PONE-D-25-53041AI-assisted quantification of dopaminergic neurons in Parkinson’s disease modelsPLOS ONE

Dear Dr. Sawamura,

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. In particular, we encourage you to strengthen the analysis to ensure the necessary rigour and robustness, which will help to substantiate the conclusions drawn and enhance the overall credibility of the study.

Please submit your revised manuscript by Dec 19 2025 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.

Please include the following items when submitting your revised manuscript:

• A rebuttal 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 https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Míriam R. García

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. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

3. 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.]

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

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?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript addresses an important issue in Parkinson’s disease research: the reliable quantification of dopaminergic neurons in the substantia nigra pars compacta. The authors demonstrate that a commercially available CNN-based artificial intelligence system can quantify dopaminergic neurons with comparable accuracy to manual counting. Notably, the validation was carried out not only in mouse models but also in marmoset PD models and human PD brain samples, which is a significant strength and adds translational value to the work.

While the concept is interesting, several methodological and interpretational concerns should be clarified before the manuscript can be considered for publication.

1．Training images appear to include repeated use of the same specimens. In Supplementary Figures, it appears that for mouse and human samples, the same images (e.g., =3 or N=4) were used multiple times as teacher images. Could the authors clarify why the same specimens appear repeatedly? Clarifying these points is important because repetitive use of identical or highly similar images in a small dataset can lead to model overfitting, and therefore the reproducibility of the AI-based quantification method may be overestimated. In addition, how many individual dopaminergic neurons were used as ground-truth labels for training?

2. Because TH-positive neuronal density can vary considerably along the rostro-caudal axis of the SNc, accurate quantification of the total number of dopaminergic neurons in the SNc generally requires unbiased stereological methods, or at least a comparison with the contralateral (non-lesioned) side to correct for section-to-section variability. In this study, it is not clearly described how the authors dealt with this issue.

3. The authors state that AI-based counting was affected by staining quality, particularly in samples with weak TH staining. However, it is unclear whether such low-intensity images were included in the training set. Were teacher images selected to include a wide range of staining intensities, including weakly stained sections? Clarifying whether the training dataset captured realistic staining variability is critical for interpreting the robustness of the AI model in routine histopathological settings.

4. In the analysis of human specimens, neither AI-based nor manual counting detected a significant reduction in dopaminergic neurons in PD patients compared with healthy controls. Given that substantial neuronal loss in the SNc is a hallmark of PD pathology, this result requires further explanation. What is the authors’ interpretation of the absence of statistical significance?

Could this be due to sampling bias related to the anatomical level of the analyzed sections? As mentioned earlier, SNc neuron density varies across rostro–caudal levels. Given that the AI and manual methods both failed to show a difference, the issue may not be the counting algorithm itself, but rather the criteria used for section selection.

5. The authors cite three prior studies that applied AI approaches to quantify dopaminergic neurons. However, to my knowledge, there is at least one additional report describing similar deep-learning–based approaches for TH-positive neuron detection (Chen et al. A user-friendly deep learning approach for robust dopaminergic neuron detection. Neurosci Lett. 2024, 836:137871. doi:10.1016/j.neulet.2024.137871.). The manuscript states that the present method performs with comparable inference accuracy, but it is not clearly discussed how this method differs from, or improves upon, those existing approaches.　What specific advantages does the proposed method (TruAI) offer over previously published models?

Reviewer #2: The authors present an interesting manuscript that highlights the use of the commercial AI-assisted cell counting software, TruAI, developed by Olympus, in dopaminergic neuron counting. The study investigates the utility of this software across three use cases—Parkinson’s disease (PD) animal models with PFF injections, marmoset models with PFF injections, and human PD samples. While the findings provide valuable insights for the broader scientific community, the study would benefit from further rigor, expanded analysis, and coherence to strengthen its conclusions. Below are specific points for consideration:

1. Comments on the Mouse PFF Model:

PFF Injection Duration:

The manuscript does not clearly state the duration of PFF injections. This information is critical to understanding the timeline for induced pathology and assessing the reliability of the reported tyrosine hydroxylase (TH) loss. Please provide explicit details about the injection schedule and post-injection timepoints.

Phospho-synuclein Staining:

To confirm the presence of synuclein pathology caused by the PFF injections, phospho-synuclein staining should be performed and shown in the results. Without such confirmation, it is difficult to ascertain that the observed changes in TH cells are the result of α-synuclein pathology.

Sampling and Section Analysis:

The authors use 8 μm thick tissue sections and analyze only one section per mouse (10 sections from 10 mice). This limited sampling is insufficient for assessing the substantia nigra (SN), where TH+ neurons may vary greatly from section to section due to technical artifacts and the topology of the SN.

Although the authors mention that they captured images from many additional sections, they did not analyze these. This represents a missed opportunity to provide a more comprehensive analysis.

To justify claiming TH loss in the PFF mouse model, the authors need to evaluate the entire SN or at least perform a more systematic analysis across different rostro-caudal levels. As it stands, the authors can only claim to have counted TH cells in a limited number of sections.

Discrepancies in Sample Details:

There are inconsistencies between the sample details in the Methods section and the data presented in figures. For instance, the Methods state “10 mice (4 PBS and 6 PFFs), 5 marmosets, and 6 human (1 PD and 5 controls),” but figures show a higher number of data points than described. Please clarify and align the information throughout the manuscript.

