Operation Manual MB 11 Classic
8100498-4_OM_MB11_Classic_3.18_EN_16a.docx
66
01/16
halving the amplitudes of the cosines with the lowest and the highest
frequency. In contrast to the standard click generated in the time domain
the new stimulus is a steady-state signal.
The patented stimulus constructed from individual cosines offers the
possibility of introducing a frequency-dependent phase correction in order
to compensate for the propagation time in the cochlea. From this phase
correction a synchronized activation of the whole basilar membrane can be
expected.
Figure 33 b) shows to the left the time course of a phase corrected click. The
corresponding amplitude spectrum is shown at the right side. The red arrows
mark the position of the first six spectral harmonics of the brainstem
response evoked by the phase corrected stimulus. These harmonics were
analyzed by the objective statistical test implemented in the automatic test
algorithm of the MB 11 Standard (Classic)
software. It is easy to see that five
of the six harmonics of the brainstem responses are overlapping with the
stimulus frequencies. This overlapping could lead to false results of the
statistical test when electrical stimulus artifacts occur. For elimination of this
source of error a frequency offset is introduced into the cosines. This
frequency offset causes a displacement of the stimulus frequency spectrum.
In the time domain, the frequency offset leads to an alternating stimulus.
With this stimulus a better detection time and shorter test time could be
realized.
For further optimization the question of the adequate cochlea model for
calculating the phase correction was addressed. The cochlea model from de
Boer is based on measurements at ears of dead bodies with non physiological
high stimulus levels. Due to this fact the cochlea delay derived from the de
Boer model has to be considered as not optimal. For this reason a new model
function of the frequency-dependent propagation time in the cochlea was
calculated based on the latency data of frequency specific ABR recorded at
low stimulus levels.
In this context the question arises as to whether the delay of the different
models has to be considered to reflect the group delay or the phase delay of
the cochlea. The phase delay describes how much a steady-state cosine of a
specific frequency is delayed from the input to the output of a system (of the
cochlea). The group delay describes how much the envelope of a signal
formed by a narrow group of frequencies is delayed through the system.
The results of a study confirmed the assumption that the latencies of the
frequency-specific ABR reflect the group delay. Since the stimulus shall
compensate for the cochlear delay of single frequencies the phase delay
calculated from the group delay was used for the phase correction of the
