6.2 Signals of the Muscles and Motions
|
229
Fig. 6.11: Determination of two instants of time for the synchronisation called trigger points; KB:=
knee bend, force (lined), knee angle (dotted).
Design of MATLAB algorithm
Based on manual analysis [11, 73] of the points of synchronisation an algorithm was
developed which automates the first steps of synchronisation [69]. Attention should
be paid on the patient data, because the results achieved, can vary individually. The
selected features should minimise the inaccuracy of the synchronisation based on the
measured data.
As above mentioned there are characteristic patterns in the force and joint angle
data which can be used as features. Prominent features are the times of rest between
the knee bends, which are further referred to as standing phases. In these phases the
patient legs are mostly extended in an upright position. In Figure 6.9 the standing
phases are circled black. The time difference between the standing phases in the force
and joint angle curve is roughly the same thus the centres of the standing phases can
be used as anchors for synchronising force and joint angle data. The centre points of
the standing phases are further referred to as trigger points (TP). Regarding the joint
angles we will discuss furthermore the knee joint angle because the data of the joint
angle of hip, knee and ankle are in already synchronous.
Synchronisation procedure
In the following part a variety of algorithms will be presented regarding the finding
of the relevant characteristic points discussed previously. In MATLAB there are sev-