Efficiency and Validity of the AI-Based rGMFM-66 in Assessing Gross Motor Function in Children with Cerebral Palsy

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

I commend the work and would like to offer the following constructive suggestions to improve the clarity, methodological rigor, and clinical applicability of the paper:

1.- Title: Consider clarifying that the study focuses on a specific tool (rGMFM-66) to more accurately reflect its scope.

2.- Abstract: We recommend softening the claim regarding a learning effect and adding a brief mention of the need for further external validation.

3.- Introduction: Condense overlapping information on GMFM-66 and GMFM-88. Including references to other emerging digital tools would enrich the context.

4.-Methods:

• Provide a more detailed description of the AI algorithm used (type, training, selection criteria).
• More explicitly address the limitations related to partial blinding.
• Include or acknowledge the lack of a sample size calculation.

5.-Results:

• Clearly separate primary (time efficiency) and secondary (validity) outcomes.
• Consider stratified analysis by GMFCS level or suggest it for future studies.

6.-Discussion:

• Expand on practical clinical implications (cost, ease of use, acceptability).
• Discuss the clinical relevance of the slight score increase with the rGMFM-66.
• Reflect on additional limitations from the study design.

7.- Conclusions: Recommend including a sentence emphasizing the need for multicenter validation studies.

8.- Style and formatting: Address minor typographic issues and ensure consistency in reference formatting.

9.-Additional content (optional but recommended):

• Include a comparison table of GMFM-66 vs rGMFM-66.
• Expand slightly on the ethical implications of AI use in pediatric care.

The manuscript holds significant potential, and with these adjustments, it could make a stronger contribution to the field of pediatric rehabilitation and applied clinical AI.

Comments for author File: Comments.pdf

Author Response

Please see the attachment 

Kind regards

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Thank you for providing me the opportunity to review the study entitled: “Efficiency of artificial intelligence-based technologies to quantify gross motor function in children with cerebral palsy”. I tried to offer impartial criticism, hoping I achieved that aim.

This study aimed to assess the efficiency and validity of the reduced Gross Motor Function Measurement (rGMFM-66). However, the methods, results, and discussion sections require significant revisions. Therefore, I believe major changes are needed before this manuscript can be accepted.

Specific comments can be found in the file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment 

Kind regards 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

  Dear authors
 I have read a REALLY INTERESTING manuscript, leading in the new world of AI. 
Your manuscript is WELL PREPARED, it has clear description and provides accurate assessment for the initially presented question regarding the efficiency of AI technologies.
As a PEDIATRIC ORTHOPAEDIC SURGEON, dealing with CP children, I would like to pose the TYPICAL questions of the surgeon, dealing with CP Children:
 Can you justify the ABSENCE of an orthopaedic surgeon in the evaluation of these children, since there is no OBVIOUS report for the presence of an orthopaedic surgeon?
1.     I can understand that the evaluation and measurement of GMFM is mainly performed from physio BUT the assessment and evaluation of spasticity versus tendon-muscle shortening is extremely difficult and it can CHANGE the measurements of movements ( kneeling – standing). Accurate measurements are NOT ALWAYS reproducible in the SAME day and it may influence the results.
2.    I cannot understand the decision WHICH ITEMS ARE OMISSED- WITHDRAWN in
RGMFM. You mention the number 34 items, please explain the selection method.
       3   the same applies for the selection of 7 items in order to classify in 3 groups. Can you provide information for the decision of which items you have used. Is it an AUTOMATIC decision?

Your study design is well presented, but the use of WBV may influence the initial GMFM result since it is a TRAINING THERAPEYTIC method.

I find ALWAYS unpleaseant to comment on statistics, you can present ANY result occasionally, the way it is available and fits the results!!

Your DISCUSSION is well formed, well presented, focusing on the accuracy and the reduced time for the evaluation.

BUT CLINICAL EVALUATION OF PATIENTS REMAINS AN ESSENTIAL TOOL FOR MEASUREMENTS AND AUTOMATIC EVALUATION OF ITEMS FROM AI CAN INTERFERE WITH EXPERIENCE OF THERAPISTS. So please comment for the importance of accurate CLINICAL ASSESSMENT AND experience of the therapists.
As well, THERAPISTS AND DOCTORS ( MAINLY SURGEONS) WILL BE EVALUATED FROM AI SYSTEMS FOR THE ACCURACY OF THEIR TREATMENT and METHODS. Please comment on this and for importance of ethics when we use AI measurements.

In GENERAL I found VERY INTERESTING your manuscript and it is worth to be published after minor revision!

 

Author Response

Please see the attachment 

Kind regards

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for the revised version of your manuscript "Efficiency and Validity of the AI-Based rGMFM-66 in Assessing Gross Motor Function in Children with Cerebral Palsy." The paper is significantly improved and presents a clear, well-structured and relevant study with valuable clinical implications.

I particularly appreciate the improvements in:

• The clarity of the introduction.

• The more detailed description of the AI algorithm.

• The thorough explanation of the study design, standardization and limitations.

• The inclusion of the comparison table between GMFM-66 and rGMFM-66.

• The expanded discussion on clinical applicability and ethical considerations.

At this stage, I only have a few minor suggestions to further strengthen your manuscript:

Blinding / Potential bias:
While you already discuss the partial blinding, I suggest slightly reinforcing in the Limitations section that the awareness of the assessor conducting the rGMFM-66 could have consciously or unconsciously contributed to faster administration times.

Order of tests and learning effect:
You appropriately mention the potential learning effect. I recommend emphasizing more clearly that counterbalancing test order (when ethically and practically possible) should be a priority in future studies.

Generalizability to non-spastic forms of CP:
Since your sample was mainly composed of children with spastic CP, it would be useful to explicitly state that further validation is needed in non-spastic CP subtypes.

Ethical frameworks for AI:
You have a strong paragraph on ethical considerations. It would further enrich the discussion to cite specific AI ethics guidelines in pediatrics (e.g., recent position papers or frameworks) to provide additional context.

Supplementary materials:
It would be helpful to clarify whether the video or tables in the supplementary materials contain examples of how the AI selects items or is implemented in clinical practice.

These are minor revisions. Overall, the manuscript is of excellent quality and contributes meaningfully to the field. I recommend acceptance after minor revision.

Author Response

Please see the attachment

Kind regards 

Stefanie Steven 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The manuscript titled “Efficiency of artificial intelligence-based technologies to quantify gross motor function in children with cerebral palsy"   has been sufficiently improved to warrant publication in Applied Sciences.

Author Response

We would like to sincerely thank the reviewer for the positive evaluation of our manuscript titled “ Efficiency and Validity of the AI-Based rGMFM-66 in Assessing Gross Motor Function in Children with Cerebral Palsy.”

We are pleased that the manuscript is considered sufficiently improved and suitable for publication.
Since no further revisions were suggested, we have not made any additional changes.

Thank you again for your time and support.

Kind regards,
Stefanie Steven