Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The study evaluates the efficacy of two methods for measuring chronic wound areas—one using a 2D calibrator and the other augmented reality (AR). These methods are compared with traditional techniques and a digital planimetry gold standard. Conducted at a single center with 41 patients, the investigation involved capturing 67 wound images using mobile devices. The results demonstrated a high correlation with the gold standard, indicating substantial accuracy and reliability. The research confirmed the repeatability and reproducibility of the software, validating its safety and usability for clinical practice.

 

However, I have the following concerns,

 

(1)First, since this is a pre-marketing clinical test to meet the manufacturer's expected use, I recommend double-checking for any potential conflicts of interest. If applicable, the authors should disclose the manufacturer's identity. Additionally, as some non-published material is not written in English, I suggest translating these materials if they contain relevant information.

(2) I suggest the authors provide more details about the Machine Learning methods used. Specifically, it is unclear which model was implemented, such as Mask R-CNN or Segment Anything. More information on the model is necessary to better estimate the accuracy and understand the performance of the proposed methods.

(3)Could the author explain why most points in Figure 11(b) form a straight line? Is a linear relationship expected in Bland-Altman plots? Why don't the other figures show this linear relationship? In lines 319 to 320, the author mentions "larger measurements in larger lesions (increasing and linear pattern)," but if this is the case, how do we explain the two outlier points? Were there any errors or inconsistencies in the operations for these two points? 

(4)In line 122, the author writes, "in square centimeters (cm2)." This should be written as cm^2, with the 2 as a superscript.

(5) Figures should be centered, and Figure 2 is too small to see the details clearly.

Comments on the Quality of English Language

English looks good to me.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript discusses two novel methods developed using Machine Learning techniques to measure chronic wound area. The following are my comments and recommendations:

In general:

- According to the authors, they evaluated chronic wound area measurement with Machine Learning techniques, but they don't talk about these techniques specifically. What type of Machine Learning is used for analysing and processing the images? Digital measurement of wound images with devices/applications has already been used to measure the wound area. How is your work different from those already published?

Introduction:

- The Introduction Section is concise, it doesn't clarify what is already known, what is not known, and why this study exists.

- Lines 25 and 30:  Reference [3] has been cited twice in two sentences in a row. Please, remove the first one (line 25).

- Figure 1: On page 3 of the informed consent, the authors say: (La enfermera medirá la longitud y la anchura de su lesión cutánea con una regla. La regla no entrará en contacto con su piel) ''The nurse will measure the length and width of your skin lesion with a ruler. The ruler will not come into contact with your skin'', but it can be seen in the figure that the ruler is in contact with the skin. 

- Lines 99-102: Please explain the segmentation methods and Machine Learning tools applied to the images in more detail.

- Line 151-152: the authors say ''the measurements of the wound area were performed by different trained nurses''. Generally speaking, the same person must make the measurements to ensure the reliability of wound measurements and provide reliable and valid results. Only in this way can human error be minimized. Have the authors taken this error into account?

- Lines 153-154: Please explain in more detail the meaning of photo A, photo B, and photo C.

Results Section:

- Table 1: Explain in the table's title the abbreviation (DE). Please, instead of DE put SD (Standard Deviation). Moreover, represent the standard deviation's value as "±11.65"

- Line 284: There is an error here, it is not Figure 2 but Table 2.

Discussion Section:

- In the Discussion Section, the authors must interpret the significance of their findings regarding what was already known about the research concern being investigated. The discussion must explain the obtained results, and if possible, compare the results with the findings from other studies.

References: 

[2] Please correct the reference: González de la Torre, H.; Soldevilla-Ágreda, J.; Verdú-Soriano, J. Clinical Practice: The status of wound care units in Spain. Wounds International 2017, 8(4): 28-33

[26] The doi does not correspond to this document. Please remove it.

Comments on the Quality of English Language

Minor editing of some grammatical mistakes is required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Revision looks good to me and just one small suggestion, equation (1) is written as "Area = Width Height 0.785" and please add a multiple sign into it as "Area = Width * Height * 0.785" for clarity.

Comments on the Quality of English Language

English looks good to me.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have adequately addressed the reviewer's comments thoroughly and satisfactorily. However, I have a couple of comments to make.

- Line 74-75: Please correct the 'typing error', instead of Mohhamed the correct name is Mohammed.

- Line 105: What does the number 1 at the end of the phrase ''... adjusting by Kundin's constant (0.785) 1.'' mean? Maybe Eq.1.

- In subsection (3.4. Repeatability Analysis), why have the authors modified the values ​​in Table 3 (intraclass correlation coefficients used to determine the repeatability of research software )? The values ​​in the previous version were the following:

3D software method iPhone photograph A vs 3D software method iPhone photo B 0.946 (0.913, 0.967), but the authors have modified them to 0.996 (0.993, 0.997). Why?

The same happens with:

2D software method iPhone photograph A vs 2D software method iPhone photo B, the values were changed from 0.971 (0.952, 0.982) to 0.996 (0.993, 0.997).

- Lines 474-476: The authors have added the following comment in the revised version:

"This paper was presented as a poster at the 6th International Conference on Advances in Signal Processing and Artificial Intelligence (ASPAI’ 2024), held on April 17-19, 2024, in Funchal, Madeira Island, Portugal [35]."

Ref. [35]: Casanova, L.; Reifs, D.; Reig, R.; Grau, S. Assessing Chronic Wound Area Measurement with Machine Learning Techniques in a Single Center, Non-randomized Controlled Clinical Trial. In Advances in Signal Processing and Artificial Intelligence: Proceedings of the 6th International Conference on Advances in Signal processing and Artificial Intelligence (ASPAI’ 2024); Sergey Y. Yurish., Ed.; IFSA, 2024; pp. 121–123

From my point of view, a published conference abstract and an article can not have the same title. My recommendation is to change the title of this paper since a different title would ensure that readers don't get confused between the journal paper and the published conference abstract.  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf