Multifractal Analysis of Choroidal SDOCT Images in the Detection of Retinitis Pigmentosa
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsAuthors have described a Multifractal analysis of choroidal SDOCT images in the detection of retinitis pigmentosa in human. They used a generalized Renyi point-centered dimensions to measure of inhomogeneity of OCT data between patients with RP and the control group. Its a simple but interesting paper. However, there are many unfactual statements in the manuscript and need a major revision to be considered for acceptance.
1. Line 60: "The retina appears as..." . this is a not a true . Please correct it.
2. Line 80-81- Need references
3.Line 85-87- Need to add for human as well as for animal eye imaging. use references . Authors may use better ref like:
https://doi.org/10.1109/JSTQE.2016.2556226
4. Line 96: "As a disadvantage ..." -Incorrect sentence
5. Line 126- incorrect sentence- SD-OCT is not an extension of OCT.
6. Authors need to mention about the two emerging optical modality for detection of RP disease, which has already proven in animal models such as directional optical coherence tomography (dOCT) and multimodal retinal imaging reported.
https://doi.org/10.1038/s41598-021-95320-z
7. Use references for all those equations copied from other sources.
8. line-183; What is the base of the logarithm used?
Author Response
Below, the line numbers on the reviewer's comments are of the initial manuscript and on responses, the line numbers and reference numbers are of the improved manuscript.
- Line 60: "The retina appears as..." . this is a not a true . Please correct it.
Response: Done (please see line 60).
- Line 80-81- Need references
Response: We have added the reference 10.
3. Line 85-87- Need to add for human as well as for animal eye imaging. use references . Authors may use better ref like:
https://doi.org/10.1109/JSTQE.2016.2556226
Response: In this paper, we have not dealt with the SS OCT technique nor have we studied animal eye imaging so we only give some references in lines 100, 134, 145, 147.
- Line 96: "As a disadvantage ..." -Incorrect sentence
Response: We have corrected it (please see line 104)
- Line 126- incorrect sentence- SD-OCT is not an extension of OCT.
Response: We have corrected It (please see lines 131-133)
- Authors need to mention about the two emerging optical modality for detection of RP disease, which has already proven in animal models such as directional optical coherence tomography (dOCT) and multimodal retinal imaging reported.
https://doi.org/10.1038/s41598-021-95320-z
Response: We have inserted lines 148-151 and references 17-19.
- Use references for all those equations copied from other sources.
Response: We have inserted reference numbers in lines 184, 205, 211, 231 and 242.
- line-183; What is the base of the logarithm used?
Response: It can be in any base (please see line 198).
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper applies a fractal study on retinal images retrieved with an OCT device. The main idea could be interesting; however, some sections of the paper do not provide relevant information for the study.
1. In the text, it is mentioned that the retinal blood vessel’s shape is fractal. An inexpert reader needs to see an image to probe this statement. Besides, is the observed fractal behavior coming from the retina’s axial or surface view?
2. To cite a reference, instead of: “In [7] the authors …” use, In reference 7, the authors… or Yu et al. [7] summarize…
3. The first bullets of page 3 are confusing. Please reorganize this section (an image could help).
4. line 117, page 3, “levels of gray color…” If they are gray levels, they do not have color; commonly, they cover from black to white.
5. The term DF is not defined in line 119, page 3.
6. In general, the introduction moves from OCTA to SDOCT abruptly. Please enhance the narrative of this section, or it will be hard for a reader to follow.
7. Figure 1: what is the purpose of the red lines?
8. Not sure, but if section 2 describes a similar model to the one applied to bone with fractal, it may reduce this section.
9. Section 4, page 8. The authors are using images already published by a different group. However, no further information about the OCT specs that generate those images exists. A typical noise in OCT is the speckle that appears in these images. How does it affect the fractal calculation?
10. As the paper is about an application, the reviewer suggests including selected post-processed retina images for both groups. This is important to observe the BW result with the noise.
