SCGAN-generated skin-tone-balanced extension of the HAM10000 dataset for fairer dermatological AI

Safrina Kabir

doi:10.1371/journal.pone.0331081

Peer Review History

Original SubmissionAugust 11, 2025
26 Dec 2025 Decision Letter - Xiaohui Zhang, Editor Dear Dr. Kabir, Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process. Please submit your revised manuscript by Feb 09 2026 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org . When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file. A letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'. A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'. An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter. If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at . Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols .. We look forward to receiving your revised manuscript. Kind regards, Xiaohui Zhang Academic Editor PLOS One Journal Requirements: When submitting your revision, we need you to address these additional requirements. 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf 2. Please note that PLOS One has specific guidelines on code sharing for submissions in which author-generated code underpins the findings in the manuscript. In these cases, we expect all author-generated code to be made available without restrictions upon publication of the work. Please review our guidelines at https://journals.plos.org/plosone/s/materials-and-software-sharing#loc-sharing-code and ensure that your code is shared in a way that follows best practice and facilitates reproducibility and reuse. 3. Please update your submission to use the PLOS LaTeX template. The template and more information on our requirements for LaTeX submissions can be found at http://journals.plos.org/plosone/s/latex. 4. Thank you for stating the following in the Acknowledgments Section of your manuscript: “This research received no external funding. The work was conducted independently by the sole author, a student researcher, without institutional or grant-based support.” We note that you have provided funding information that is not currently declared in your Funding Statement. However, funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form. Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows: “The author(s) received no specific funding for this work.” Please include your amended statements within your cover letter; we will change the online submission form on your behalf. 5. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please ensure that your ethics statement is included in your manuscript, as the ethics statement entered into the online submission form will not be published alongside your manuscript. 6. If the reviewer comments include a recommendation to cite specific previously published works, please review and evaluate these publications to determine whether they are relevant and should be cited. There is no requirement to cite these works unless the editor has indicated otherwise. [Note: HTML markup is below. Please do not edit.] Reviewers' comments: Reviewer's Responses to Questions Comments to the Author 1. Is the manuscript technically sound, and do the data support the conclusions? Reviewer #1: Yes Reviewer #2: No ******** 2. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #2: No ****** 3. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #1: Yes Reviewer #2: Yes ****** 4. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: No ****** Reviewer #1: The manuscript SCGAN-Generated Skin-Tone-Balanced Extension of the HAM10000 Dataset for Fairer Dermatological AI presents a synthetic dataset of 10,000 dermoscopic images generated with a tone-conditioned GAN, intended to balance medium and dark skin tones while preserving the lesion-class distribution of HAM10000. The topic is timely and important, and a well-documented synthetic dataset with explicit skin-tone conditioning could be valuable for both fairness research and teaching. However, I have several major concerns about methodological clarity, validation, and how the fairness claims are supported. Comments: (1) The manuscript repeatedly positions the dataset as contributing to “fairer dermatological AI” and enabling “bias mitigation” and “equitable performance across skin tones.” However, the only evaluation presented is qualitative (visual inspection, tone histograms, and class counts). To support these fairness-related claims, I encourage at least a small quantitative experiment: e.g., train a simple baseline classifier on HAM10000 only vs. HAM10000 + synthetic images and compare performance across tone conditions. If such experiments are not feasible, the language should be toned down to clearly state that this is a hypothetical fairness resource and that the benefit for bias mitigation is an anticipated, not yet empirically demonstrated, outcome. (2) In the Introduction the authors state that images are generated “conditioned on both lesion class and skin tone,” and refer to a “Skin-Conditioned GAN (SCGAN).” However, in the Methods the conditioning is described only in terms of a 2-D one-hot tone vector [medium: (1, 0), dark: (0, 1)], with no explicit lesion-class conditioning. Please clarify is the generator explicitly conditioned on lesion class (e.g., via one-hot vectors, class embeddings, or class-specific batches), or is lesion class only used afterward to assign labels to generated images? (3) The authors define medium vs dark tones via a one-hot label (2-D vector) and post-generation RGB tint blending with fixed color vectors and blend strengths (40% and 80%). For dermatology and fairness work, this is a crucial aspect that needs more careful justification how do these RGB tints and blend ratios map to established skin tone scales (e.g. Fitzpatrick, Monk, or other multidimensional measures)? (4) Some useful details are provided (128×128 