Development of a multi-body computational model of human head and neck

Experimental studies using human volunteers are limited to low acceleration impacts while whole cadavers, isolated cervical spine specimens, and impact dummies do not normally reflect the true human response. Computational modelling offers a cost effective and useful alternative to experimental methods to study the behaviour of the human head and neck and their response to impacts to gain insight into injury mechanisms. This article reports the approach used in the development of a detailed multi-body computational model that reproduces the head and cervical spine of an adult in the upright posture representing the natural lordosis of the neck with mid-sagittal symmetry. The model comprises simplified but accurate representations of the nine rigid bodies representing the head, seven cervical vertebrae of the neck, and the first thoracic vertebra, as well as the soft tissues, i.e. muscles, ligaments, and intervertebral discs. The rigid bodies are interconnected by non-linear viscoelastic intervertebral discs elements in flexion and extension, non-linear viscoelastic ligaments and supported through frictionless facet joints. Eighteen muscle groups and 69 individual muscle segments of the head and neck on each side of the body are also included in the model. Curving the muscle around the vertebrae and soft tissues of the neck during the motion of the neck is also modelled. Simulation is handled by the multi-body dynamic software MSC.visuaNastran4D. Muscle mechanics is handled by an external application, Virtual Muscle, in conjunction with MSC.visuaNastran4D that provides realistic muscle properties. Intervertebral discs are modelled as non-linear viscoelastic material in flexion and extension but represented by‘bushing elements’ inVisualNastran 4D, which allows stiffness and damping properties to be assigned to a joint with required number of degrees of freedom of the motion. Ligaments are modelled as non-linear viscoelastic spring–damper elements. As the model is constructed, the cervical spine motion segments are validated by comparing the segment response to published experimental data on the load–displacement behaviour for both small and large static loads. The response of the entire ligamentous cervical spine model to quasi-static flexion and extension loading is also compared to experimental data to validate the model before the effect of muscle stiffening is included. Moreover the moment-generating capacity of the neck muscle elements has been compared against in vivo experimental data. The main and coupled motions of the model segments are shown to be accurate and realistic, and the whole model is in good agreement with experimental findings fromactualhumancervical spine specimens. It has been shown that the model can predict the loads and deformations of the individual soft-tissue elements making the model suitable for injury analysis. The validation of the muscle elements shows the morphometric values, origins, and insertions selected to be reasonable. The muscles can be activated as required, providing a more realistic representation of the human head and neck. The curved musculature results in a more realistic representation of the change in muscle length during the head and neck motion.