Peer Review History
Original SubmissionSeptember 3, 2019 |
---|
PONE-D-19-24806 Myosoft: an automated muscle histology analysis tool using machine learning algorithm utilizing FIJI/ImageJ software. PLOS ONE Dear Dr. Choo, 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. We would appreciate receiving your revised manuscript by Nov 10 2019 11:59PM. When you are 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. If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols Please include the following items when submitting your revised manuscript:
Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. We look forward to receiving your revised manuscript. Kind regards, Thomas Abraham, PhD Academic Editor PLOS ONE Journal Requirements: When submitting your revision, we need you to address these additional requirements. 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 http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf [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: No Reviewer #2: Partly ********** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: Yes Reviewer #2: 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: No ********** 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 paper presents a semi-automated method for the morphometric and fiber type analysis in muscle sections stained with fluorescent antibodies estimating the size and type in histological muscle preparations is useful for quantifying key indicators of muscle function and for measuring responses to a variety of stimuli or stressors. The method is interesting since it might help clinicians who presently manually perform this analysis; however, this is a laborious, time consuming task, affected by inter and intra-personal variability. Results are compared to state of the art works; however, I think that several points must be modified/extended before accepting the article. section Introduction --------------------- The introduction is clear, the problem is well stated. However, I think some state of the art works are missing. More precisely, in lines 70-72, authors list some automatic segmentation techniques: "To offset this obstacle, several groups have developed software that automates analysis of muscle histology (16-19)." My advice is to shortly summarize the main steps of the cited methods (I believe they are the mostly related to this work). Moreover, in subsequent lines, authors mention that learning methods are the solution to overcome staining artifacts and they say (line 75): "Machine learning offers unprecedented potential in 76 resolving these present limitations in automated image analysis (20)." I would firstly recall some works where machine learning methods are used to solve the problem of histochemical image segmentation. As an example, please have a look (and I would cite) the following works, which exploit machine learning methods for stained image analysis. [1] Madabhushi A, Lee G. (2016). Image analysis and machine learning in digital pathology: Challenges and opportunities. Medical image Analysis, 33: p. 170-175. [2] Gurcan MN, Boucheron L, Can A, Madabhushi A, Rajpoot N, Yener B. (2009). Histopathological Image Analysis: A review. IEEE Rev Biomed Eng., 2: p. 147-171. [1] Casiraghi E, Cossa M, Huber V, Tozzi M, Rivoltini L, Villa A, et al. (2017) "MIAQuant, a novel system for automatic segmentation, measurement, and localization comparison of different biomarkers from serialized histological slices", Eur J Histochem. [2] Casiraghi E, Huber V, Frasca M, Cossa M, Tozzi M, Rivoltini L, et al. (2018). A novel computational method for automatic segmentation, quantification and comparative analysis of immunohistochemically labeled tissue sections. BMC BioInformatics. 19(Suppl 10): p. 357. Next I would insert citation (20) and recall the main features of Trainable Weka Segmentation. Why authors did choose this learning strategy, instead of others? section Image acquisition -------------------------- At the end of this section I would put a table, or anyway I would clarify: - how many sections where acquired, - the dimensions of the acquired sections - how many subimages where extracted from those sections - the dimension of the extracted subimages - how many subimages where used for developing (and training) the system - how many subimages where used for testing it Please put some images to show: - the whole section - some of the working subimages - some of the discarded subimages (discarded as explained in line 133-134: "If an image was randomly selected and had significant fluorescence artifacts or tissue damage, this area was either excluded, or a new image was chosen.") - some good stain quality subimages (stain ratio>5) - some poor stain quality subimages (stain ratio<=5). A small note on line 138: I believe poor stain are less OR EQUAL TO, isn't it? Otherwise, what about subimages with stain ratio = 5? section Four components of the Myosoft pipeline ------------------------------------------------ At first, I would insert Fig.1, the schematic drawing to recall the main step of the myosoft pipeline, at the beginning of the section and, obviously, I would start the section with lines 185-196. line 156: which contrast enhancement technique and which convolutional (filter) matrix are applied as preprocessing? The latest might be a gaussian/median filter to remove high band/salt-and-pepper, randomic noise... however please clarify. What is the iterative classifier mentioned in line 165? Neural network, SVM, convolutional neural network, kNN, bayesian tree,... I won't go in details with the remaining part of the algorithm described in this section, since the note is that each step must be better explained. All the image processing steps are just mentioned. Therefore, if a reader wants to reimplement it, he must understand the code, which means the paper is not clear at all. Briefly, I would suggest that authors start mentioning Fig.1 and listing the main steps of the algorithm. Next, for each step, I would insert a subsection describing in detail the processing and the algorithms used in the step. section Adjustable Parameters ------------------------------ Since the parameters are manually chosen by users, I would suggest that the author mention which procedure is used to choose the parameter values. section Fiber typing is determined by gating of MyHC fluorescent intensity distributions ----------------------------------------------------------------------------------------- Are the thresholds automatically chosen by automatic methods (e.g. Otsu or Arbib), or are them manually set by users? If they are manually set, what happens if any automatic thresholding or clustering method is used? sections on Results and Discussion ---------------------------------- These sections are well written and well presented, in my opinion, though I can't really understand the test sample size. How many sections where used for testing? There must be at least 50 sections to consider the work as convincing. Comparison to existing software would be more convincing if the number of test images and mean performance ratios were reported (e.g. with a table). Reviewer #2: In this study, the authors developed myosoft, an improved automated tool, for the analysis of muscle myofibers size and classification of fiber type populations in histological samples stained with fluorescent antibodies. Myosoft implements a machine learning approach, which appears to overcome the limitation of existing technologies that rely on the need of having a high quality staining to achieve accurate analysis. The overall work is exciting and the tool could be of great use for the biomedical field. However, considering the authors are introducing a new tool for general research use, information describing the requirements/steps for the use of this tool is necessary, as well as a fine-tuned approach for the detection of hybrid fibers. Specific comments: Given than the authors are introducing a new method for the histological analysis of muscle sections, a lot of details regarding the image requirements as well as the detailed steps for the use of the tool from beginning to end need to be provided. For instance: 1. There are very few details regarding the requirements for: a) image acquisition (i.e. exposure time), b) format of the images needed to be analyzed by the tool (TIFF, 8 bit, 16 bit, merged image, individual channels, etc), c) organization of the image-files for multiple sample analysis, d) any pre-treatment of the images (it was not clear whether the multiple supplementary txt files are run before the analysis), or e) the computer needs for running the program. 2. There was no tutorial with troubleshooting instructions attached as supplementary data in this manuscript or at the github site. The file attached as S2 included only a few lines of instructions with not sufficient details for a novice user. A tutorial with clear, detailed instructions of the use of the tool (preferably with images/screen shots), that includes a descriptive list of the output files obtained with an explanation of the information contained in each of them, is necessary for the correct use of the tool and an accurate interpretation of the results. Finally, it was not clear how the multiple files provided as supplementary materials should be used for the analysis of one sample (multiple samples) or for what purposes are these used. This point needs to be expanded and clarified. 3. For this tool to be pertinent to the user that requires fast information, the analysis should include an output file (or a log) that contains a summary of the following information: a) Number of cells/µm2 (or biopsy size [mm2] so this information can be calculated), b) Total number of cells c) Number of cells (and %) corresponding to each fiber type d) Average CSA of all cells (µm2) e) Average CSA of each fiber type population (µm2) 4. It would also be important to define what is considered a sub-optimal quality staining and define the limits at which the tool can still provide an accurate analysis. 5. The authors compared myosoft with several available tools that perform similar tasks as the developed method. However, at least 2 recent publications (2019) describing automated tools using Image/J (doi.org/10.1186/s13395-018-0186-6 and doi: 10.1186/s13395-019-0200-7) have been overlooked. To have a better perspective of myosoft capabilities as compared to more recent tools, I recommend these tools should at least be mentioned in the discussion part of the manuscript. 6. One of the main issues of current tools for automated detection of fiber populations in muscle sections is the detection of hybrid fibers. Biologically fibers can adapt and switch from one fiber population to another in response to different stimuli. While I understand how you propose to quantify I/IIa fibers, I am concern about the significance of finding I/IIb or IIa/IIb or I/IIa/IIb fibers. More than a biological phenomenon, the presence of these populations together is a reflection of staining issues that needs to be addressed. More importantly is to define the criteria to select one population vs the other (i.e. in the case of I/IIb are these cells really type I (slowest) or IIb (fastest)). Finally, hybrid fibers for IIa/IIx or IIx/IIb populations cannot be accounted for with the proposed staining protocol. Minor: 1. It would be a good reference to include the time that the tool takes to analyze a single sample (whole section) 2. Besides CSA, it is not described which information regarding fiber size (perimeter, diameters, etc) can be obtained with this program. 3. In the discussion part it is not clear whether myosoft can clearly identify and quantified central nuclei or if this is a possibility that has not been properly tested. Please expand 4. Line 97 - C57Bl /6 J or N?, also what is the number of mice used? 5. Line 274. Myovision 6. The overall quality of the images and tables in the figures, especially those in color need to be improved. ********** 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: 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 be viewed.] 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 us at figures@plos.org. Please note that Supporting Information files do not need this step. |
Revision 1 |
Myosoft: an automated muscle histology analysis tool using machine learning algorithm utilizing FIJI/ImageJ software. PONE-D-19-24806R1 Dear Dr. Choo, We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements. Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication. Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. 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 enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and 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. With kind regards, Thomas Abraham, PhD Academic Editor PLOS ONE Additional Editor Comments (optional): Reviewers' comments: |
Formally Accepted |
PONE-D-19-24806R1 Myosoft: an automated muscle histology analysis tool using machine learning algorithm utilizing FIJI/ImageJ software. Dear Dr. Choo: I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, 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. For any other questions or concerns, please email plosone@plos.org. Thank you for submitting your work to PLOS ONE. With kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Thomas Abraham Academic Editor PLOS ONE |
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 .