Biomedical Engineering Reference
In-Depth Information
of continuum-mechanical models is their ability to represent structural aspects
of a muscle's anatomy. Combined with macroscopic constitutive laws describ-
ing the stress-strain relationship of active and passive skeletal muscle tissue, the
governing equations of finite elasticity, and the Finite Element (FE) method, the
continuum-mechanical models provide new insights into stress and strain distribu-
tions (Blemker et al.,
2005
), predictions of a muscle's shape (Johansson et al.,
2000
;
Oomens et al.,
2003
; Böl and Reese,
2007
) or differences between 3D and 1D skele-
tal muscle models (Röhrle and Pullan,
2007
).
Besides the Hill-type and continuum-mechanical models, there exist models that
represent sub-cellular processes. For example, Hodgkin-Huxley-like models (e.g.,
Hodgkin and Huxley,
1952
; Shorten et al.,
2007
) describe the ionic currents and ki-
netics on the (sub-)cellular scale using ordinary differential equations (ODE). Such
models provide further insights into related processes or results of pathological con-
ditions. Bridging the scales between cellular and continuum-mechanical skeletal
muscle models opened recently a new field of modeling the skeletal muscle's elec-
tromechanical behavior by driving a contraction through muscle recruitment prin-
ciples (Röhrle et al.
2008
,
2012
; Röhrle,
2010
). The first and still only model that
describes the electromechanical behavior of skeletal muscles and accounts for the
unique manner in which skeletal muscles are activated, specifically the fact that
neighboring fibers are electrically isolated and act independently, has been proposed
by Röhrle et al. (
2008
,
2012
) and Röhrle (
2010
).
However, all studies using state-of-the-art continuum-mechanical models instead
of lumped-parameter models focus on inverse dynamics. No framework that ap-
peals to forward dynamics and principles of continuum mechanics has been pro-
posed so far. The research community investigating the dynamics of the muscu-
loskeletal system is mainly focusing on analyzing motion (gait) and exhibits limited
interest in computationally more expensive but structurally and functionally more
accurate continuum-mechanical models. Considering the already significant com-
putational costs of forward-dynamics simulations with lumped-parameter models,
it is comprehensible that the lumped-parameter models cannot only be replaced by
continuum-mechanical models. Novel and computationally efficient methodologi-
cal approaches need to be developed.
For lumped-parameter models, alternative and cost effective strategies have been
sought to estimate muscle forces. One of these alternative strategies is to use EMG
recordings to predict muscle activity to be used as input in forward-dynamics sim-
ulations. Such EMG-driven forward-dynamics simulations (using rigid-body dy-
namics and lumped-parameter skeletal muscle models) have been proposed, among
other musculoskeletal systems, to investigate elbow motion (Koo and Mak,
2005
).
While EMG data can reduce the computational cost and hence support the use of
inverse dynamics within a clinical setting, concerns remain about EMG-driven ap-
proaches, in particular about inaccuracies of linking EMG data to muscle parameters
of lumped-parameter models. The challenge of using EMG data in musculoskeletal
simulations is often related to measuring, processing, and quantifying EMG signals.
The aim of this publication is to present in more detail two different method-
ological approaches of modeling skeletal muscle mechanics spanning from the cel-
lular level, e.g., modeling the electrophysiological properties of a half-sarcomere,