2. Technical Questions on the Software and Methodology:

Effect of Signal Contrast:

The authors state that signal contrast impacts the accuracy of AI-based detection. Please provide specific details on:

How signal contrast was manipulated.

The magnitude of changes in contrast that led to failure in detection.

The percentage of counting errors introduced by differences in contrast.

Experience of the Testers:

Please elaborate on the qualifications and experience of the “experienced testers” who were responsible for validating the AI-based counts. Information about their expertise in histology, microscopy, or AI-assisted analysis will help gauge the reliability of their findings.

Segmentation of the Substantia Nigra (SNpc):

The images provided show a diffuse segmentation of the SNpc by the AI algorithm. How was the manual segmentation of the SNpc aligned with the AI algorithm’s segmentation? If there were discrepancies between the two methods, how were these addressed, and how do they impact the reported cell counts?

Handling of Overlapping/Dense TH+ Cells:

The TH counting algorithm seems to rely on the nuclei of individual TH cells for detection. How does the software resolve overlapping or dense TH+ cells, which are common in the SNpc and often introduce challenges in automated analysis? The representative images provided do not appear to depict such scenarios. Please discuss and provide evidence that TruAI can handle these situations, or clarify the limitations of the software in this regard.

3. Clarifications on Results and References:

Line 166:

The statement regarding a “38% decrease in 2 months” is unclear. Does this pertain to 2 months post-PFF injection, or is it referring to 2-month-old animals? Please clarify and provide appropriate citations to support this claim.

Additional Literature Review:

The literature review could be expanded to provide a more comprehensive discussion of related studies on dopaminergic neuron analysis and AI segmentation. Papers by Haghighi et al. and Barzekar et al. (2023) may provide useful comparisons and insights. Consider incorporating relevant findings from these studies into Table 3 and elsewhere in the discussion.

Training Dataset and Parameters:

Since the study aims to showcase the application of TruAI for dopaminergic neuron counting, it would be valuable to provide the training dataset and parameters used by the algorithm. This transparency will enable the broader scientific community to replicate or adapt these methods for similar applications.

4. Limitations of the Study and Overstated Conclusions:

Marmoset Model Data:

The conclusions drawn for the marmoset model appear to be accurate and appropriately supported by the data. However, the analyses of mouse and human samples do not align with the same level of rigor and should be revised accordingly. Please ensure that conclusions are aligned with the data presented.

TH Cell Counting vs. TH Loss:

The data in the manuscript convincingly demonstrate the ability of TruAI to count TH cells in different species. However, the authors’ claim that they have demonstrated the ability of the software to examine TH cell loss in the PD samples (animal models and human tissues) is not sufficiently supported. This conclusion requires additional evidence and analysis.

Human PD Sample Size:

The study analyzes only one human PD sample as per methods. With an n=1, it is not statistically sound to make broader claims about the accuracy of the AI’s detection of TH cell loss in human PD tissues. Please explain how statistical analysis (if any) was performed and provide cautionary notes about these limitations in the manuscript.

Data Normalization for Statistical Analysis:

Were the data normalized prior to statistical analysis? If so, describe the normalization process and how the data were represented in the analysis. This information is important for reproducibility and transparency.

Null Hypothesis and AI-Based Analysis:

For the AI-based analyses, what was the null hypothesis? Please explicitly discuss the hypothesis tested and the conclusions drawn based on the results.

The title can be improved and a more suitable title would be something like “Evaluation of a Commercial AI-Assisted Cell Counting Software TruAI for Dopaminergic Neuron Counting across species”

In summary, while the study demonstrates potential applications of TruAI, improvements in methodology, additional analyses, and consistency in reporting are necessary to fully support the conclusions. This will enhance the impact and utility of the work for the scientific community

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). 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 Privacy Policy.

Reviewer #1: Yes:Yasuhiko Izumi

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.

PONE-D-25-53041R1Evaluation of a Commercial AI-Assisted Cell Counting Software for Dopaminergic Neurons across SpeciesPLOS One

Dear Dr. Sawamura,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we believe that it has merit, although it does not yet fully meet PLOS ONE’s publication criteria in its current form. We therefore invite you to submit a revised version of the manuscript that addresses the minor points raised during the review process.

Please submit your revised manuscript by Apr 02 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.

Please include the following items when submitting your revised manuscript:

• 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 https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Míriam R. García

Academic Editor

PLOS One

Journal Requirements:

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.

Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

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 #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: Thank you for thoroughly addressing all the comments. I understand that it is painful to address the reviewers comments but I hope this will make the manuscript stronger and will receive credits in the long run.

I have some very minor comments to better understand,

1.Were reference images also incorporated in counting the cells, if not, why was that route chosen? Better to have counts from all images considering there is limited images in the study which has been highlighted.

2. Under "AI-based counting of TH-positive cells in mouse, marmoset, and human samples" it still says 6 human samples but "Human Samples" in methods says 9. Please cross check.

3. I didn't see adding Cellpose model as a reference and addressing why this is better than that considering it is the state of art model for cell counting. Additionally, can the authors please add how the model can be made better, for e.g incorporating automated segmentation models for SN as has been shown in some recent manuscripts. That might provide the company to make their workflow and software better for future users.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). 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 Privacy Policy.

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.

Evaluation of a Commercial AI-Assisted Cell Counting Software for Dopaminergic Neurons across Species

PONE-D-25-53041R2

Dear Dr. Sawamura,

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 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,

Míriam R. García

Academic Editor

PLOS One