Effects of Loading Conditions on the Pelvic Fracture Biomechanism and Discrimination of Forensic Injury Manners of Impact and Run-Over Using the Finite Element Pelvic Model

Round 1

Reviewer 1 Report

The paper by Li et al. deals with pelvis fracture under impact loading studied by Finite Element analysis.

The paper is not well presented at all.

Figures are not visible, a too small font size is used font or too much informations are present in the figure (42 subfigures in Fig 8 for instance)

The quality of the figures is not good, the font size should be adapted to be the same as in the text and 600 dpi (at least) figures should be provided.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Authors,

The paper can be accepted for publication, but several additions and corrections are necessary.

1. Check punctuation and spacing within the paper’s text (see worldwide[1], etc.).
2. The subscripts (see sigmaY, εfail, etc.) and superscripts (see kg/m3, etc.) must be written correctly.
3. How do you confirm that the simulation results are correct? The results of a simulation can be considered correct if they have small deviations, compared to the experimental data. The comparison between the curves (experimental and simulated) illustrated in fig. 4 shows large deviations. Validation of FE numerical simulations is required to be achieved through experiments.
4. More figures (for example, Figs. 10, 11, 13, etc.) do not have a good enough resolution for publication. A higher resolution of figures (minimum 600 dpi) is required.
5. The quality of all figures must be improved. Font size of the figures must be comparable to that of the paper’s text and the same for all figures. The images obtained by simulation and all others can be processed in other editing program, to have an appropriate font size.
6. The equations of the article must be numbered (see line 256).
7. Section “5. Conclusions” needs to be developed and argued better.
8. Check the correct way of writing all the referencing, which must comply with the journal’s requirements.
9. English is good, but the flow of some sentences could be improved for better legibility.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This manuscript is a comprehensive analysis of various loading conditions on pelvic fractures, by simulation on the THUMS finite element model. A small number of issues might be considered:

1. From Figure 4, there remains a certain gap between simulated and experimental results. It might be discussed if any attempt might be made, to adjust the FE taking experimental results into consideration.

2. Brief sensitivity analyses for collision position (i.e. not exactly the midpoint) might be considered.

3. Modelled outcomes (e.g. "Both back and frontal run-over can result in characteristic open-book fractures", Line 213) might be substantiated with reference to clinical evidence (e.g. case studies), where possible.

4. The discrimination of injury manners (Line 321, Figure 18a) might be clarified further. Does this refer to using the retrieved principal component values, to predict injury manners? If so, what are the classes of injury manners predicted here?

5. Some minor grammatical issues:
(Line 23) "while no characteristic under high-velocity conditions" -> "...while no such characteristic was observed under high-velocity conditions"
(Line 24) "Loads on the frontal and back" -> "Loads on the front and back"
(Line 408) "discriminating ability" -> "discriminative ability"
etc.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The figures have been slightly improved but do not match the required level for publication, for instance why are there two legend in Fig 3 ? (one of which is not visible)

The finite element analysis is carried out without any convergence analysis, which is not acceptable.

Moreover, the analysis provided by the authors is not always clear. Why would the authors studied Von Mises stress, they do not provide any evidences of the material behavior being similar to that of a metal for instance.

It is crucial to conduct a mesh convergence analysis especially when deleting elements. Element deletion results in a loss of mass in the model, which is absolutely not physical.

I thus still do not recommend the paper for publication in Applied Science

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear Authors,

The authors have provided clarifications on their work, but some additions and corrections are still needed.

1. The sensitivity must be marked with a letter, the equal sign and the value of a/this parameter cannot be used after a word (see line 29, etc.).
2. The font size of all the figures must be comparable to that of the paper’s text, for ALL elements of the drawing/image (see X, Y, Z in Fig. 1, etc.).
3. In fig. 3, “Gravity” will be written rotated by 180º.
4. The explanation of a figure’s parts (see fig. 5, etc.) must be written next to the figure’s text, not separately below it. Example: “Figure 5. Pelvic impact loading conditions: a. Vertical positions of the plate impactor; b. – h. Represent the impact directions of 0°, 30°, 60°, 90°, 148°, 270°, 300°, and 330°, respectively.”
5. The comparison between the curves (experimental and simulated) illustrated in fig. 4 shows smaller deviations after correction. The maximum deviation size between the two data strings must be determined. A personal opinion: It is not completely right that the validation of one's own FE numerical simulations is done by comparison with experiments of other authors, because the experimental conditions used will not always be fully known.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

We thank the authors for addressing our previous concerns. However, while it is stated that the THUMS pelvic model has been scaled to better mimic the empirical curve (Figure 1), the following results (e.g. Figure 10, Table 4) appear unchanged from the previous manuscript. This might be clarified.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx