Investigating CT scan quality: Appendices

APPENDIX 1

Methods

The full version of VGSTUDIO MAX v.3.1 has a wider range of tools for optimizing surface determination. We tested the impact of 10 surface determination calculations on shell volume, the methods for which are described in Table A1.

Results

Although the different surface determination modules found different absolute values, they capture the same trend of shell volume increasing with exposure time. All the ROI refinement methods except one show the opposite trend, with shell volume decreasing with increased exposure time. All the analyses in the main manuscript were performed using the simple surface determination module which falls near the center of the range of values calculated from the 10 different surface determination modules (Figure A1.1). Data are available in Appendix 3.

Discussion

Shell volume calculations are based on the number of pixels that fall within the selected greyscale range for shell material. By selecting different methods to determine where the surface between shell and background, or shell and organic material lies, the calculated volume of shell material will change. Although the absolute values vary, all of the surface determination and refinement methods capture the same trend: increasing shell volume with increasing exposure time. This is likely due to the fact that increasing exposure time increases the amount of data that is collected, making it easier to distinguish the three materials.

The smoothing refinement methods show the opposite trend, with a slight decrease in shell volume with increased scan time. The smoothing measurements are also all lower than the surface determination and distance refinement methods. This is because smoothing erodes some of the pixels from the edge of the shell material, decreasing the calculated volume of shell material. The higher the smoothing number, the more material is removed, lowering the resulting calculated shell volume. If these new surfaces are visualized, the smoothing can sometimes create false holes in the pteropod shells because the shell thickness has been smoothed to nothing.

As the automatic surface determination falls near the center of the range of values calculated using the ten different methods (Figure A1.1 and Figure A1.2), and the fact that it is available on all versions of VGSTUDIO MAX, all the calculations in the main manuscript are based on measurements made using the automatic surface determination module.

REFERENCES

VGSTUDIO MAX Reference Manual, 2018. Volume Graphics, Heidelberg, Germany.

Oakes et al., 2019 - Testing the impact of two key scan parameters on the quality and repeatability of measurements from CT scan data

APPENDIX 2

Shell volume calculations from the 10 surface determination modules. The data are presented visually in Figure A1A-A1B. (Data for Appendix 2 and 3 presented in zipped spreadsheet files).

APPENDIX 3

Greyscale measurements used for signal-to-noise ratio, contrast-to-noise ratio, and beam shift calculations. (Data for Appendix 2 and 3 presented in zipped spreadsheet files).

APPENDIX 4

Data for Figure 1

TABLE 1A4. Shell volume calculations from the 11 trial scans in the scan parameter experiment and the 16 trial scans in the repeatability experiment. The data are presented in Figure 1 in the main manuscript.