AUDITORY EVOKED POTENTIALS IN CHILDREN WITH COCHLEAR IMPLANTS: A PRELIMINARY STUDY
Emily Jo Venskytis · D-Scholarship@Pitt (University of Pittsburgh) · 2013
This is a preliminary and exploratory study designed to investigate the use of electrophysiologic recordings of cortical auditory evoked potentials (CAEPs) as an objective test measure for patients who have a cochlear implant (CI), with a more specific goal of learning best practices for extracting the CI artifact from the collected data. The particular evoked potential of interest was the P1-N1-P2 complex. It was hypothesized that normal latencies and peak amplitudes of the late-evoked potential (P1-N1-P2) would reveal characteristics that may correlate with behavioral data to indicate the level of benefit with a cochlear implant. The study included three participants, each with one cochlear implant on the right side: two males, age 13, and one female, age 10. The study also evaluated electrophysiologic data from two normal-hearing volunteers, age 21. Data was collected via a 64-channel electrode cap with two reference electrodes placed on the mastoid and two additional upper and lower vertical eye channels. The stimulus was created using the guidelines by Dimitrijevic et al. (2008, 2011) and consisted of two test blocks of 250 Hz and 4000 Hz, respectively, with frequency modulations at 0% (no change), 2%, 4%, 10%, 25%, and 50% every 1.4s, lasting for approximately 100ms. Evoked potentials were recorded in response. The ongoing EEG was decimated and converted into MATLAB® format to run an independent component analysis (ICA) using the runica algorithm (Bell & Sejnowski, 1995). A method of waveform extraction was developed in MATLAB® and EEGLAB in order to evaluate the peak amplitudes and latencies. Following ICA, independent components presumably affected by the cochlear implant were identified and rejected according to methods adapted from Gilley et al. (2006). Data was highly variable across the small number of subjects, raising several questions about age at implantation, length of CI use, etiology of hearing loss and speech/language processing abilities in the pediatric user. From this study, it appears that the artifactual component from the implant can be removed from the electrophysiologic response so that CAEPs in children with CIs can be investigated across a larger number of individuals from this pediatric population.