High-Fidelity Auditory Evoked Potential Recording System: Design and Clinical Application to Objective Tinnitus Diagnosis Research

Abstract

The auditory evoked potential (AEP) is an electrical activity of the auditory system following the auditory stimulus presentation. The AEP is broadly applied not only as a clinical use for the testing of hearing threshold or auditory neuropathy, but also as a research tool for the investigation of fundamental mechanisms in many neurodevelopmental disorders. For these purposes, essential requirements for the high-quality AEP recording are the low system noise and precise auditory stimulus presentation due to its very low amplitude and stimulus dependence. Additionally, flexible stimulus control and real-time AEP data processing are advantageous for the researches on customized auditory stimulus paradigms and novel signal processing algorithms. The first objective of this thesis was to design the flexible, high-performance AEP recording system as the form of a single platform in order to increase the operational stability. The low noise analog front-end and power regulation circuits were developed for the inference minimization. The various auditory stimuli generated from the integrated circuit were accurately calibrated complying with the international standard. Moreover, the parallel loop structure of the software enabled real-time AEP data processing. The evaluation results indicated that the developed system can be used for the high-fidelity AEP recording in terms of the system noise level and stimulus accuracy. In the real auditory brainstem response (ABR) and auditory late response (ALR) recordings from human subjects for further validation, clear waveform morphologies were confirmed and they were reproducible in all subjects. The second objective of this thesis was to apply the developed system to the real clinical research by utilizing its flexibility in the stimulus control and AEP data processing. Toward objective tinnitus diagnosis, the customized gap-prepluse inhibition (GPI) paradigm used in animal studies was modified in the context of human subjects with the ALR recording. In the first normative study with healthy normal-hearing subjects, the N1-P2 complex of the ALR best reflected the GPI in terms of the inhibition ratio and test-retest reliability. The minimum required number of stimuli repetitions for the stable GPI ratio was also found to shorten the test time. Using these practical findings, the discriminative stimulus condition showing the effect of tinnitus presence was found in the second comparative study with tinnitus patients. Thus, this novel approach using the ALR with the GPI paradigm may hold a promise as an objective measure of tinnitus in humans.