11. Labels for the control and affected groups are weird; please simplify them.
12. Tables are not numbered; please add them numbers. The figure of the fractal dimension in bars is presented for each eye. It does not make sense; a combination will be more helpful. Otherwise, the table already has this information for each eye.
13. Page 9. The comment about how RP affects one eye more than the other is preliminary, as the authors use only three samples. A higher number of samples is required to reach such a conclusion.
14, Figures with no number on page 10 (they need it) show the lack of samples. It is hard to conclude something with so few samples. The figures’ tendency is not conclusive.
15. In general, a major concern is with the number of samples analyzed with the proposed method. They are not statistically significant, but if the analysis is valid in a small number of samples, it should be mentioned in the text.
Comments on the Quality of English Language
Small details
Author Response
- In the text, it is mentioned that the retinal blood vessel’s shape is fractal. An inexpert reader needs to see an image to probe this statement. Besides, is the observed fractal behavior coming from the retina’s axial or surface view?
Response: Done (please see added lines 67-70, figure 1 and reference 6).
- To cite a reference, instead of: “In [7] the authors …” use, In reference 7, the authors… or Yu et al. [7] summarize…
Response: Done (please notice the difference throughout the paper).
- The first bullets of page 3 are confusing. Please reorganize this section (an image could help).
Response: Done (please see lines 110-130 and figure 2.
- line 117, page 3, “levels of gray color…” If they are gray levels, they do not have color; commonly, they cover from black to white.
Response: Please see lines 118-130 and figure 2.
- The term DF is not defined in line 119, page 3.
Response: It is corrected to FD (Fractal Dimension). Please see line 120.
- In general, the introduction moves from OCTA to SDOCT abruptly. Please enhance the narrative of this section, or it will be hard for a reader to follow.
Response: Done. Please see lines 131-134.
- Figure 1: what is the purpose of the red lines?
Response: The red lines indicate where the machine scans the vertical sections, since it cuts the 3D organ simultaneously in the horizontal and vertical directions. We have omitted images of vertical sections because they do not show fractal behavior.
- Not sure, but if section 2 describes a similar model to the one applied to bone with fractal, it may reduce this section.
Response: We have explicitly stated that it can be omitted on a first reading. Please see lines 166-170 also.
- Section 4, page 8. The authors are using images already published by a different group. However, no further information about the OCT specs that generate those images exists. A typical noise in OCT is the speckle that appears in these images. How does it affect the fractal calculation?
Response: Please see lines 251-254 and lines 284, 285. A new related reference has been added also (ref. 28).
- As the paper is about an application, the reviewer suggests including selected post-processed retina images for both groups. This is important to observe the BW result with the noise.
Response: We have added Fig. 4 and 5.
- Labels for the control and affected groups are weird; please simplify them.
Response: Done. Now we have Subject 1, Subject 2 for the healthy subjects and Patient 1, Patient 2, Patient 3 for the patients (please see lines 287-309, tables 1, 2 and figures 6, 7, 8)
- Tables are not numbered; please add them numbers. The figure of the fractal dimension in bars is presented for each eye. It does not make sense; a combination will be more helpful. Otherwise, the table already has this information for each eye.
Response: Done (please see the whole paper and in particular figure 6).
- Page 9. The comment about how RP affects one eye more than the other is preliminary, as the authors use only three samples. A higher number of samples is required to reach such a conclusion.
Response: Corrected (please see lines 328-330).
14. Figures with no number on page 10 (they need it) show the lack of samples. It is hard to conclude something with so few samples. The figures’ tendency is not conclusive.
Response: Done (please see Fig. 7, 8)
- In general, a major concern is with the number of samples analyzed with the proposed method. They are not statistically significant, but if the analysis is valid in a small number of samples, it should be mentioned in the text.
Response: Please see lines 379, 380.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsAuthors have addressed all the raised concerns adequately. Revised Manuscript could be accepted now.
