise from a single train-validation-test split and provides a more comprehensive assessment of the model's generalization performance.
By implementing cross-validation, you'll be able to showcase the stability and consistency of your proposed approach across different data partitions. This can lead to more robust conclusions and lend greater confidence to your results.
I wanted also to add F1 score to your results. F1-Score: The harmonic mean of precision and recall. It provides a balanced measure that considers both false positives and false negatives.
You proposed a segmentation enhancement and I want to evaluate the proposed segmentation method using the measures: Dice Coefficient or Intersection over Union (IoU): These measures are used for segmentation tasks and quantify the overlap between predicted and ground-truth segmented regions.
I wanted to draw your attention to a concern in Figure 7 of your paper. It appears that there is an overfitting issue, which could potentially affect the validity of the results. I recommend investigating this matter to ensure the accuracy and reliability of your findings. Please fix the issue of over-fitting by applying some approaches to resolve the issue and reduce the differences between the training and validation.
The paper introduces improvements in segmentation and transfer learning, which are well-structured. However, the proposed segmentation method lacks evaluation. I believe it is essential to assess and compare this method with approaches presented in other related research papers on the same topic.
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.