Modeling and validation of a detailed FE viscoelastic lumbar spine model for vehicle occupant dummies

Abstract

The dummies currently used for predicting vehicle occupant response during frontal crashes or whole-body vibration provide insufficient information about spinal loads. Although they aptly approximate upper-body rotations in different loading scenarios, they overlook spinal loads, which are crucial to injury assessment. This paper aims to develop a modified dummy finite element (FE) model with a detailed viscoelastic lumbar spine. This model has been developed and validated against in-vitro and in-silico data under different loading conditions, and its predicted ranges of motion (RoM) and intradiscal pressure (IDP) maintain close correspondence with the in-vitro data. The dominant frequency of the model was f = 8.92 Hz, which was close to previous results. In the relaxation test, a force reduction of up to 21% was obtained, showing high agreement in force relaxation during the in-vitro test. The FE lumbar spine model was placed in the HYBRID III test dummy and aligned in a seated position based on available MRI data. Under two impulsive acceleration loadings in flexion and lateral directions with a peak acceleration of 60 m/s2, flexion responses of the modified and original dummies were close (RoMs of 29.1° and 29.6°, respectively), though not in lateral bending (RoMs of 34.1° and 15.6°, respectively), where the modified dummy was more flexible than the original. By reconstructing a real frontal crash, it was found that the modified dummy provided a 10% reduction in the Head Injury Criterion (HIC). Other than the more realistic behavior of this modified dummy, its capability of approximating lumbar loads and risk of lumbar spine injuries in vehicle crashes or whole-body vibration is of great importance.