Inertial Measurement Unit-Based Real-Time Adaptive Algorithm for Human Walking Pattern and Gait Event Detection

Round 1

Reviewer 1 Report

Manuscript ID: electronics-2575416

Title: On-board Short-term Model Training Strategy for IMU-based Real-Time Human Walking Pattern and Gait Event Detection

This manuscript presents a gait detection method with two inertial measurement units (IMUs) on the foot and thigh and results were evaluated. The paper is well written, explained, and presented the results. I’ve following comments:

1. There are few English mistakes, like on Page 1, ‘Abstract’ Line 17–sentence starting with ‘In which’ has meaning issue. Authors are advised to look carefully for such mistakes.

2. Authors should discuss the data inclusion criteria, if any.

3. How did authors prevent data leakage during training?

4. Why different x-y-z axis directions have been chosen for sensor placed at foot and thigh?

Overall, the paper is well written and contains interesting and revealing content with clear proof.

There are minor English language meaning issues, authors are advised to look carefully for such sentences.

Author Response

Dear Reviewer,

Thank you for taking time out of your busy schedule to review the revised manuscript (Electronics, electronics-2575416, “IMU-based Real-Time Adaptive Algorithm for Human Walking Pattern and Gait Event Detection”.) and give feedback.

These comments are all valuable and very helpful for revising and improving our manuscript, as well as the important guiding significance to our research. We have carefully studied the comments and have made corrections which we hope meet with approval.

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The article is interesting and valuable. The article is of high practical value. The article has a correct IMRaD structure. The article is important to colleagues working in the field of biomechanics The number of references is proper and references are modern. A classical statistical apparatus was used. No errors in reasoning. The work is thematically coherent. The results were presented in an aesthetic method. The article is scientifically credible. The work demonstrates scientific theorems that are consistent with experience, the laws of nature and do not have of a normative nature, i.e. they answer the question: how is it? not how it should be?

What is the accuracy of the IMU sensors ?

Major:

The title of the article is unclear.

In the introduction do not articulate the research gap found in science.

The vocabulary used differs from that used in biomechanics of gait.

The results were not validated by another experimental method.

The article should be directed to a journal dealing with the biomechanics of gait, although a link was found to the journal electronics

Minor:

You should reedit sentence "Table 3 shows..." etc. Improve greatly the English style.

There are editorial errors in the article. The work is not carefully edited!

Put some text below Figure 9.

Conclusions:

The use of IMU sensors in biomechanical research has been enthusiastically received. The correct logical operations were performed and a train of thought was carried out leading to a rational conclusion. Work is going in the right direction. Yeap that's it.

 

 

The article checked by an experienced linguist will be more accessible to Readers

Author Response

Dear Reviewer,

Thank you for taking time out of your busy schedule to review the revised manuscript (Electronics, electronics-2575416, “IMU-based Real-Time Adaptive Algorithm for Human Walking Pattern and Gait Event Detection”.) and give feedback.

These comments are all valuable and very helpful for revising and improving our manuscript, as well as the important guiding significance to our research. We have carefully studied the comments and have made corrections which we hope meet with approval.

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper proposes a gait detection system based on IMU and force-sensing resistors attached to the lower limb. According to the authors, data processed by a short-term, online training algorithm provides a robust and accurate classification of typical walking patterns.

The overall concept looks reliable and fits the Journal’s scope. However, some aspects need improvements: 

1)    Sec. 1: Please comment on the advantages and limitations of optical tracking methods for motion capture;

2)    Sec. 2.1: Fig. 1 depicts some wires connecting sensor moduli, which may impair the user’s movements. Furthermore, the IMU and magnetic transducers probably offset during gaiting events, yielding cumulative errors in the output data. How do you circumvent such problems?

3)    Sec. 2.3.2: How does the walking pattern detection module behave if the user remains static or performs unpredicted motions, like jumping? How does it ensure that new training data is trustworthy for classification and may update the previously stored one?

4)    Sec. 2.4. How many repetitions did you perform per volunteer? Are the results reproducible after donning/doffing the sensors?

The manuscript contains some writing and style mistakes, demanding revision by a professional editing service

Author Response

Dear Reviewer,

Thank you for taking time out of your busy schedule to review the revised manuscript (Electronics, electronics-2575416, “IMU-based Real-Time Adaptive Algorithm for Human Walking Pattern and Gait Event Detection”.) and give feedback.

These comments are all valuable and very helpful for revising and improving our manuscript, as well as the important guiding significance to our research. We have carefully studied the comments and have made corrections which we hope meet with approval.

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

This paper proposes a gait detection system based on IMU and force-sensing resistors attached to the lower limb. The authors improved the manuscript following our suggestions. However, some problems remain unsolved: 

1)  Reviewer 3, comment 2: I did not ask about the volunteers’ physiological characteristics. My question regards the wires connecting sensors and their occasional displacement due to repetitive motion, which may impair the repeatability of IMU and magnetic transducers. Please discuss such aspects;

2)  Reviewer 3, comment 4: How many repetitions did you perform per user? Are the results statistically robust? Please explain.

The manuscript contains some writing and style mistakes, demanding revision by a professional editing service.

Author Response

Dear Reviewer,

Thank you for taking time out of your busy schedule to review the revised manuscript (electronics-2575416, “IMU-based Real-Time Adaptive Algorithm for Human Walking Pattern and Gait Event Detection”). These comments are all valuable and very helpful for revising and improving our manuscript, as well as the important guiding significance to our research. We have carefully studied the comments and have made corrections which we hope meet with approval. Please see the attachment.

Author Response File: Author Response.pdf