Development of an Automatic Functional Movement Screening System with Inertial Measurement Unit Sensors

Round 1

Reviewer 1 Report

The paper is very well writen and structured.

Recommendations:

  - The abstract should not show numerical results but descriptive results.

  - Define the acronym ROM

  - Table 1 is confusing because there is not separation between lines of information. Consequently, is very hard to distinguish the information clasification.

  - Line 237 has text in red. Reference [23] in line 276 is also in red

  - The conclusion should be more "conclusive" as a summary of all the previous results. In fact, the conclusion shown in the abstract is more explicit than the shown in the conclusions.

Author Response

Please see the attachment.

Author Response File: Author Response.doc

Reviewer 2 Report

The article describes the use of an automatic scoring system for the functional movement screen assessment. The paper is concise and matches with the scope of the Journal of Applied sciences. The work overall has rationality and completeness of research problem and sufficient explanation of findings. English and writing are of academic level. The discussion is appropriate but there are a few points in the methods that require further details and justification:

1. Lines 60-62: are the definitions of mobility and stability here defined by the authors? If yes, why are they in quotation marks? If not, citations are missing.
2. Lines 66-78: in my opinion, I do not believe that this paragraph has much to add to the paper. The authors reference studies using IMUs and machine learning technology in scoring sport performance, but the referenced manuscripts are not related to FMS scoring and no technical details are drawn from them or used in the authors’ work. If the authors agree, I would recommend deleting this paragraph entirely.
3. Line 112: “The sensor measurement angle error was less than ±1° [21]”, please make clear that this error was not tested in the present study but you are referring to the manufacturer’s specs.
4. Lines 134-235: “In addition, the shoulder mobility test was excluded because it was determined to be less suitable for measurement by IMU sensors.” Determined by whom and how? If the system measuring RoM can calculate shoulder flexion, abduction and rotation with an “angle error was less than ±1°” why was that test excluded?
5. Lines 159-160: “Prior to modeling, min–max normalization of ROM data was conducted, and the data were scaled to the interval.” Is the min-max here refers to the min-max values of the RoM among all the participant’s recordings of the same exercise? It is not clear how this normalisation was done. I have the impression that by RoM you are not referring to a single value measuring the amount of the joint’s movement.
6. Lines 172-173: “The root-mean-square error (RMSE) metric was then used to compare the five models and determine the optimal one.” Your test suggests that you are performing classification rather than regression. RMSE and R-square are measures associated with continuous variables. To my understanding, it would be preferable to use MAE, recall, precision or F1 score to measure the performance of the model and not RMSE. Is there a specific reason behind the selection of the RMSE?
7. Line 223: “Higher R2 values indicate better goodness of fit.” I believe that goodness of fit is not an appropriate index to measure the performance of your model. R-squared tests how well a linear model (that is a continuous response variable as a function of one or more predictor variables) fits a set of observations. To my understanding, your response is not continuous and the model is not linear.
8. Lines 237: “presents the stepwise regression analysis of the ROM data”. There might be a violation here of the regression analysis assumptions that require variables to be measured in the continuous level. Perhaps your data are better fit for Ordinal logistic Regression analysis?
9. Table 4, I believe that measurement units are missing.

Author Response

Please see the attachment.

Author Response File: Author Response.doc

Round 2

Reviewer 2 Report

The authors have addressed all of my concerns and comments.

Author Response

Thank you for your suggestion.