Research Question:​ Human surrogates such as anthropomorphic test devices and computational models are common tools used by the automotive industry to characterize injury mechanisms and design countermeasures. The surrogates currently available were designed to be representative of certain sex and percentile of the population, such as the 5th- and 50th-percentile females, and the 50th- and 95th-percentile males. Great improvements were achieved in crash protection thanks to these models, but there is a need for more refined models that take into account a greater variation of the human diversity, such as the geometric variation due to aging. Therefore, the objective of this study was to determine how the length of ribs varies with age and sex.

Methods and Data Sources:​ A total of 103 asymptomatic volunteer subjects aged 0 to 84 years old were included in the study. First, each volunteer was imaged either with a standard clinical CT-scan, or in a standing position with the EOS imaging system, a low-radiation X-ray system that acquired one frontal and one sagittal high resolution images of the ribcage. Second, a custommade software toolbox was used to create a subject-specific geometrical 3D model of each subject bony ribcage by registration of a statistical parametric ribcage model. Third, the rib length and mean thoracic index were extracted from the 3D ribcage models.

Results:​ The two main results were that the thoracic index (depth to width ratio of the thorax) was found to be fairly ageindependent, and that the ribs length increased linearly with age between 0 and 20 years old before reaching a plateau. The growth rate of the ribs increased between each rib level from rib 1 (4 mm/year) to rib 10 (10.5 mm/year). This indicates a change in the size and shape of the ribcage during growth.

Conclusion and relevance:​ The study provides a quantitative characterization of age and sex-induced variation in the ribcage geometry, based on asymptomatic adult volunteer data. This study addresses the need for the geometric data required to build a set of computational models that represent male and female subjects of various age.