Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript addresses a relevant methodological question in manual wheelchair propulsion biomechanics by comparing phase durations derived from wheel and hand acceleration. The topic is well motivated, and the speed-related differences may be useful for methodological and clinical interpretation. However, several conceptual, statistical, and reporting issues need to be addressed before the findings can be interpreted with confidence. Overall, this paper is a good work. Once the following concerns are adequately resolved, the manuscript will be suitable for acceptance.

1. The two approaches use different operational definitions of the propulsion phase, with Approach 1 based on wheel acceleration and Approach 2 on hand–pushrim contact. The manuscript should therefore avoid implying validation or interchangeability between the methods and clarify this distinction throughout. The term “agreement” should also be used cautiously.
2. The proposed explanation that Approach 2 captures non-propulsive hand–pushrim contact is plausible but is not directly verified in this study. If no independent force or contact reference is available, this mechanism should be presented as a hypothesis supported by previous literature rather than as a demonstrated finding.
3. Three slow-speed trials were excluded because Approach 2 failed to detect events, and only cycles successfully detected by both approaches were analyzed. Please report the number and percentage of excluded trials/cycles and the reasons for exclusion. A sensitivity or participant-level analysis would help assess potential selection bias.
4. The “slow,” “preferred/normal,” and “fast” conditions are not objectively characterized. Please report the instructions and, where available, actual speed, cadence, loading/resistance, and power output. The fixed preferred–slow–fast order and the limited generalizability of the stationary test-stand setup should also be acknowledged.
5. Multiple cycles from the same participants appear to have been treated as independent observations. The analysis should account for within-participant clustering using an appropriate repeated-measures approach or, at minimum, participant-level averaged results. Spearman correlation should not be interpreted as agreement, and the speed-dependent effect should be supported by an explicit statistical test.
6. Additional methodological details are needed for reproducibility. Please clarify the calculation, filtering, zero-crossing detection, and marker-occlusion handling for Approach 1, and report the main CWT and event-detection parameters for Approach 2. The handling of different native IMU sampling rates and resampling to 2000 Hz should also be described.
7. The practical implications should be clarified by explaining which research or clinical questions are better addressed by each operational definition. The authors should also provide guidance on reporting these methods in future studies. Given the speed-dependent differences, pooling results across approaches or applying a universal correction factor should be discouraged without further validation.
8. Many of the cited works on motion capturing techniques are relatively dated. To better position the novelty of this paper, the authors should review and discuss more contemporary research. For instance, recent work in perception in extreme environments, such as A DMD-based adaptive modulation method for high dynamic range imaging in high-glare environments. This is the basic input forvision-based motion capturing techniques. This will allow the proposed method to have more application scenarios.
9. Table 1 provides limited information on participant and wheelchair characteristics. Relevant neurological characteristics and key wheelchair configuration variables should be reported where available, as these factors may influence hand–pushrim interaction. Unavailable variables should be acknowledged as limitations.
10. The conclusions should be restricted to the studied population, test-stand setting, propulsion conditions, and the two specific detection methods. The findings should not be generalized to all wheelchair users or propulsion environments. Without an independent criterion measure, neither approach can be considered universally more accurate or superior.

Author Response

Thank you very much for taking the time to review this manuscript, and for your helpful comments. Please find the detailed responses in the attached file, and the corresponding revisions/corrections, which are highlighted in blue in the resubmitted manuscript.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Abstract

Consider using the word one and two rather than the numerical option for approaches.

Introduction

Line 54: The glenohumeral joint.

Paragraph line 53-60: Reword for clarity. Each sentence can be more descriptive and concise.

Line 64: Different to what? You listed detection mechanisms but did not clarify what they are different too. Work on sentence clarity.

Line 76: It is not clear to the reader how positive wheel acceleration is not detected in marker-based camera systems or IMU’s.

Line 81: Do not start a sentence with a citation.

Hand-pushrim should be defined. The phases of wheelchair propulsion should be defined.

Line 86: Speed? So, no direction?

Methods

What was participant exclusion criteria?

Table 1. lowercase n.

Line 110: what did the participants do upon arrival to the lab?

Line 113: How much time were participants given?

Line 117: Who placed the markers? Were all markers placed by the same researcher?

How were the three speeds determined? Was preferred speed determined via a cost of transport measurement? How was preferred speed consistent between participants? Were slow and fast speeds the same between each participant?

Figure 1: the length of propulsion and recovery phase should be different for each individual. Wheel chair design and limb length would have an affect on this.

How much rest was between cycle?

Results

The first paragraph of the results section should not be included in results.

Spacing between table 1 and table 2 is not consistent.

Discussion

Include limitations.

Author Response

Thank you very much for taking the time to review this manuscript, and for your helpful comments. Please find the detailed responses in the attached file, and the corresponding revisions/corrections, which are highlighted in blue in the resubmitted manuscript.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised version provides detailed responses to my questions. I have no additional questions. The revised version can be accepted for publication.

Author Response

We thank the reviewer for the positive assessment of our revised manuscript and for the helpful comments during the review process.

Reviewer 2 Report

Comments and Suggestions for Authors

Using three different researchers to place markers is a major limitation. Using a different researcher to "double check" the placement is not valid. 

Choosing preferred speeds for all three scenarios is also a major limitation. This does not reflect difference between individuals preference. 

Author Response

We thank the reviewer for the careful reading of our revised manuscript and for the constructive comments, which have helped us to further improve the clarity and transparency of our work.

 

• Using three different researchers to place markers is a major limitation. Using a different researcher to "double check" the placement is not valid. 

Response: We agree that a second researcher checking the placement does not remove inter-rater variability. We agree that for markers attached to the body, for example when joint kinematics or inverse dynamics are computed, placement by a single researcher is the appropriate standard, because small placement differences propagate directly into the results. In the present study, however, the markers were mounted on the wheel, i.e., on a rigid body. Approach 1 derives angular acceleration from their rotation about the wheel axis, and for a rigid body the angular acceleration does not depend on where exactly the markers are placed on the wheel. Small placement differences therefore do not affect the zero crossings used to define the phase boundaries.

• Choosing preferred speeds for all three scenarios is also a major limitation. This does not reflect difference between individuals preference. 

Response: We thank the reviewer and agree that self-selected speeds are a limitation, which we address in the Limitations (see below). To make clear that the three conditions were defined relative to each individual’s own preferred speed, we have revised the instruction in the Methods (see below): participants were instructed to propel at their preferred speed, at a slower speed than preferred, and at a faster speed than preferred, with no target speeds prescribed. The resulting absolute speeds differ between individuals (mean ± SD: 0.43 ± 0.23, 0.86 ± 0.45, and 1.59 ± 0.90 m/s for slow, preferred, and fast), which reflects inter-individual differences.  For this reason, the speed dependence was not analyzed using the condition labels (preferred, slow, and fast speed) but using the actual propulsion speed as a continuous covariate (repeated-measures ANCOVA). The results thus do not rely on slow, preferred, and fast corresponding to the same speed in all participants.

Methods: “Participants were instructed verbally to propel their wheelchairs in three conditions, in this order: at their preferred speed, at a slower speed than preferred, and at a faster speed than preferred. Speeds were self-selected in all conditions, and no target speeds were prescribed. The preferred speed was not determined by a separate measurement (e.g., cost of transport).”

 

Limitations: “First, propulsion speeds were self-selected rather than standardized across participants. Absolute speeds therefore differed between individuals which may have influenced the detected phase durations.”