(E-184) Feasibility of Automated Methods for the Characterization of Rib Cage Shape Variability

Introduction: Motor vehicle crashes (MVCs) frequently result in serious thoracic injuries that lead to poorer outcomes [1]. To better understand injury tolerance in MVCs, researchers often use finite human body models that are designed to match a target demographic. However, these models typically rely on a single reference subject, overlooking the diverse rib cage shapes found in adults. While previous studies have indicated age-related rib cage shape changes in smaller samples [2-4] or individual rib shape differences [5], comprehensive analysis of whole rib cage shape variability, regardless of age, remains limited in larger populations. In this study, we developed an automated pipeline to characterize rib cage shape variability in a large adult population.


Materials and Methods: IRB approval was obtained to retrospectively analyze chest CT scans from 4,800 adults aged 25-45 years. An open-source automated segmentation algorithm was used to extract rib cage segmentations, and custom MATLAB code was developed to extract nine measure describing rib cage shape from each segmentation. Accuracy of the automated results was evaluated via 3D surface deviations from manual segmentations, percent differences to manually collected rib cage measurements, and qualitative visual inspection of segmentations.


Results, Conclusions, and Discussions: Preliminary results were collected from a subsample of 25 individuals (12 males, 13 females). Mean height was 180.9 ± 6.7 cm for males and 161.2 ± 7.0 cm for females. Mean weight was 88.1 ± 23.7 kg for males and 68.8 ± 17.9 kg for females. The analysis pipeline took 15s per scan to run. The mean positive 3D deviation was 3.1 ± 2.1 mm, while the mean negative 3D deviation was -1.9 ± 1.4 mm. On average, manual and automated measurements differed by less than 6% across all nine rib cage measures. Visual inspection revealed that the segmentation algorithm effectively captured rib cage shape, with the exception of the rib head, which was excluded in all segmentations. Measures tended to be smaller in females than in males and showed trends with BMI, a result consistent with prior findings. The pipeline developed in this study can consistently, accurately, and efficiently extracted features describing rib cage shape. These findings are based on a limited sample size and serve as a foundation for future expansion for the final study population.


Acknowledgements (Optional): This work was supported by the Global Human Body Models Consortium, LLC and NHTSA under GHBMC Project No. WFU-006.


References (Optional): 1. Benhamed, A., et al., Road traffic accident-related thoracic trauma: Epidemiology, injury pattern, outcome, and impact on mortality—A multicenter observational study. PLOS ONE, 2022. 17(5): p. e0268202.

2. Weaver, A.A., S.L. Schoell, and J.D. Stitzel, Morphometric analysis of variation in the ribs with age and sex. J Anat, 2014. 225(2): p. 246-61.

3. Gayzik, F.S., et al., Quantification of age-related shape change of the human rib cage through geometric morphometrics. J Biomech, 2008. 41(7): p. 1545-54.

4. Kent, R., et al., Structural and material changes in the aging thorax and their role in crash protection for older occupants. Stapp Car Crash J, 2005. 49: p. 231-49.

5. Holcombe, S.A., S.C. Wang, and J.B. Grotberg, The effect of age and demographics on rib shape. J Anat, 2017. 231(2): p. 229-247.