Auditory Brainstem Evoked Potentials Peak Identification by Finite Impulse Response Digital Filters

Abstract

Linear phase finite impulse response (FIR) filtering can be used to differentiate auditory brainstem evoked potentials (ABEP) components. The power spectrum of ABEP at high intensities indicates that they contain 3 frequency bands that can be distinguished by applying appropriate digital filters with the following characteristics: up to 240 Hz (revealing slow components), 240-483 Hz (resulting in medium components) and above 500 Hz (leaving only fast components). The results using these filters, indicate that the medium components coincide with peaks I, III and V and that the slow filter results in a 'pedestal' whose peak coincides with peak V. These findings were used for automatic identification of ABEP peaks. A coincidence of the 'pedestal' peak with a medium component was sought and labelled peak V. The preceding medium peaks were labelled, in order of decreasing latency, III and I. Validation of this procedure was conducted on ABEP from normal subjects, using different stimulus rates and intensities, as well as from selected neurological patients with lesions affecting the brainstem. Provided the waveform included a 'pedestal', the results proved this procedure to be reliable and in very good agreement with manual identification and measurement of ABEP peaks.

Key Words:

• Auditory brainstem evoked potentials
• Digital filters
• Peak identification