DeepBreastCancerNet: A Novel Deep Learning Model for Breast Cancer Detection Using Ultrasound Images
Round 1
Reviewer 1 Report
1. Explicitly state your novelty and motivation in the Introduction section.
2. The literature review section is badly organized and difficult to follow. The state-of-the-art is not very well covered. I recommend that authors should add a summary of limitations of previous works as a motivation of this study.
3.Figure 2 is not displayed clearly.
4. Did you do some hyperparameter optimization/tuning? The ablation study is needed.
5. The ablation study of proposed method is needed.
6. This study lack of novelty and the experiment is too simple.
7. The proposed method does not have obvious performance advantage compared with the selected competing methods.
Author Response
Many thanks to you for reviewing our manuscript (applsci-2149796) entitled “DeepBreastCancerNet : A Novel Deep Learning Model for Breast Cancer Detection Using Ultrasound Images” The reviewers’ comments were beneficial to improve the quality of our manuscript, and therefore we revised our manuscript accordingly. As suggested by the reviewers, we modified the original manuscript and edited it thoroughly.
Author Response File: Author Response.docx
Reviewer 2 Report
In this article, a novel DeepBraestCancerNet DL model for breast cancer detection and classification has been introduced. Breast cancer is very common and has become a global issue which causes hundreds of women's deaths each year.
DeepBraestCancerNet DL model is a non-invasive solution which is a good contribution to the field of ML applications and medical science advancement. However, a few fundamental misunderstandings have been observed, which need to be corrected before acceptance. Suggested revisions are:
-
1) A typo error has been observed in the title of the article and needs to be corrected.
-
2) The literature review needs to be updated. Most of the references are from conferences which can be replaced by journal articles. Some suggestions are below;
-
[1] Abdul Halim, Ahmad Ashraf, et al. "Existing and Emerging Breast Cancer DetectionTechnologies and Its Challenges: A Review." Applied Sciences 11.22 (2021): 10753.
[2] Syeda, Iqra Hassan, et al. "Advance control strategies using image processing, UAV and AI in agriculture: a review." World Journal of Engineering (2021).
-
3) In this paper, 70% of the data has been used for training and 30% for testing. It is therefore suggested that 80% of the data may be used for training and 20% for testing because the strength of this research is based on the accuracy of the model.
-
4) Many Grammatical errors have been observed. It is therefore suggested that the article may be reviewed again by all the authors for corrections.
Comments for author File: Comments.pdf
Author Response
Many thanks to you for reviewing our manuscript (applsci-2149796) entitled “DeepBreastCancerNet : A Novel Deep Learning Model for Breast Cancer Detection Using Ultrasound Images” The reviewers’ comments were beneficial to improve the quality of our manuscript, and therefore we revised our manuscript accordingly. As suggested by the reviewers, we modified the original manuscript and edited it thoroughly.
Author Response File: Author Response.docx
Reviewer 3 Report
The authors provide an interesting deep learning-based method for classifying breast cancer detection. I strongly suggest that the journal Applied Sciences publish this article. However, I have some reservations about the findings, and the authors may need to revise their assertion in light of the following considerations.
Comments for author File: Comments.pdf
Author Response
Many thanks to you for reviewing our manuscript (applsci-2149796) entitled “DeepBreastCancerNet : A Novel Deep Learning Model for Breast Cancer Detection Using Ultrasound Images” The reviewers’ comments were beneficial to improve the quality of our manuscript, and therefore we revised our manuscript accordingly. As suggested by the reviewers, we modified the original manuscript and edited it thoroughly.
Author Response File: Author Response.docx
Reviewer 4 Report
Manuscript is about “DeeBreastCancerNet : A Novel Deep Learning model for Breast Cancer Detection using Ultrasound Images”. The main contributions are :
According to authors:
We proposed a DeepBreastcancerNet deep learning model for breast cancer detection and classification.
We illustrate that by the adopting pre-trained ImageNet models, TL may achieve excellent results in BC detection.
3. We use data augmentation to increase model performance and avoid the problem of overfitting.
4. We evaluate and compare the performance of various DNN-based BC identification techniques using four different performance matrices: such as 114 accuracy, precision, recall, and F-score.
I don’t agree with the contribution # 1 and 4 as they are non-scientific, I would like to suggest here that author must write/summarize as 02 contributions of proposed work.
Values presented must be statistically verified
Mathematical model requires more explanation.
Citations are missing in some tables such as table-5
Parameter section details require detailed explanation with significance of each parameter
Authors are encouraged to cite and discuss the recent deep-learning models
‘Efficient and Low-Cost Skin Cancer Detection System Implementation with a Comparative Study Between Traditional and CNN-Based Models’
‘A Hybrid CNN for Image Denoising’
‘Stud Pose Detection Based on Photometric Stereo and Lightweight YOLOv4’
‘Novel Approach to Evaluate Classification Algorithms and Feature Selection Filter Algorithms Using Medical Data’
There are grammar errors in this manuscript and it must be corrected.
How authors have selected the research used for comparison, this is not clear and require justification.
How authors divide the research study already published in the field of Breast Cancer Detection using Ultrasound Images
Author Response
Many thanks to you for reviewing our manuscript (applsci-2149796) entitled “DeepBreastCancerNet : A Novel Deep Learning Model for Breast Cancer Detection Using Ultrasound Images” The reviewers’ comments were beneficial to improve the quality of our manuscript, and therefore we revised our manuscript accordingly. As suggested by the reviewers, we modified the original manuscript and edited it thoroughly.
Author Response File: Author Response.docx
Reviewer 5 Report
Figure-2 and figure-5 are not clear.
Texts in Figure-2 are not readable
overall good work
Author Response
Many thanks to you for reviewing our manuscript (applsci-2149796) entitled “DeepBreastCancerNet : A Novel Deep Learning Model for Breast Cancer Detection Using Ultrasound Images” The reviewers’ comments were beneficial to improve the quality of our manuscript, and therefore we revised our manuscript accordingly. As suggested by the reviewers, we modified the original manuscript and edited it thoroughly.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
The authors have made the corrections. The manuscript seems good.
Author Response
Thank you very much for accepting the replies in response to your comments. We appreciate your dedication and commitment to research community.
Kind Regards
Reviewer 3 Report
The authors provide an interesting deep learning-based method for classifying breast cancer detection. I
strongly suggest that the journal Applied Sciences publish this article. However, I have some reservations
about the findings, and the authors may need to revise their assertion in light of the following
considerations.
1) The author did not highlight the change part in the abstract in the revised manuscript.
2) The quality of figures 2. and 4. is not good, the author needs to change the figures.
3) The author should compare the manuscript with the same methodology i.e DCNN method in the comparison table.7.
Author Response
Thank you, for the valuable comments that help us to improve the quality of the manuscript. The detailed reply to the comments is attached below. We are hopeful that this version of the manuscript will satisfy the needs of the esteemed reviewer. Kind Regards
Author Response File: Author Response.docx