resolution, 100 epochs, Adam optimizer, LR=0.0002, β1=0.5, β2=0.999), which is good. However, reproducibility would benefit from further elaboration of batch size, training/validation split, number of iterations per epoch and hardware description (e.g., single GPU model, training time). (5) The manuscript cites several relevant GAN-based medical image synthesis works. Can the authors further clarify how does SCGAN differ conceptually and architecturally from existing multi-domain or conditional GANs (e.g. StarGAN-style frameworks cited in the references)? Reviewer #2: This paper presents a synthetic skin lesion dataset generated using a conditional GAN (SCGAN) to address the underrepresentation of darker skin tones in the HAM10000 dataset. The authors generated 10,000 images at 128×128 resolution, equally divided between medium and dark skin tones, while preserving the lesion class distribution of the original HAM10000 dataset. The work addresses an important and timely issue in medical AI fairness and provides an accessible synthetic resource. However, the methodological novelty, validation rigor, and clinical relevance of the generated data require further clarification and strengthening. 1. The main text includes detailed folder tree structures and file-naming examples, which are more appropriate for a dataset release webpage or repository README. Including such implementation-level details in the body of a journal article detracts from the professionalism of the manuscript and disrupts its scholarly presentation. 2. The manuscript does not provide principled reasoning for several critical methodological decisions, such as the choice of the specific GAN architecture. These design choices require clearer justification to be scientifically grounded. 3. The assessment of the generated images relies almost exclusively on visual inspection, without quantitative analysis or objective metrics. This makes it difficult to rigorously evaluate the realism, diversity, and utility of the synthetic data. 4. For a dataset-oriented paper, it is generally expected to include baseline experiments showing that the proposed dataset provides measurable benefits in downstream tasks. The absence of such experiments limits the ability to assess whether the dataset meaningfully contributes to improving model performance or fairness. ****** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our Privacy Policy .--> Reviewer #1: No Reviewer #2: No ******** [NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.] To ensure your figures meet our technical requirements, please review our figure guidelines: https://journals.plos.org/plosone/s/figures You may also use PLOS’s free figure tool, NAAS, to help you prepare publication quality figures: https://journals.plos.org/plosone/s/figures#loc-tools-for-figure-preparation. NAAS will assess whether your figures meet our technical requirements by comparing each figure against our figure specifications. https://doi.org/10.1371/journal.pone.0331081.r001
Revision 1
9 Feb 2026 Author Response Response to Reviewers I sincerely thank the Editor and both reviewers for their careful evaluation of our manuscript and for their constructive and insightful comments. Below, I provided a detailed point-by-point response to each comment. All changes have been incorporated into the revised manuscript. Reviewer #1 Comment (1): Fairness claims lack quantitative validation 1. Added a quantitative baseline classification experiment: A simple ResNet-18 classifier was trained under two settings: (i) using HAM10000 only, and (ii) using HAM10000 augmented with the proposed synthetic dataset. We report accuracy and macro-averaged F1-score for both settings. 2. Toned down fairness-related claims throughout the manuscript: The language has been revised to clearly state that the dataset is intended as a controlled experimental resource for fairness-aware research, rather than claiming direct or guaranteed bias mitigation in clinical systems. Comment (2): Clarification of lesion-class conditioning The generator is explicitly conditioned on skin tone only via a two-dimensional one-hot vector. Lesion class conditioning is not injected into the generator. Instead, lesion labels are assigned post-generation, proportionally to match the original HAM10000 class distribution. This design choice was made to preserve dataset-level class balance without enforcing class-specific generation constraints. All statements implying explicit lesion-class conditioning within the generator have been corrected for accuracy. Comment (3): Justification of RGB tint blending and skin tone definition In the revised manuscript, we explicitly state that the tone definitions used in this work are approximate and operational, rather than strict mappings to clinical scales such as Fitzpatrick or Monk. The RGB tint vectors and blend strengths are presented as controlled visual proxies designed to introduce consistent, distinguishable tone variation in the absence of reliable skin tone annotations in HAM10000. We now emphasize that these choices support controlled experimentation and fairness analysis, rather than clinical skin tone classification. Comment (4): Reproducibility details (batch size, hardware, etc.) We have expanded the Methods section to include: ● Batch size ● Number of iterations per epoch ● Training duration ● Hardware configuration (single-GPU training environment) Additionally, a code availability statement has been added to ensure full reproducibility, in line with PLOS ONE guidelines. Comment (5): Clarification of SCGAN vs existing GAN frameworks In brief, SCGAN was chosen because it enables explicit attribute-controlled synthesis without requiring paired domain translation or well-defined source–target mappings, which are prerequisites for frameworks such as StarGAN or CycleGAN. SCGAN allows