-
Notifications
You must be signed in to change notification settings - Fork 530
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Paper: PredxTools: Dispelling the Mystery in Histopathological Image Processing #922
Conversation
Curvenote Preview
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Here are just a couple MyST-related suggestions. I think they make the final build slightly tidier, but feel free to ignore if you don't like them!
Sent review reminders to @KrishnaRekapalli and @arushimisra99 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @bkf15 , my name is Hongsup Shin, the editor of your paper. Unfortunately, we have had hard time getting reviews from the assigned reviews, so I and another SciPy proceedings co-chair will be stepping in as reviewers of your paper. I already left some comments in the manuscript for clarification.
Here are also my general comments:
- Given that the SciPy conference highlights creative and novel uses of various Python libraries, and your proposed tool is built on various existing packages, I would appreciate if you can add more details on how the packages cited in your paper (like scikit-learn, for instance) were used to build your tool.
- I think the title of the paper is a bit exaggerating what the tool does. It's hard to find what exactly the "mystery" is. After reading the manuscript, I think the novelty here is that the proposed tool does multiple things altogether that the previous tools didn't. So please consider changing it to be more specific and accurate.
- Even though the summary table helps comparing your tool and the existing tools, while reading the manuscript, it was difficult to follow exactly how your tool is different from those. If you can add more detailed comparison, I think that would really help especially because the current manuscript very much reads like a manual of your tool.
- I am also wondering whether you have any data on how much efficiency your tool provides. For instance, your tool has some automation steps and how many man-hours can it save compared to manual approach (and also compared to other tools)? Is your tool faster than existing tools? If yes, then by how much?
- You mention challenges in dealing with "unseen data" but I wasn't able to find any ML elements. Does your tool alleviate these specific challenges with unseen data?
@bkf15 By the way, even though the Open Review Period ends tomorrow (Aug 7), since this manuscript didn't have any reviews so far, you can continue making changes to the manuscript. We will provide one more review by the end of this week. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi @bkf15, I am the other SciPy proceedings co-chair stepping in to provide a review for your paper. Thanks for your patience in waiting for your submission to be reviewed. I do not work in the field of histopathological image processing so my comments are mainly around making the paper more accessible to a reader from another domain.
papers/brian_falkenstein/main.md
Outdated
# Ensure that this title is the same as the one in `myst.yml` | ||
title: "Predx-Tools: Dispelling the Mystery in Histopathological Image Processing" | ||
abstract: | | ||
Histopathological image contains insights into disease state, but applying automatic image analysis is challenging. Often, human intervention in the form of manual annotation or quality control (QC) is required. Additionally, the data itself vary considerably in available features, size, and shape. Thus, a streamlined and interactive approach is a necessary part of any digital pathology pipeline. We present PredX-Tools, a suite of simple and easy to use python GUI applications which facilitate all parts of a digital pathology pipeline, from image QC and labeling to visualization and analysis of results to allow for a deeper understanding of the data. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Because SciPy is an interdisciplinary conference, the abstract to this paper can be made more accessible to readers by defining terms such as "histopathological image", by clarifying who the intended user of the tool suite is, etc.
Additionally, here are a few specific comments on the text.
- Please revise the first sentence to make it grammatically correct.
- "data itself vary" -> either "data themselves vary" or "data itself varies" for plural or singular self-consistency
- In the last sentence I believe it could help a reader unfamiliar with "all parts of a digital pathology pipeline" if it were directly stated why "a suite of simple and easy to use python GUI applications" would "allow for a deeper understanding of the data".
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I suggests adding information about your intended user/audience. Who could benefit from this tool? Be sure to call them out clearly and early so you reach your intended reader.
