Intelligent Brushing Monitoring Using a Smart Toothbrush with Recurrent Probabilistic Neural Network
Round 1
Reviewer 1 Report
The paper proposes a recurrent probabilistic neural network (RPNN) for toothbrush posture recognition. The RPNN model is trained for toothbrush posture recognition and brushing position, and then to monitor the correctness and integrity of Bass brushing technique.
I think the paper can be accepted after concerning the following comments:
- Please explain the meaning of Eq. (15) clearly.
- The format of the tables should be improved, such as Table 2.
Author Response
Reviwer1:
Q1: Please explain the meaning of Eq. (15) clearly.
Author response: Thank you Reviewer comments. The plan will also assess the integrity of brushing of each subject, and the formula for assessing the integrity of brushing times is as follows:
Q2: The format of the tables should be improved, such as Table 2.
Author response: Thank you Reviewer for your guidance and valuable comments. Next we will define the performance evaluation indicators for brushing posture recognition, in machine learning (ML), information retrieval (IR), accuracy, precision, recall, and F1 (F1-Measure). It is widely used to evaluate the pros and cons of different algorithms and models. Before understanding the above evaluation methods, it is necessary to define true positive (TP, true positive), true negative (TN, true negative), false positive (FP, False positive), false negative (FN, false negative), four classifications of the dichotomy, as shown in Table 2.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Overall the topic is an interesting application of machine learning, though I think the major “selling point” is the development of an efficient algorithm, not the application to tooth-brushing, per se. It was very difficult to follow the research report since I could not get past all the fragmented writing and what seems to be information we could find in the literature – it was hard to extract what was original here.
Line 20 - “… requirement at edge devices like smart phone …” Please clarify
Line 33 – I thought you were going to mention the higher propensity for cardiovascular disease as a consequence of oral disease.
Line 61 – “…Space, the busy life of people …” Please clarify
Line 73 – I thought you were going to discuss how semi-smart toothbrushes (those that time duration of brushing) has raised awareness of the 3-minute rule!
Line 100 – Is the phrase complete? “… The C??? Neural Network (CNN) …”
Lines 111- 113
“ … We will plan an innovative neural network classifier based on deep learning based on the lack of cost 112 of existing smart toothbrushes, low accuracy of brushing area, and lack of intelligence. Design a 113 forward-looking smart toothbrush based on Bass Brusng Technique….”
Please revise for clarity.
Line 166 – Transition verb needed? “ …but in real life. Many …”
Line 176 – Excellent comment. The CNN is popular in imaging and you are taking the CNN to a new level.
Line 172/176 Is the reference for CNN applications here the same as for the CNN reference on Line 100?
Line 178 – “…Two-dimensional picture. …” please clarify.
Line 220 – Please clarify “Sigmoid” - Is this an integral, transfer function, or other operator?
Lines 217 – 238 should eb entirely revised for clarity; use a flowchart or something besides the narrative in which the reader has to imagine the process. Script is very fragmented.
Lines 170 – 311 should be seriously reconsidered. How much of this is repeating lessons that can be learned form the literature? How much is relevant to the application at hand? Is there a clearer way to explain why these 3 approaches were chosen and their essential features?
Line 322 – Is this a clinical trial requiring an IRB approval?
Line 338 – Please put the chart on one page and please explain in the Caption what is the significance.
Line 394-399 – Excellent outcome!!
Author Response
Thank you Reviewer for your comments. Because there are many changes, we provide them as attachments. Please provide valuable comments again. Thank you. happy New Year.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Thank you for your effort in undertaking a significant revision of the paper and for the many improvements and clarifications. It will be important to note the policy under which you have indicated the bench trials are exempt from Institutional Review Board (IRB) approval. Generally even laboratory bench tests that are subsequently used in publishing results (for validation) are subject to IRB protocol approval. If your University has a policy waiving this requirement, it is essential that this be explicitly stated in the paper.
Author Response
Hello, dear editor:
Thank you for your valuable comments. We asked the translation company to edit the entire paper in English, and attach the translation certificate.
Author Response File: Author Response.pdf
