The current status of the use of fast Fourier transform for auditory evoked response

Abstract

The current status of the application of the fast Fourier transform (FFT) for the auditory evoked response was summarised under for items of importance to the auditory brain stem response (ABR) as follows. (1) Digital filter: The most common way to use FFT, which facilitates quick isolation of fast and slow components from data of the same ABR wave. In addition, the proper setting of digital filter increase the rate of detection of waves II and IV, the frequencies of which have conventionally been considered to be low. (2) Power spectrum analysis: It is difficult to obtain clinically significant information from power spectrum in the present situation. However, it seems significant in ABR to isolate and analyze slow, middle (tentative name) and fast components from the Fourier components revolving around each peak of the power spectrum with a three-peak pattern by inverse fast Fourier transform (IFFT). (3) Cross correlation function: This method is the way to analyze the relationship between two signals from the viewpoint of the phase. Clinical application including application to the diagnosis of multiple sclerosis, application to automatic detection for ABR, etc. are expected in the future. (4) Phase specteral analysis: The synchrony measure method (Fridman, 1982) is one of the phase spectral analysis made by means of FFT. In this method, phase variance for selected Fourier components are calculated among the averaging groups of 200 sweeps under the same stimulating conditions, in order to determine the presence or absence of response by the degree of the timelocked condition of the Fourier components for stimuli. Its clinical application to automatic evaluation of ABR, etc. is being studied.