papers/brian_falkenstein/main.md
Outdated
|
||
## Introduction | ||
|
||
Histopathological image analysis reveals insights into disease state and has potential to harness the immensely complex and heterogeneous tumor microenvironment to guide understanding of disease progression, response to treatment, and aid in patient stratification for clinical trials {cite:p}`doi:10.4132/jptm.2019.12.31, doi:10.3389/fonc.2022.889886, doi:10.1016/j.irbm.2019.06.001, doi:10.3390/cancers15194797, doi:10.1038/s41379-021-00919-2`. However, applying automatic image analysis routines continues to be challenging {cite:p}`doi:10.3390/cancers13123031, doi:10.1308/rcsann.2023.0091`. Often, human intervention in the form of manual annotation or quality control (QC) is required. Additionally, the data itself varies considerably in available features, size, and shape. Thus, a streamlined and interactive approach is a necessary part of any digital pathology pipeline {cite:p}`doi:10.1111/his.14356`. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you give more detail about or some examples of variations in "features, size, and shape"?
papers/brian_falkenstein/main.md
Outdated
# Ensure that this title is the same as the one in `myst.yml` | ||
title: "Predx-Tools: Dispelling the Mystery in Histopathological Image Processing" | ||
abstract: | | ||
Histopathological image contains insights into disease state, but applying automatic image analysis is challenging. Often, human intervention in the form of manual annotation or quality control (QC) is required. Additionally, the data itself vary considerably in available features, size, and shape. Thus, a streamlined and interactive approach is a necessary part of any digital pathology pipeline. We present PredX-Tools, a suite of simple and easy to use python GUI applications which facilitate all parts of a digital pathology pipeline, from image QC and labeling to visualization and analysis of results to allow for a deeper understanding of the data. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I suggests adding information about your intended user/audience. Who could benefit from this tool? Be sure to call them out clearly and early so you reach your intended reader.
papers/brian_falkenstein/main.md
Outdated
|
||
## Introduction | ||
|
||
Histopathological image analysis reveals insights into disease state and has potential to harness the immensely complex and heterogeneous tumor microenvironment to guide understanding of disease progression, response to treatment, and aid in patient stratification for clinical trials {cite:p}`doi:10.4132/jptm.2019.12.31, doi:10.3389/fonc.2022.889886, doi:10.1016/j.irbm.2019.06.001, doi:10.3390/cancers15194797, doi:10.1038/s41379-021-00919-2`. However, applying automatic image analysis routines continues to be challenging {cite:p}`doi:10.3390/cancers13123031, doi:10.1308/rcsann.2023.0091`. Often, human intervention in the form of manual annotation or quality control (QC) is required. Additionally, the data itself varies considerably in available features, size, and shape. Thus, a streamlined and interactive approach is a necessary part of any digital pathology pipeline {cite:p}`doi:10.1111/his.14356`. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I’m not medical, and I could use a few words about what pathology and specifically histopathology, actually are. I recommend breaking down the first sentence in the introduction. First introduce histopathology itself and be sure to tell us what the images are of! I had to look up that the images are of tissues, which in the past were probably viewed as slices on slides. Only then go on to say that histopathology is now often done with digital imaging instead of class slides and how that makes it amenable to computational analysis. Then you can go into the rest of your first paragraph, which is very nice.
Hi @bkf15 , just a friendly reminder that all initial reviews are in. I highly recommend you start responding to the comments soon since it'd take time for the reviewers to respond to the changes. Remember that the open review period ends on Sep 2, and you will not be able to make any changes to the manuscript after that point. If you have any questions, please let me know! |
Hello. Thank you for all the comments. I will be providing my thoughts and answering questions as well as pushing updates to the paper. Thank you for your patience. |
I apologize for the delay. I will be pushing my changes to the paper shortly. I hope I have addressed all questions and concerns. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Addressing reviewer comments.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@bkf15, thanks for the updates to make this paper more accessible to a wider audience. It looks great to me.
papers/brian_falkenstein/mybib.bib
Outdated
author = {{PySimpleGUI Team}}, | ||
title = {PySimpleGUI}, | ||
year = {2024}, | ||
url = {https://zarr.readthedocs.io/en/stable/}, |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this the intended URL for PySimpleGUI?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for bringing this to my attention. I just made the fix and pushed it. I'm not sure if it's too late. Apologies.
Thanks @mepa @nicole-brewer for reviewing the paper! |
If you are creating this PR in order to submit a draft of your paper, please name your PR with
Paper: <title>
. An editor will then add apaper
label and GitHub Actions will be run to check and build your paper.See the project readme for more information.
Editor: Hongsup Shin @hongsupshin
Reviewers: