Auditory brainstem responses are used to detect hearing defects in aud
iology and otoneurology. The use of computer programs for the analysis
of such recordings is increasing. To identify their detailed properti
es a pattern recognition algorithm implemented in an analysis program
must be highly reliable. For the recognition process, some preprocessi
ng phases after recording are necessary, such as filtering and often a
lso segmentation. In the following, we will explore segmentation, whic
h can be used in preprocessing of biomedical signals after filtering.
We studied linear segmentation, where slopes of short signal segments
are computed and divided into different classes according to their val
ues. A segment length of 8 samples for a sampling frequency of 50 kHz
employed was best according to our tests and error criteria. Using clu
stering, we found that less than 10 segment classes is suitable for pa
ttern recognition.