Adult Head and Neck Dynamics: A Sensitivity Analysis Study during Frontal Impact

II. METHODS The significance of each model parameter of the current study was analysed through sensitivity analysis. Each parameter was varied plus and minus one standard deviation as reported in the literature. A subset of these parameters is given in Table 1. TABLE I Parameter Min Scale Factor Max Scale Factor Source Neck length 0.957 1.043 Gilad and Nissan 1986 Muscle max isometric stress 0.400 2.00 Winters and Stark 1988 Muscle attachment – x ‐1.000 1.000 Chancey et al. 2003 Head centre of gravity (CG) location was varied relative to the occipital condyle joint. Intervertebral joint stiffness for the upper and lower cervical spines were varied jointly. The 22 muscle insertions, origins, and physiological cross sectional areas (PCSAs) were varied by their respective standard deviations. Muscle attachment locations were varied jointly in each of the three degrees of freedom, with the same percentage of their standard deviation in the same direction. Latin hypercube sampling was used to select the value of the variables and 50 simulations were run per variable [4‐5]. With 31 model variables, a total of 1,550 simulations were run with additional simulations when boundary conditions were included. Model development and activation states are detailed in [6]. Model boundary conditions for the NBDL simulations are given in [7]. Briefly, the horizontal acceleration and sagittal rotation kinematics from the first thoracic vertebrae (T1) were applied to the model. Simulations were run using LS‐DYNA (LSTC, Livermore, CA).