How Can Radiomics Help the Clinical Management of Patients with Acute Ischemic Stroke?
Round 1
Reviewer 1 Report
The manuscript acknowledges challenges in interpreting radiomics studies and emphasizes the importance of prospective studies to validate the diagnostic value of radiomics. It also discusses the potential impact of factors like thrombectomy techniques, clot composition, and imaging heterogeneity on radiomics analysis. The authors discuss how radiomics can be applied to different types of medical imaging, such as MRI and CT, for the management of AIS patients. The authors emphasize that radiomics does not require alternative diagnostic procedures and can be used in conjunction with standard imaging techniques. Radiomics is proposed to offer a wide range of possibilities, including predicting clinical status, identifying ischemic lesions, predicting the outcome of reperfusion techniques, and differentiating complications of AIS. Radiomics is suggested to play a role in supporting clinical criteria for diagnosing AIS and aiding in decision-making. The manuscript also discusses the evolving role of segmentation software and the potential standardization of feature extraction methods in radiomics. The authors express optimism about the future of radiomics, indicating that it could potentially change the paradigm of diagnostic imaging. However, they emphasize the need for further research, particularly prospective studies, to demonstrate the specific diagnostic value that radiomic analysis adds to medical imaging. The present study is valuable and carefully written, and it seems that it can be published with few changes that will be made.
The main problem with the introduction and problem statement is that it lacks a clear and concise statement of the actual problem that the research aims to address and its length and excessive level of detail. The work does not provide a clearly defined aim nor research questions that it aims to answer. The authors state "This review aims to summarize published articles on the performance of radiomics in patients with AIS ..." which is far too unspecific. The authors should clearly define the research questions that are answered in this review.
The introduction length and excessive level of detail are the other problems. While providing comprehensive information is important, this section seems to go into overly specific details about various studies and their results, which can make it overwhelming for readers and may detract from the main focus of the paper. Also, readers may struggle to understand the overall purpose of the study and how the various studies mentioned contribute to that purpose.
There are several other review papers with very similar topics. Such as: “MRI Radiomics and Predictive Models in Assessing Ischemic Stroke Outcome—A Systematic Review” and “Radiomics-based infarct features on CT predict hemorrhagic transformation in patients with acute ischemic stroke”, So what is the need to write this review article?
The accuracy of radiomics in stroke diagnosis is almost constant, what is the reason for this review? What factors cause the importance and necessity of review in this field?
In Table 1, the type of imaging modality is mentioned under the title of technique. This title causes a misunderstanding. Please justify it.
Author Response
Please see the attachment. Thank you very much.
Author Response File: Author Response.pdf
Reviewer 2 Report
In this review the authors describe the current literature on studies on the use of radiomics in the field of cerebrovascular disease. They explore the potential applications of radiomics in patients with acute ischemic stroke and theorise how radiomics may change the paradigm for these patients in the future.
The paper is mostly well written and conclusive in its content. However, there are few concerns on this work that need to be clarified.
First of all, as you state, these are all retrospective studies , most of them with small numbers of patients and a high heterogenity. It is difficult (or maybe too early) to draw a meaningful conclusion here. Please provide a statement.
It is not easy to read, as many small studies are listed one after the other. Please skip a few and concentrate on the important ones.
Could you provide a sentence on costs?
The differentiation of hemorrhage from intracranial contrst extracasation after MT is very important and has an immediate therapeutic impact. What is the advantage of radiomics compared to the most widely used dual energy CT?
minor editing of english language required
Author Response
Please see the attachment. Thank you very much.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors do a review of more than 40 papers which describe use of machine learning. Significant results are extracted and presented in text and in a table. The review is carefully written and the lot of work put into it is apparent.
------------
line 16/17/18 should be expressed in more sober words.
line 41 diffusion MRI is what came after perfusion CT, machine learning/radiomics was late
line 43 unprecedented boom - hyperbole should be avoided
line 54 ... something is missing
line 59 which articles?
line 64 ... something is missing
line 67 there are articles, please reference them
last paragraph of intro, good, but please avoid the forefront thing
2. Radiomics - line 76 to 116 the review does start with a non-review, this should be explained and referenced individually.
line 116 just a subheading and a flow chart without explanation. Readers should be given more.
line 171 ff They they they - please revise. This paragraph has a lot of AUC numbers. This is not attractive and some kind of discrimination of papers should be tried. A few sentences at the end describing if there is a development in these studies; if more progress may be expected - this would help.
Same holds true till 2.9
line 336 burst into our lives, more sober language please
line 344 creates diagnostic advance
368 lack of prospective studies. This is true, no doubt, but are there any at all?
399 future is bright, be sober please
Table of references should be structure according to subheadings.
---------
In conclusion the review is written with care and diligence; lots of references were read. The paper could be much improved if the authors dared to classify those papers and try to identify lines of research. As it is it is a careful collection of materials. Readers might like opinions on these papers somewhere.
Author Response
Please see the attachment. Thank you very much.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
The authors did what they could and the paper is improved.