Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

A very well-written review article. A few things/questions, 

1. The content is lengthy. I suggest making it concise. Briefly introduce each modality. Shorten or summarize each paper reviewed. The clinical application part of each modality is lengthy as well. Please summarize.
2. As a clinician, I do not find the examples for each modality particularly helpful or particularly strong in their implications, such as figures 6 and 7. Consider making readers understand why those examples matter. 
3. In the section on "Trends in Computed Thermography Imaging", the example figure 9 and the related works in 5.2 are all CT scans. This discrepancy creates confusion. Please explain why using CT examples/works in the section on thermography.
4. In the "other image modality" section, it is crucial to review the nuclear medicine scan, an important modality for cardiac imaging. Please consider including it. 
5. The overall discussion and conclusion parts are lengthy. Consider making it concise and to the point. It is helpful to distinguish clinical validation from experimental results, mention some prospective studies, and consider less speculative AI trends. 

Author Response

Please see the attched file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1.The originality and the scientific value of the subject are good. An important problem having direct application, as mentioned previously, is treated. 
2.The abstract is a good summary of the paper since the material used, the performed experiments, the methods, the results, the conclusions of this work are all clearly mentioned. 
3.The introduction is very satisfactory. The materials, their applications, the aim of the work, previous research on this subject, applied methods and especially the use of the investigated material are explained in detail. 
4.Presentation of materials is very adequate. The composition, the origin of the material used, dimensions of samples etc are all mentioned. 
5.Presentation of the experimental work is in detail.

Author Response

Please see the attched file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript categorizes the literature according to imaging modalities, including ultrasound, MRI, X-ray, CT, ECG, and other types. The datasets, deep learning techniques, clinical targets, and evaluation metrics used in the reviewed works are analyzed. The authors also attempt to identify research trends and gaps and propose future directions such as the use of transformers, explainability, and uncertainty quantification. Overall, the manuscript surveys a broad range of related works. Several aspects could be further improved, as detailed below.

1. Although the manuscript outlines the process of collecting and filtering papers, the search strategy and inclusion criteria are not sufficiently detailed. The authors should clarify the search queries, screening steps, and selection criteria to improve transparency and reproducibility.
2. In Fig. 5, ECG is treated as an imaging modality, which may cause conceptual confusion since ECG is a physiological signal rather than an imaging technique. A clearer classification of imaging and non-imaging data sources would improve the organization of the review.
3. The manuscript compares metrics such as AUROC and Dice scores across different studies, but these results are based on different datasets and experimental settings. The authors should clarify that such metrics are not directly comparable.
4. Several deep learning models such as CNNs, U-Net, Mask R-CNN, and YOLO-based frameworks are discussed, but it lacks clear architectural illustrations. It is suggested to include one or two schematic diagrams showing the typical network pipeline would help readers better understand the model structures.
5. The manuscript mixes segmentation, classification, detection, and regression tasks in the discussion without clearly separating them. Since these tasks have different objectives, evaluation metrics, and model architectures, the authors should explicitly categorize studies according to the task type to improve technical clarity.
6. The paper is quite long and contains extensive descriptions of individual studies, which sometimes leads to redundancy and reduced readability. It is suggested to condense the literature descriptions and focusing more on synthesizing insights from the reviewed works.

Author Response

Please see the attched file.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This is a review article that, by examining the diagnosis of heart diseases using different types of medical imaging, proposes future trends in the application of deep learning methods in this field. The article is generally clear in its presentation, well-organized in its data compilation, and effectively addresses the relevant issues. However, if predictive methods and models were employed in the analysis of developmental trends, it would better support its arguments. From a technical perspective, the level of innovation still requires further improvement.

Author Response

Please see the attched file.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for the reply. A few points that I did not fully agree, 

I still find this manuscript somewhat challenging to read due to its length and potential repetition. I unfortunately do not agree with the author's reply: "contemporary scientific reading does not rely solely on linear, cover-to-cover consumption." "Readers increasingly use AI-assisted tools, keyword searches."... I believe a scientific paper should be concise and to the point. The current manuscript can be difficult for readers to perceive the points. 
The CT section in the previous manuscript was erroneously titled as "thermography". This was revised in the current manuscript. The thermography of reference 86 was unfortunately cited erroneously in the CT section.    I would consider adding a cardiologist or cardiac imaging specialist to the authorship to enhance clinical relevance.    I would like to ask again, possibly to make the manuscript concise and prevent redundant and possibly repetitive information. 
         

Author Response

Please see attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript has been adequately revised in response to the reviewers’ comments, and I recommend its acceptance for publication.

Author Response

Please see attached file.

Author Response File: Author Response.pdf