Wave V detection using continuous wavelet transform of auditory brainstem response signal

Abstract

Auditory brainstem response (ABR) audiometry is well known tools for detection of hearing loss. It is a neurologic test of ABR in response to the stimuli such as click sound, tone burst or chirp. The non-invasive electrodes are used to obtain the ABR signals. In ABR, wave V of the ABR signals is used as indicator of hearing loss because it is the most dominant and robust peak wave. In current device, there are about 2000 sweeps of the ABR signals have to be averaged in order to get the meaningful signal which required about 4 minutes for one single stimulation level, which is consumed a lot of time. Therefore it is compulsory to have a fast detection method to reduce the recording time. Several types of signal processing techniques have been introduced by researchers to obtain the best technique to detect the ABR signals such as Fast Fourier Transform (FFT) and discrete wavelet transform (DWT). In this study, the continuous wavelet transform (CWT) of ABR signal had been introduced as a marker to identify the ABR waves. Study showed that the CWT of auditory brainstem response can be used a marker to identify the ABR waves with the sensitivity, 0.63 and the specificity, 0.83.