Efficient Fuzzy Image Stretching for Automatic Ganglion Cyst Extraction Using Fuzzy C-Means Quantization

Round 1

Reviewer 1 Report

The author presents an algorithm for ganglion detection using fuzzy stretching and C-means-quantization, and compares the accuracy of ganglion detection with ART2. The algorithm is well explained, which is a good point. On the other hand, the clinical challenge of ganglion detection is the need to differentiate them from tumors and vascular masses. Therefore, machine learning needs to consider not only the detection of the mass but also the differentiation from similar diseases. If comparisons with these are being made, we think they need to be detailed in the article. Also, the presence of a daughter cyst (another small cyst) of a preliminary ganglion arising around the joint capsule must be noted. Some articles say that the ganglion itself is easily differentiated by its soft, irregular appearance on examination and by the presence of proliferative rheumatoid synovitis in the extensor tendons. I believe the author needs to confirm this.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This article proposes a method for automatic removal of ganglion cysts based on diffuse stretching and diffuse C-means quantification, with a success rate of 95.6% without human intervention.

• In the introduction the importance of the research is described, it is recommended to mention if there are in the literature applications of Fuzzy C-Means in the detection of similar pathologies.
• It is also recommended to include at the end of the introduction a description of what has been covered in the other sections.
• It is recommended to better organise the algorithm shown in equation 4.
• It is recommended to add an explanation of the construction of the Fuzzy Inference Rules for Fuzzy Stretching.
• In Figure 10, it is recommended to mention who is a) and who is b). Also, it is recommended to enter an image with the experimental setup.
• Although the comparison of the proposed method with ART2 is presented, it is recommended to extend the argumentation of the results and discussion.
• A more detailed conclusion on the results needs to be added.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The introduction of the manuscript has been well improved. The manuscript states that extraction of ganglions is useful for preoperative studies, is it difficult to distinguish correct from incorrect with the naked eye? Is it difficult to distinguish between correct and incorrect results with the naked eye, and what types of cases are likely to be incorrect? After reading this paper, I believe that the interpretation and discussion of the extraction results need to be done more fully as reference information for the reader to effectively conduct research.

A small point: fix the Caption "Lowe"(r) in the upper right corner of Figure 3.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

Improvement was seen in the explanation of the results. For the future application of image extraction, it is strongly recommended to set a task for diseases that can only be achieved by image detection, not for diseases for which there are other diagnostic methods, such as ganglion detection, which was the subject of this study.