EVALUATING THE THORACIC RESPONSE IN A HUMAN BODY MODEL, COMPARED TO CADAVER EXPERIMENTS

Fatalities caused by car accidents accounted for 1.2 million worldwide in 2004. Chestinjuries are the second death cause, after head injuries, in vehicle accidents. Devices topredict risk of injury are fundamental to develop and evaluate restraint systems that canmitigate the injury severity and reduce the fatalities. This can be achieved withmathematical Human Body Models (HBM). An HBM needs to be biofidelic, both in termsof size and biomechanical response. Herein, the biofidelity of the thorax region in an HBMis evaluated. The evaluation focuses on whole body kinematics and chest response duringfrontal car accidents.Finite element simulations with an HBM were performed with the code LS-DYNA [1].A modified version of Total HUman Model for Safety v3.0 (THUMS) [2] has been the basisfor the work presented here. THUMS was modified to improve its robustness andnumerical stability. The meshes of the rib cage, skin around the ribcage, and intercostalmuscles were refined. THUMS represents a 50th percentile male and has approximately150,000 elements. It was compared with cadaver pendulum impacts [3], table top tests [4],and sled tests [5]. The pendulum test consist of a cylinder with diameter 125 mm, mass of23.4 kg and an initial speed of 4.3 m/s impacting on the middle of the sternum. The tabletop tests included four loading conditions of the chest: hub, diagonal belt, double diagonalbelt, and distributed load. The sled test was performed at 40 km/h, using a three pointstandard seat belt.The model has been compared to cadaver experiments on different load cases, whichare representative of modern automotive restraint systems. In general, THUMS showed agood agreement with the experimental corridors for the pendulum and table top tests. Thechest response in the sled test differed for the lower chest, possibly due to the absence ofbiofidelic fracture simulations in the THUMS. After comparing the kinematic and dynamicresponses of THUMS with cadaver experiments it was concluded that the model isadequate to simulate the human response under frontal impacts. The next step is to identifyparameters that can predict the risk of rib fracture and then become a tool to evaluate newrestraint systems