Foot and Ankle Model for Surgical Treatment - Computational Biomechanics of the Musculoskeletal System - page 40

which may be an important factor in bone stress distribution. The plantar ligament was represented by
a solid model and was assumed to be linearly elastic, which is much different from reality. The cadav-
eric experiment was carried out to validate the FE model. They found that the release of the plantar
fascia decreased arch height without causing the total collapse of the foot arch and the longitudinal
foot arch was destroyed after sectioning of the four major plantar ligaments. The FE model indicated
that the release of the plantar fascia may relieve focal stress associated with heel pain.
An FE study was conducted to evaluate the effects of different foot postures on fifth metatarsal
fracture healing (Brilakis et al. 2012). The model combined 53 plantar and dorsal ligaments, the
plantar fascia, 28 bones, and surrounding soft tissue, all of which were set as homogeneous, isotro-
pic, and linearly elastic. The muscle attachment points were not positioned exactly at their anatomi-
cal locations, and boundaries conditions, including the muscle forces, were simplified to be of the
same value in three different foot postures. As stress in the bone is mainly caused by the transfer of
ground reaction forces (GRFs) and muscle contraction, inaccurate application of GRFs and muscle
forces may lead to erroneous results. Different postures were simulated by simply turning the model
through different angles in relation to the ground, instead of simulating the actual interaction of dif-
ferent segments during different postures. The study concluded that different foot postures did not
significantly influence the peak strains at the fracture site of the fifth metatarsal and eversion of the
foot caused further torsional strain on the fracture site of the fifth metatarsal.
Although the aforementioned studies offer much insight into the biomechanical environment
within the foot and corresponding forces between segments, there are inherent limitations to these
models that may affect their reliability. It is necessary to develop a more accurate model of the foot
and ankle for clinical application.
3.3 FInIte element SImulatIon oF tarSometatarSal JoInt FuSIon
An FE model of the foot and ankle was modified from a previous study, the details of which are
provided in Chapter 1. The FE model was based on the right foot of a normal female adult with
body weight and mass of 164 cm and 54 kg, consisting of 28 bony segments, 72 ligaments, and
the plantar fascia embedded in a volume of encapsulated soft tissue (Cheung and Zhang 2005;
Cheung et al. 2005). For simplicity, it was assumed that the TMT joint fusion does not change the
foot kinematics much during gait, based on the fact that the motion of TMT joints is quite limited
in a normal foot.
3.3.1 B oundary and l oadinG c onditionS durinG W alkinG
The kinematic and kinetic information during walking was obtained from human locomotion anal-
ysis using a motion monitoring system, force platform, and electromyography (EMG) measure-
ments. The gait tracking system provides kinematic information on the lower limbs with markers
positioned on regions of interest. The curve of the GRFs obtained from the force platform, with the
first peak, midstance, and the second peak, are shown in Figure 3.1.
The instant of the first peak in terms of the vertical GRF is at about 25% and the second peak
located around 70% of the stance phase. The midstance instant was chosen as the valley in the curve
between the first and second peak. To simulate different gait instants, the active extrinsic muscle
forces, in addition to the GRF, were applied. The muscle forces were estimated from physiological
cross-sectional areas (Dul 1983) of muscles and EMG data with a linear EMG-force assumption
(Kyu-Jung Kim et al. 2001). All of the boundary and loading conditions obtained from the four
simulated instants are listed in Table 3.1.
The muscles in the FE model were represented by lines connecting the anatomical attachment
points of muscles to bones. The Achilles tendon was represented by five axial connector elements, on
which five equivalent force vectors were applied at the points of insertion. Other muscle forces were
applied to the corresponding muscle structures represented by solid lines, as shown in Figure 3.2,