Rapid neonatal hearing screening using maximum length sequence auditory brainstem response, chirps and automated signal detection

Universal Newborn Hearing Screening (UNHS) programs have been implemented in many countries throughout the world, including Australia and New Zealand. Currently the most widely accepted tests for UNHS are the Otoacoustic Emissions (OAE) and the Auditory Brainstem Response (ABR). Whilst both are popular, the ABR is often preferred because it tests a greater proportion of the auditory pathway with a higher sensitivity and specificity (Dort et al., 2000). Despite its advantages, the ABR has limitations. Of particular interest to UNHS is the time required to complete an ABR assessment. A standard ABR screening assessment can usually be completed in approximately 8 minutes for both ears (including subject preparation). This falls to approximately 3 minutes if a simple pass/fail result is all that is required, provided the infant is in natural sleep or quiet rest (Hahn et al., 1999). Part of the ABR test time is taken up by the signal averaging process. Several thousand auditory responses are required to obtain a reliable ABR at or near threshold levels. To make matters worse, each new response cannot be obtained until after the previous response has finished. Whilst these delays may seem small per neonate, they are large per UNHS program, particularly in programs where tens of thousands of neonates must be screened per year. Such delays result in significant cost to the organisations responsible for UNHS (National Center for Hearing Assessment & Management (NCHAM)), and go against the UNHS guidelines provided by the American Speech-Language and Hearing Association (ASHA) in 1997 (American Speech-Language Hearing Association Audiologic Assessment Panel 1996, 1997) which state that a screening test should be short in duration, comfortable for patients, and of minimal cost to the organization providing the screening service. This paper describes three methods that could reduce ABR acquisition time: Maximum length sequences (MLS), chirp stimuli and automated signal detection