stable, large-scale generation of tone-balanced data even when real medium- and dark-tone reference domains are sparse or inconsistently labeled. We presented the reasons of choice in the revised manuscript. Reviewer #2 Comment (1): Excessive implementation details in main text Detailed folder structures and file-naming conventions have been removed from the main text and are now referenced implicitly via the dataset README hosted in the public repository. This improves readability and aligns the manuscript with scholarly presentation standards. Comment (2): Justification of GAN architecture choice We have expanded the Methods and Discussion sections to provide clearer justification for the architectural design, including the choice of a tone-conditioned GAN, ResNet-based generator blocks, and PatchGAN-style discriminator. The discussion now explicitly links these design choices to the study’s goals of controllable tone synthesis, lesion feature preservation, and computational efficiency. Comment (3): Lack of quantitative evaluation We included the following: 1. Fréchet Inception Distance (FID) as a reference-level distribution similarity metric 2. A baseline classification experiment demonstrating the impact of synthetic augmentation We clearly state the limitations of these metrics and frame them as reference indicators, not as measures of diagnostic realism. Comment (4): Absence of baseline downstream task evaluation We agree and have added a baseline classification experiment, as described above. While intentionally simple, this experiment demonstrates that the synthetic dataset can positively influence downstream learning under controlled conditions. Closing Statement I again thank the reviewers for their constructive feedback. I believe the revised version more accurately reflects the capabilities and limitations of the proposed dataset and will serve as a useful resource for fairness-aware dermatological AI research. Sincerely, Safrina Kabir Sole Author and Corresponding Author Attachments Attachment Submitted filename: Response to Reviewers.pdf https://doi.org/10.1371/journal.pone.0331081.r002
3 Mar 2026 Decision Letter - Xiaohui Zhang, Editor SCGAN-Generated Skin-Tone-Balanced Extension of the HAM10000 Dataset for Fairer Dermatological AI PONE-D-25-42787R1 Dear Dr. Kabir, We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements. Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication. An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at Editorial Manager® and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact and clicking the ‘Update My Information' link at the top of the page. For questions related to billing, please contact billing support .. If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. Kind regards, Xiaohui Zhang Academic Editor PLOS One Additional Editor Comments (optional): Reviewers' comments: Reviewer's Responses to Questions Comments to the Author Reviewer #1: All comments have been addressed Reviewer #2: All comments have been addressed ******** 2. Is the manuscript technically sound, and do the data support the conclusions??> Reviewer #1: Yes Reviewer #2: Yes ****** 3. Has the statistical analysis been performed appropriately and rigorously? -->?> Reviewer #1: Yes Reviewer #2: Yes ****** 4. Have the authors made all data underlying the findings in their manuscript fully available??> The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.--> Reviewer #1: Yes Reviewer #2: Yes ****** 5. Is the manuscript presented in an intelligible fashion and written in standard English??> Reviewer #1: Yes Reviewer #2: Yes ****** Reviewer #1: The authors have addressed all the comments from the reviewers. I think the paper is good for publishment. Reviewer #2: This manuscript is well written, and I believe this manuscript meets the standards for the publication. ****** what does this mean? ). If published, this will include your full peer review and any attached files.). If published, this will include your full peer review and any attached files. If you choose “no”, your identity will remain anonymous but your review may still be made public. Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our For information about this choice, including consent withdrawal, please see our Privacy Policy .--> Reviewer #1: No Reviewer #2: No ******** https://doi.org/10.1371/journal.pone.0331081.r003
Formally Accepted
Acceptance Letter - Xiaohui Zhang, Editor PONE-D-25-42787R1 PLOS One Dear Dr. Kabir, I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS One. Congratulations! Your manuscript is now being handed over to our production team. At this stage, our production department will prepare your paper for publication. This includes ensuring the following: * All references, tables, and figures are properly cited * All relevant supporting information is included in the manuscript submission, * There are no issues that prevent the paper from being properly typeset You will receive further instructions from the production team, including instructions on how to review your proof when it is ready. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few days to review your paper and let you know the next and final steps. Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org. You will receive an invoice from PLOS for your publication fee after your manuscript has reached the completed accept phase. If you receive an email requesting payment before acceptance or for any other service, this may be a phishing scheme. Learn how to identify phishing emails and protect your accounts at https://explore.plos.org/phishing. If we can help with anything else, please email us at customercare@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Xiaohui Zhang Academic Editor PLOS One https://doi.org/10.1371/journal.pone.0331081.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .