PONE-D-20-33334

Different cell imaging methods did not significantly improve immune cell image classification performance

PLOS ONE

Dear Dr. Takahashi,

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 26 2021 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: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Chi-Hua Chen, Ph.D.

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. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

3.

Thank you for stating the following in the Competing Interests: "NO authors have competing interests"

We note that one or more of the authors are employed by a commercial company:  Epistra Inc, Japan.

(1) Please provide an amended Funding Statement declaring this commercial affiliation, as well as a statement regarding the Role of Funders in your study. If the funding organization did not play a role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript and only provided financial support in the form of authors' salaries and/or research materials, please review your statements relating to the author contributions, and ensure you have specifically and accurately indicated the role(s) that these authors had in your study. You can update author roles in the Author Contributions section of the online submission form.

Please also include the following statement within your amended Funding Statement.

“The funder provided support in the form of salaries for authors [insert relevant initials], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ‘author contributions’ section.”

If your commercial affiliation did play a role in your study, please state and explain this role within your updated Funding Statement.

(2) Please also provide an updated Competing Interests Statement declaring this commercial affiliation along with any other relevant declarations relating to employment, consultancy, patents, products in development, or marketed products, etc.  

Within your Competing Interests Statement, please confirm that this commercial affiliation does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests) . If this adherence statement is not accurate and  there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include both an updated Funding Statement and Competing Interests Statement in your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

[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?

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: Partly

Reviewer #2: Yes

Reviewer #3: Partly

**********

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

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

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

Reviewer #3: 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

Reviewer #3: 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: I do not find enough merit to support the publication of the paper, as the contribution is not significant enough.

The authors do not compare the proposed approach with actual state of art. Several methods are proposed in the literature.

It is not clear the novelty of the method.

Are not clear the pros and cons of the proposed approach respect the actual state of the art.

Overall impression of study is just below average; the study could be useful and deals with significant problem, but the style way in which paper is written does not match the journal standards.

Reviewer #2: Different cell imaging methods did not significantly improve immune cell image

classification performance

Summary:

This paper studies the use of three microscopy imaging modalities on accuracy of image-based cell classification. The three imaging modalities are: differential interference contrast (DIC), phase contrast (Ph) and bright-field (BF). The cell types are lymphoid-primed multipotential progenitor (LMPP) and pro-B cells. The cell classification methods are custom-designed convolutional neural networks (CNNs) and support vector machine (SVM) using cell size and cell contour shape as inputs to SVM. The ground truth cell classification is established using Alexa Fluor 594 stain, fluorescent imaging, and simple image analysis.

The conclusions of this study are: (1) the accuracy of image-based cell classification is invariant to the three microscopy imaging modalities and their combinations, (2) the CNN classifiers outperform the SVM classifiers, and (3) the accuracy of image-based cell classification is invariant to a chosen focal position. The conclusions are supported by three replicates (three dishes) and multiple spatial fields of views (FOVs) and multiple time points (time lapse FOVs) in two experiments.

Major comments:

• The authors should include a clear description of experimental and observational factors and their levels in this experimental design study. Right now, the reader cannot easily derive such critical information to this study.

o For example, it is not clear whether the cells are going through mitosis during time lapse imaging and therefore the cell state is ignored as a factor.

o The focal position is listed as a factor. The levels of this factor are captured as z-stacks (± 3.0 �m range, 0.3-�m interval). First, the authors should explain how they samples the focal positions (e.g., frequently occurring deviation from a perfect focus, visually indistinguishable image appearance). Second, the claim about classification accuracy being invariant to focal position should be limited to the chosen range (otherwise it might not be true).

• The authors should explain why they used a custom designed CNN architecture as opposed to using a widely used U-Net CNN architecture.

o The authors could include a picture of the custom-designed CNN architecture.

o If the authors are not using the U-Net, then they could compare the CNN architectural designs (custom vs U-Net)

• The authors should include confidence intervals in their reported results.

o For example, “CNN showed the best classification performance with DIC images (AUC ~0.9).” on page 15. How do we know that the classification results obtained using DIC images are statistically any better than the results obtained by the other imaging modalities?

Minor comments:

• Page 8: “These regions were removed from the train/test data by removing regions ≤ 2000 pixels and ≥ 8000 pixels”. Can you provide physical dimensions accompanied by some apriori knowledge?

• Page 9: “…fluorescence images were divided by the intensity of the blurred Alx594-conjugated ..” Who defines which frame in a z-stack is blurred?

• Page 11: “Cell images were masked with a circle of radius 50 pixels …” Why are the cell images masked with a circle if you have a segmentation mask? I would also recommend clarify the terminology for raw cell images, mask images from segmentation (ground truth), and mask images created by inserted circles into background color.

• Page 16, line 258: “different numbers of cell images (214, 708, 2,144, 7,077 and 21,444 cells)” Please, use semicolons to separate the numbers if you are using commas to separate thousands. Include the numbers of cell images for CNN training into a list of factors explored.

• The authors claim that the data are available, but there is no URL pointing to the data.

Reviewer #3: In this work, the authors studied the effects of deep learning-based cell classification using images from several common bright field imaging methods. These imaging methods include differential interference contrast (DIC), phase contrast, and regular bright field. They used lymphoid-primed multipotential progenitor (LMPP) and pro-B cells as their model system. Their results showed that the performance of Deep learning classifier from different imaging methods performed similarly. Overall, this is an interesting study, and the results suggested that regular bright-field images without contrast optics (i.e., phase, DIC) can provide sufficient information for Deep learning. However, the authors only demonstrate this result in one biological model system (LMPP vs. proB) with one simple deep learning architecture. Hence, it is not clear if the reported findings represent a general theme or restricted to this very system that the authors demonstrated in the study. As it is well known that the performance of deep learning is highly associated with network architectures. Thus, it is not clear if or how different network architecture (such as ResNet, google net, mobile net …etc.) of deep learning can affect the classification performance in recognizing different types of biological specimens/cell samples under different imaging modalities. Overall, more thorough studies should be conducted to better support their claims.

**********

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: No

Reviewer #2: Yes: Peter Bajcsy

Reviewer #3: 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.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PONE-D-20-33334R1

Different cell imaging methods did not significantly improve immune cell image classification performance

PLOS ONE

Dear Dr. Takahashi,

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 Sep 05 2021 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: http://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,

Chi-Hua Chen, Ph.D.

Academic Editor

PLOS ONE

[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 #1: (No Response)

Reviewer #2: All comments have been addressed

Reviewer #3: (No Response)

**********

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 #1: No

Reviewer #2: Yes

Reviewer #3: No

**********

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

Reviewer #1: No

Reviewer #2: Yes

Reviewer #3: N/A

**********

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: No

Reviewer #3: 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 #1: Yes

Reviewer #2: Yes

Reviewer #3: 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 #1: I do not find enough merit to support the publication of the paper, as the contribution is not

significant enough.

your tests are not sufficient to validate your conclusions. It is not clear the novelty of the method.

Are not clear the pros and cons of the proposed approach respect the actual state of the art.

Overall impression of study is just below average; the study could be useful and deals with

significant problem, but the style way in which paper is written does not match the journal

standards.

Reviewer #2: Different cell imaging methods did not significantly improve immune cell image

classification performance

Minor comments:

• Line 94: “three glass-bottom dishes (Eppendorf, Germany). For” – remove the period

• Line 95: “Experiment 1 and 2, respectively.” – Table 1 includes description of Experiment 1. I cannot find anywhere information about Experiment 2. Can you clarify?

• Previous comment: “The authors claim that the data are available, but there is no URL pointing to the data.”

o Response: “The data will be made available from the corresponding author upon reasonable request. Since the size of the datasets is larger than 1 TB, it is difficult to put it on a public place with a URL.”

o Recommendation: add a sentence that summarizes the data size reaching 1TB as a multiplication of image pixel size x number of fields of view taken per dish x number of dishes x etc.

• Previous comment: “The authors should explain why they used a custom designed CNN architecture as opposed to using a widely used U-Net CNN architecture.

� The authors could include a picture of the custom-designed CNN architecture.

� If the authors are not using the U-Net, then they could compare the CNN architectural designs (custom vs U-Net)”

o Response:” Our CNN network is a plain CNN (no special connections such as skip connections and attention are added), and only changes made were on network parameters such as kernel sizes. Parameters used are given in the section ‘Training and testing of CNN and SVM’

(Main text lines 210-213).

As far as we understand, U-Net is a network for image transformation and is not often used for identification. Although the main focus of this study is to investigate the effect of different observation methods on the accuracy of identification, it is an interesting question whether our findings hold for other architectures (such as ResNet). We have added this point to Discussion (Main text lines 296-302).”

o Recommendation: add a sentence that relates “your definition of a plain CNN” (i.e., there is no published definition of a plain CNN) to a well-known published CNN architecture. For example, you should state how your CNN architecture is different from AlexNet or LeNet and why you made those changes.

Reviewer #3: The reviewer's comments were not directly addressed in the revised version of the manuscript. The main claim authors attempting to establish is that regular bright-field images without contrast optics (i.e., phase, DIC) can provide sufficient information for Deep learning. Yet, this was only demonstrated in one biological model system (LMPP vs. proB) with one simple deep learning architecture. Thus, I still don’t think the present data provide sufficient evidence to support the claim.

**********

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 #1: No

Reviewer #2: Yes: Peter Bajcsy

Reviewer #3: 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.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PONE-D-20-33334R2Different cell imaging methods did not significantly improve immune cell image classification performancePLOS ONE

Dear Dr. Takahashi,

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 Dec 23 2021 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,

Chi-Hua Chen, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

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 #1: All comments have been addressed

Reviewer #2: (No Response)

**********

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 #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: No

**********

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 #1: Yes

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 #1: Revision well done, the authors have addressed concerns of all the reviewers improving the submitted paper.

Reviewer #2: Minor comments:

From the rev 1 and rev 2:

+++++++++++++

Concern #2

Previous comment: “The authors claim that the data are available, but there is no URL

pointing to the data.”

o Response: “The data will be made available from the corresponding author upon

reasonable request. Since the size of the datasets is larger than 1 TB, it is difficult to

put it on a public place with a URL.”

o Recommendation: add a sentence that summarizes the data size reaching 1TB as a

multiplication of image pixel size x number of fields of view taken per dish x number of

dishes x etc.

A part of the dataset is now available at the following URL.

https://figshare.com/articles/dataset/Minimum_dataset_for_Different_cell_imaging_meth

ods_did_not_significantly_improve_immune_cell_image_classification_performance_/14

789811

1TB is the total size of the original images. The cropped images were used for

training. 0.6% of the original images are now available at the URL above. The cropped

images were excluded from the public data because they can be generated from the

original images using the method described in the paper.

+++++++++++++++++

The authors claimed that “A part of the dataset is now available at ..”. However, The URL just shows a list of file folders (dish1, dish3, dish2) and a list of file names. There is no way to download or view the images.

From rev1 and rev2:

+++++++++++++++++++

Concern #3

Previous comment: “The authors should explain why they used a custom designed

CNN architecture as opposed to using a widely used U-Net CNN architecture.

The authors could include a picture of the custom-designed CNN architecture.

If the authors are not using the U-Net, then they could compare the CNN architectural

designs (custom vs U-Net)”

o Response:” Our CNN network is a plain CNN (no special connections such as skip

connections and attention are added), and only changes made were on network

parameters such as kernel sizes. Parameters used are given in the section ‘Training

and testing of CNN and SVM’

(Main text lines 210-213).

As far as we understand, U-Net is a network for image transformation and is not often

used for identification. Although the main focus of this study is to investigate the effect

of different observation methods on the accuracy of identification, it is an interesting

question whether our findings hold for other architectures (such as ResNet). We have

added this point to Discussion (Main text lines 296-302).”

o Recommendation: add a sentence that relates “your definition of a plain CNN” (i.e.,

there is no published definition of a plain CNN) to a well-known published CNN

architecture. For example, you should state how your CNN architecture is different

from AlexNet or LeNet and why you made those changes.

We provided the information on the CNN architecture in Main text (lines 212-). As the

purpose of our work is not to propose a novel network architecture, we think that our

description contains enough information to reproduce our work. However, we added a

citation to the paper by Xu et al (2017), in which readers can find the exact architecture

ours is based on.

+++++++++++++++

I agree that your description contains enough information about the changes of the UNet architecture. My comment was about explaining why you had to make those changes instead of using the published UNet as is. This comment takes into account your description presented at the lines 211-219:

+++++++++++++++++++++++++++

211 The CNN architecture is based on Xu’s study [6]. The CNN had four convolution

212 layers (16 channel kernel size 5, 16 channel kernel size 5, 32 channel kernel size 5, 32 channel

213 kernel size 5, but when using images with ≥ 2 channels, the numbers of channels were all

214 doubled) and two fully connected layers (2,000 nodes, 2 nodes). ReLU (see Eq (3)) was used

215 as the non-linear activation function in all hidden layers. In Eq (3), x is the output of the prior

216 layer. The softmax function (see Eq (4)) was applied at the output of the neural network, and

217 here xi is the ith node of output layer x. A max-pooling layer with kernel size 3 and slide 2

218 was placed behind each convolution layer. A drop rate of 0.1 dropout layer was added in all

219 gaps in the layers.

+++++++++++++++++++++++

**********

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 #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.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Different cell imaging methods did not significantly improve immune cell image classification performance

PONE-D-20-33334R3

Dear Dr. Takahashi,

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 for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

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,

Chi-Hua Chen, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments: