Pratt Fellow Garrett Wood Models Cervical Spine Muscles
This article is part of Summer Stories, a special, online issue of Dukengineer Magazine, in which students wrote about their experiences in the Summer of 2007 during their time away from Duke.
by Garrett Wood, BME/ME’08
Every year, millions of serious injuries in the U.S. are caused by automobile crashes. Of those incidents that lead to long-term disabilities or death, head and neck injuries stand as a major culprit. Therefore, it is important for automobile manufacturers and automobile owners alike to understand more about the causes and prevention of these injuries. Here at Duke, as part of the Pratt Undergraduate Research Fellows program, I spent the summer using engineering to gain a better understanding of the causes and characteristics of neck injury in order to promote automobile safety.
Automobile impact safety is normally tested using Anthropomorphic Testing Devices (ATDs), more commonly known as “crash test dummies”. These ATDs are useful and give safety researchers data on the severity of a car crash, but the ATDs are not perfect. A very important example of these imperfections is the ATD neck as engineers are faced with the difficult task of modeling the highly complicated human neck with manufactured parts. In order to design a mechanical system to mimic the human neck, a better physical and quantitative understanding must be achieved.
An improved biomechanical understanding of the human neck is achieved today through the use of computational modeling. Researchers have used finite element analysis of the human in recent years to quantify loads associated with external accelerations or forces. Through the use of biomedical imaging, a computational model of the osteoligamentous system (the bones and ligaments) can be created, and used to simulate responses to external forces. However, this model loses its effectiveness under certain loading due to the absence of neck musculature. Previous graduate students in Duke’s Injury and Orthopaedic Biomechanics Laboratory have created a human head and neck computational model consisting of bone, ligament and muscle. This model however, does not perform accurately under some conditions.
Therefore, my efforts have focused on the identification of deep musculature of the neck that may play a role in the stabilization of the head under the effects of gravity. In order for muscles' segments to be included into a computational model, we must identify important characteristics of a muscle’s geometry and structure. Those characteristics include fiber length, physiologic cross-sectional area (PCSA), as well as origin and insertion points. The maximum force that a muscle can possibly exert is determined by its PCSA--a function of muscle volume, fiber length, and the angle of attachment, or "pennation angle." These muscle characteristics have been found by dissecting cadavers and through biomedical imaging techniques, such as CT and MRI.
Once the required muscle characteristics are found, they are modeled using computational elements in a finite element model. These muscles consist of a parallel series of nonlinear springs and dampers, representing the constant tissue force exerted by a muscle and the force exerted by muscle flexion. I have used reverse optimization to determine muscle activation levels that correspond to minimum muscle fatigue or energy use in computational simulations. Reverse optimization methods allows an engineer to set boundary conditions and constraints to a system, and then through an iterative process minimize or maximize a defined function according to the desired use of the results.
Through quantitative analysis of muscle characteristics and the use of computational modeling coupled with reverse optimization techniques, a head and neck model is used to simulate the human behavior in response to external loading. This summer I have worked to model and optimize several intersegmental muscle pairs to improve the Duke Head and Neck computational model. It is theorized that these smaller intersegmental muscles play an important role in the stabilization of the neck under small external forces. No conclusive results have been achieved to date, but if the correct muscles can be identified and implemented it may be possible to physically model these elements in a new ATD. The more accurate ATDs become, the safer automobiles can be designed and used in the future.