Validation of human spine model under the low-speed rear-end impact

Track 5. Occupational and Impact Injury Biomechanics 5132 Mo, 14:15-14:30 (PIO) Superficial and deep neck muscle activity during isometric, voluntary and reflex contractions G.P. Siegmund 1,2, J.-S. Blouin 2, J.R. Brault 1, S. Hedenstierna3, J.T. Inglis 2. 1MEA Forensic Engineers & Scientists, Richmond, BC, Canada, 2School of Human Kinetics, University of British Columbia, Vancouver, BC, Canada, 3Division of Neuronic Engineering, Royal Institute of Technology, Huddinge, Sweden Increasingly complex models of the neck neuromusculature need detailed muscle and kinematic data for proper validation. Our goal was to measure the electromyographic (EMG) activity of superficial and deep neck muscles during isometric, voluntary and reflexively-evoked contractions of the neck muscles. Three male subjects (28-41yrs) had EMG fine wires inserted into their left sternocleidomastoid, levator scapulae, trapezius, splenius capitis, semispinalis capitis, semispinalis cervicis, and multifidus muscles. Surface electrodes were placed over the left sternohyoid muscle. Subjects then performed (i) maximal voluntary contractions in eight directions (450 intervals) from the neutral posture, (ii) 50N isometric contractions with a slow sweep of the force direction through 720°, (iii) voluntary oscillatory head movements in flexion/extension, and (iv) initially-relaxed reflex muscle activations to an abrupt forward acceleration while seated on a sled (apeak = 1.5g; At =60 ms; Av = 0.5 m/s). Isometric contractions were performed against a load cell and movement dynamics were measured using 6DOF accelerometry and motion tracking on the head and torso. In all subjects, the two anterior neck muscles had similar preferred activation directions and acted synergistically in both dynamic tasks. With the exception of splenius capitis, the posterior and posterolateral neck muscles also showed consistent activation directions and synergies during the voluntary motions, but not during the sled perturbation. These findings suggest that the common numerical-modeling assumption that all anterior muscles act synergistically as flexors is reasonable, but that the related assumption that all posterior muscles act synergistically as extensors is not. The data presented here can be used to validate a neck model at three levels of increasing neuromuscular-kinematic complexity: muscles generating forces with no movement, muscles generating forces and causing movement, and muscles generating forces in response to induced movement. These increasingly complex data sets will allow researchers to incrementally tune their neck model's muscle geometry, physiology, and feedforward/feedback neuromechanics. 5743 Mo, 14:30-14:45 (P10) Occupant awareness affects whiplash biomechanics B.D. Stemper, N. Yoganandan, F.A. Pintar. Medical College of Wisconsin and VA Medical Center, Milwaukee, WI, USA Neck muscles contract under dynamic whiplash loading to stiffen the headneck complex and reduce spinal motions. Occupant awareness affects muscle contraction timing, wherein aware occupants contract muscles prior to impact and unaware occupants contract muscles reflexively in response to impact. Due to physiological delays in the muscle contraction mechanism, reflex contraction may take 150 msec after impact to reach optimum contraction levels. The biomechanical effect of these contraction schemes on localized cervical spine soft tissue deformations during whiplash has not been investigated. The present study implemented a head-neck computational model subjected to 10.5km/h rear impacts. Geometry was obtained from CT images, material properties were obtained from literature, and the model was comprehensively validated against human volunteers, full-body cadavers, head-neck complexes, and isolated cervical spine specimens in whiplash. Head extension relative to T1, level-by-level segmental angulations, and facet joint capsular ligament distractions were quantified during initial stages of whiplash kinematics. A sagittally balanced contraction scheme was implemented to balance the headneck complex. The aware occupant attained maximum muscle contraction prior to impact. The unaware occupant implemented minimum contraction levels prior to impact to maintain upright posture. After impact, the unaware occupant implemented 50 msec reflex delay, 13-msec electromechanical delay, and 81-msec muscle rise time. The unaware occupant demonstrated normal stages of whiplash kinematics: retraction, extension, and rebound. Although the aware occupant demonstrated the three phases, S-shaped spinal curvature was not present during the retraction phase. Overall head to T1 extension was decreased by 63% in the aware occupant. Facet joint capsular ligament distractions were decreased by a maximum of 75%. Due to inherent delays in the reflex contraction mechanism and the fact that whiplash injuries occur during the initial retraction phase, it is unlikely that reflex contraction in the unaware occupant can affect spinal kinematics during whiplash and decrease the likelihood of injury. However, this study demonstrated the ability of neck muscles to stabilize the head-neck complex in whiplash and reduce soft tissue deformations when contracted prior to impact.

5.2. Whiplash and Neck Injury Biomechanics

$147

4103 Mo, 14:45-15:00 (P10) Finite element analysis of head-neck kinematics and potential ligament injury under simulated rear impact E.C. Teo, Q.H. Zhang, T.X. Qiu. School of Mechanical & Aerospace Engineering, BioMedical Engineering Research Center, NanYang Technological University, Singapore

Study Design: A detailed three-dimensional head-neck (C0-T1) finite element (FE) model of head-neck complex was exercised to dictate the kinematics of each cervical spinal segment under simulated rear impact. Background: The previous experimental studies on neck injury during whiplash were generally time and cost consuming. The information on the variation of cervical ligament strains during whiplash was hardly investigated. Methods: A detailed three-dimensional C0-C7 FE model developed previously and validated globally [1] was modified to include T1 vertebra. Rear impact of half sine-pulses with peak accelerations of 5 G and 8 G [2] were applied to T1 inferior surface, respectively, to validate and investigate intervertebral rotations and variation of ligament strains under simulated rear impacts. Results: The simulated kinematics of the head-neck complex showed relatively good agreement with the experiment with most of the predicted peak values fell within one standard deviation of the experimental data [2]. Under rear impact, the whole C0-T1 structure formed a S-shaped curvature with flexion at the upper levels and extension at the lower levels at early stage after impact, during which the lower cervical levels might experience hyperextensions. The results showed that the capsular ligament (CL) should be the major concern under the rear impact condition, followed by anterior longitudinal ligament (ALL). In current study, only the peak strain values of CL were above its failure limit under both accelerations. Under 8 G, the ALL also exceeded its failure strains. The strain value of posterior longitudinal ligament kept low under both conditions, while there is definitely no tension in ligamentum flavum and inter spinous ligament. The peak impact acceleration has significant effect on the potential injury of ligaments. Under higher acceleration, more ligaments will reach failure strain at much shorter time immediately after impact. Conclusion: The current model was identified to be available for human neck injury study. References [1] Zhang QH, Teo EC, Ng HW. Development and validation of a C0-C7 FE complex for biomechanical study. J Biomech Eng. 2005 Oct; 127(5): 729-35. [2] Panjabi MM, Ito S, Ivancic PC, Rubin W. Evaluation of the intervertebral neck injury criterion using simulated rear impacts. J Biomech. 2005 Aug; 38(8): 16941701. 6213 Mo, 15:00-15:15 (P10) Validation of human spine model under the low-speed rear-end impact S. Ejima, K. Ono. apan Automobile Research Institute The recognition of neck injuries caused by whiplash motion of the head from rear-end collisions has typically been based on movements of the 1st thoracic vertebra (T1), with the impulse force to the head and neck system considered a product of only horizontal T1 movement. However, during rear-impact volunteer tests, it was indicated that the neck is also pushed vertically due to torso ramping-up motion and straightening of the spine. This study aims to further identify this motion; utilizing both experimental data from the rear impact volunteer tests and simulated data from a human lumbar spine model. The original human lumbar spine was first modified by mesh refinement in its inter-vertebral discs, increasing its flexibility and allowing for a more biofidelic bending motion. This modified spine model was then validated against the results of the low speed rear impact volunteer experiments. During these tests, the spinal motions of each vertebra were measured by means of a spinal deformation measurement system, and accelerations were recorded at key points along the spine. When analyzing the general spine motion, it was found that three separate phases were present during the initial period after impact. Up to the first 50ms after impact, the lumbar vertebrae flex while the upper thoracic spine extends, initiating the ramping up motion of the spine. After 50ms, the lower thoracic vertebrae (T6-T8) act as a pivot for the interaction between the subject's back and the seatback. This allows the lumbar vertebrae to flex, traveling backward with the pelvis due to the femoral inertia. The final phase begins at around 150 ms as the thoracic vertebrae extend, resulting in the rearward travel of the upper torso. The results from the simulated impact showed a good correlation with experimental results, however the vertebral rotations were under-predicted for the upper thoracic vertebrae (T3-T5) during the second phase of the spine motion. This discrepancy is due to the flexibility of human shoulder joints in this area, and an improvement of the model's spine and shoulder blade interaction would improve the overall biofidelity of the spine model.