Fast Classification of Thyroid Nodules with Ultrasound Guided-Fine Needle Biopsy Samples and Machine Learning

Round 1

Reviewer 1 Report

Referee Report

• Abstract section should be revised and shortened.
• Figures 3, 4,and 5 should be revised.
• If possible theoretical computations should be conducted.
• For introduction section, the new reference should be added such as:

             Quantum Computational Investigation of (E)-1-(4-methoxyphenyl)-5-methyl-N′-(3-phenoxybenzylidene)-1H-1,2,3-triazole-4-carbohydrazide, DOI:10.3390/molecules27072193.

• All figures should be revised according to resolution.
• Conclusion is very weak should be revised according to important findings.

MINOR REVISION

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In the manuscript entitled “Fast Classification of Thyroid Nodules with Ultrasound Guided-Fine Needle Biopsy Samples and Machine Learning”, the authors used experimental data from a group of 267 patients for training a machine learning algorithm (MLA) to discriminate malignant and benign thyroid nodules. The patients’ group was divided in four sub-groups, namely malignant (98), benign (110), undetermined (42), and single blind patients (17). Despite the relatively small number of samples, the authors used optimization techniques to reach an accuracy of 92%. The experimental and computational procedures are thoroughly described in the manuscript and previous related works were cited and commented. Although the contribution of this work is incremental, i.e., does not have a high level of novelty, I think that results in this work can be of interest for other researchers interested in the subject. In particular, this work can motivate further research to improve artificial intelligence aided diagnosis of thyroid nodules.

 

Before publishing this manuscript in Applied Sciences, I will as the authors to address the minor comments listed below:

 

• There are several acronyms that were used before definition or never defined like, for example, BRAF and MLP.

 

• There are some grammar issues like, for instance, “significantly comparing using” in line 19 of the abstract. There are several other typos throughout the text that authors should check.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

I’ve read with attention the paper of Wang et al. that is potentially of interest. The background and aim of the study have been clearly defined. The methodology applied is overall correct, the results are reliable and adequately discussed. The only comment I have is that the paper lacks of any statistic description and discussion of the study limitation. These section should be improved.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The manuscript investigates the use of a PESI-MS/MS-machine learning method to classify malignant and benign thyroid nodules with ultrasound-guided fine-needle aspiration biopsy samples. Overall, the work is quite interesting and well-organized. However, I have some observations that are listed in the following lines.

1. There are some typos and grammatical errors. The work would benefit from close editing.
2. Avoid using abbreviations in the abstract and conclusions.
3. Add the details of the method validation process.
4. Figure 2: The component difference marking seems incorrect. Please fix it.
5. Table 3: Add what the abbreviations stand for in the table footer.
6. Line 192: Avoid using references in the results section. Please rephrase.
7. Line 287: Avoid starting a sentence with a number that is not written out.
8. Add a list of abbreviations.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf