Experimental and numerical replication of blunt ballistic impact on a male thorax surrogate: study of rib fractures and lung injuries predictions.

Prediction of body injuries caused by non-penetrating ballistic impacts is a major challenge. Experiments on biological models are difficult to conduct due to ethical and logistical constraints and on the data that can be collected. These limitations can be overcome by using surrogates, but their results must be correlated with real injuries recorded by ballistic injury databases. The objective of this study is to establish injury prediction curves for rib fractures and lung injuries. Several ballistic impact scenarios from French and American police reports were analyzed. These cases were recreated experimentally using a biofidelic thoracic surrogate, SurHUByx, and numerically using its numerical twin, SurHUByx FEM. These surrogates represent the thorax of a 50th percentile male and were designed to reproduce the biomechanical responses of the human body to ballistic impacts. Then a scaling process adjusted the data according to body mass index to account for individual variability. The optimal cutoff for injury diagnosis was determined using the Youden method. And the injury risk curves were constructed using survival analysis according to ISO guidelines. The results show the risk of injuries according to rib deformation and pulmonary pressure over time. The injury probability was also plotted against the backface deformation. For a standard 44 mm intrusion, the risk of rib fracture was 44.84% and lung injury was 52.45%. These results highlight the limitations of current body armour standards, which may underestimate the probability of internal injuries. This study highlights the value of sharing ballistic injury cases to build a more reliable database. It also highlights the use of surrogates to replicate cases in order to develop injury prediction models.